CC Omake

CLEARCASE OMAKE MANUAL
Release 4.0 and later
Windows Edition
800-012616-000
ClearCase OMAKE Manual
Document Number 800-012616-000 December 1999
Rational Software Corporation 20 Maguire Road Lexington, Massachusetts 02421
IMPORTANT NOTICE
Copyright Notice
Copyright © 1992, 1999 Rational Software Corporation. All rights reserved.
Copyright 1989, 1991 The Regents of the University of California
Copyright 1984–1991 by Raima Corporation
Copyright 1992 Purdue Research Foundation, West Lafayette, Indiana 47907
Trademarks
Rational, the Rational logo, Atria, ClearCase, ClearCase MultiSite, ClearCase Attache, Clear DDTS,
ClearQuest, ClearGuide, PureCoverage, Purify, Quantify, Rational Rose, and SoDA are trademarks or
registered trademarks of Rational Software Corporation in the United States and in other countries. All other
names are used for identification purposes only and are trademarks or registered trademarks of their
respective companies.
Microsoft, MS, ActiveX, BackOffice, Developer Studio, Visual Basic, Visual C++, Visual InterDev, Visual J++,
Visual Studio, Win32, Windows, and Windows NT are trademarks or registered trademarks of Microsoft
Corporation.
Sun, Solaris, and Java are trademarks or registered trademarks of Sun Microsystems, Inc.
Oracle and Oracle7 are trademarks or registered trademarks of Oracle Corporation.
Sybase and SQL Anywhere are trademarks or registered trademarks of Sybase Corporation.
U.S. Government Rights
Use, duplication, or disclosure by the U.S. Government is subject to restrictions set forth in the applicable
Rational License Agreement and in DFARS 227.7202-1(a) and 227.7202-3(a) (1995),
DFARS 252.227-7013(c)(1)(ii) (Oct 1988), FAR 12.212(a) 1995, FAR 52.227-19, or FAR 52.227-14, as applicable.
Patent
U.S. Patent Nos. 5,574,898 and 5,649,200 and 5,675,802. Additional patents pending.
Warranty Disclaimer
This document and its associated software may be used as stated in the underlying license agreement, and,
except as explicitly stated otherwise in such license agreement, Rational Software Corporation expressly
disclaims all other warranties, express or implied, with respect to the media and software product and its
documentation, including without limitation, the warranties of merchantability or fitness for a particular
purpose or arising from a course of dealing, usage or trade practice.
Technical Acknowledgments
This software and documentation is based in part on BSD Networking Software Release 2, licensed from the
Regents of the University of California. We acknowledge the role of the Computer Systems Research Group
and the Electrical Engineering and Computer Sciences Department of the University of California at Berkeley
and the Other Contributors in its development.
This software and documentation is based in part on software written by Victor A. Abell while at Purdue
University. We acknowledge his role in its development.
/vobs/doc/ccase/omake/cc_omakeTTL.fm — October 8, 1999 5:44 pm

Contents
Preface .......................................................................................................................................xv
About This Manual ..........................................................................................xv
ClearCase Documentation Roadmap .......................................................... xvi
Typographical Conventions ........................................................................ xvii
Online Documentation ................................................................................ xviii
Technical Support ........................................................................................ xviii
1. Introduction ........................................................................................................................... 1
1.1 File Manifest........................................................................................................ 1
1.2 Configuring omake ............................................................................................ 2
1.3 Notes on Using omake ...................................................................................... 2
omake’s Configuration Management Features.............................................. 2
Configuration Lookup................................................................................ 3
Derived Object Sharing .............................................................................. 3
Configuration Record Creation................................................................. 3
Express Builds ............................................................................................. 3
Command-Line Options ............................................................................ 3
Location of Temporary Files...................................................................... 4
Non-MVFS Dependencies ......................................................................... 4
Differences Between omake and Standard Make Tools ...............................4
Running omake .................................................................................................. 5
Makefiles.............................................................................................................. 5
Supporting Both omake and clearmake .................................................. 5
Parallel and Distributed Build Operations..................................................... 6
Build Scripts and the rm Command................................................................ 6
Pathnames in CRs .............................................................................................. 6
1.4 Auditing 16-bit Tools......................................................................................... 7
Contents iii
/vobs/doc/ccase/omake/cc_omakeTOC.fm — September 10, 1999 12:08 pm

2. Overview of Using omake ..................................................................................................9
2.1 The omake Process .............................................................................................9
2.2 Invoking omake from the Command Line ...................................................10
2.3 Initialization Files and Makefiles ...................................................................11
The Initialization File .......................................................................................11
The Makefile ......................................................................................................12
Continued Lines ........................................................................................12
Comments...................................................................................................12
Contents of Initialization Files and Makefiles ..............................................13
2.4 Rules ...................................................................................................................13
2.5 Dependency Lines ............................................................................................14
Explicit and Inferred Dependencies...............................................................14
Macros in Dependency Lines..........................................................................15
The Make Process Is Recursive................................................................15
Detected Dependencies ............................................................................16
Wildcards in Dependency Lines ....................................................................16
The Dependency Line Separator ....................................................................16
A Dependency Example ..................................................................................17
2.6 Build Scripts ......................................................................................................18
Build Script Execution......................................................................................19
Auto-Detection Mode ...............................................................................19
Standard Execution Mode........................................................................20
Build-Script Line Exit Status ...........................................................................20
2.7 Macros ................................................................................................................21
Macro Precedence.............................................................................................22
Defining Macros in the Makefile ....................................................................22
Read-Time Expansion of Macros ............................................................23
Standard Macro Definition: name = [ text ] ...........................................23
Conditional Macro Definition: name ?= [ text ] ....................................23
Expanded Macro Definition: name := [ text ] ........................................24
Appended Macro Definition: name += [ text ]......................................24
Case-Sensitivity of Macro Names ...........................................................24
The Location of Macro Definitions .........................................................24
iv ClearCase OMAKE Manual

Indenting Macro Definitions ................................................................... 24
Undefining Macros ................................................................................... 24
Example Macro Definitions ..................................................................... 25
Defining Macros on the Command Line ...................................................... 25
Dynamic Macros............................................................................................... 26
Macro Modifiers ............................................................................................... 26
File Name Components............................................................................ 27
Tokenize ..................................................................................................... 27
Other Modifiers ......................................................................................... 28
Environment Variables.................................................................................... 28
Macro Expansion or Macro Referencing....................................................... 28
Run-Time Expansion of Macros.............................................................. 29
Recursive Macro Definitions ................................................................... 29
2.8 Inference Rules ................................................................................................. 29
Defining Inference Rules................................................................................. 30
The General Inference Rule Definition .................................................. 31
The Target Inherits Build Scripts and Attributes .................................31
Alternative (Suffix-Only) Form............................................................... 31
Automatic Use of Inference Rules ................................................................. 32
Inference Rules and Target Groups............................................................... 32
Multiple-Step Inference Rules ........................................................................ 33
Chained Targets Are Deleted Automatically........................................ 33
Preventing Multiple-Step Rules.............................................................. 34
Inference Rule Search Order........................................................................... 34
Overriding the Rule Ordering................................................................. 35
Rule Finding for Target Names with a Directory Component ..........35
Common Operations on Inference Rules...................................................... 35
Built-In Inference Rules ................................................................................... 36
Compatibility with Suffix Rules (.SUFFIXES).............................................. 37
Handling of Suffix Rules and .SUFFIXES.............................................. 37
2.9 Response Files................................................................................................... 37
Inline Response Files ....................................................................................... 38
Contents v

2.10 Directives ...........................................................................................................38
2.11 The Keep-Working Mode ................................................................................39
3. omake Reference................................................................................................................41
3.1 Command-Line Options..................................................................................41
Initial Command-Line Parameters: OMAKEOPTS .....................................41
The Command-Line Options ..........................................................................42
The Current Options ........................................................................................46
3.2 Locating the Initialization File ........................................................................46
Disabling the Initialization File ......................................................................46
3.3 The Make Process .............................................................................................47
Read Time ..........................................................................................................47
Run Time............................................................................................................47
Run-Time Initialization and Deinitialization ........................................48
Updating the Time Stamp ...............................................................................49
3.4 Targets ................................................................................................................49
Case-Sensitivity of Target Names ..................................................................49
A Target Has a Name and a Pathname .........................................................49
Using the Path Separator in Names ...............................................................50
The First Target in the Makefile Is the Default Target ................................50
Targets May Appear on Several Dependency Lines ...................................50
Double-Colon Dependency Lines ..................................................................51
Mixing Single-Colon and Double-Colon Dependency Lines.....................51
Targets Without Dependencies ......................................................................51
Target Groups ...................................................................................................51
3.5 Macros ................................................................................................................52
Macro Modifiers................................................................................................52
Filename Components ..............................................................................53
Absolute Pathname ...................................................................................53
Append and Prepend Strings ..................................................................54
Change Case...............................................................................................54
Expand Pathnames....................................................................................54
Include File .................................................................................................54
vi ClearCase OMAKE Manual

Include File with Regular Expression Matching .................................. 55
Member and Nonmember ....................................................................... 55
Select a Particular Element ...................................................................... 56
String Substitution .................................................................................... 56
Tokenize ..................................................................................................... 57
Wildcard Expand File and Directory Names........................................ 57
Regular Expressions......................................................................................... 58
Configuring Regular Expressions........................................................... 58
Regular Expressions for the M Modifier ............................................... 58
Regular Expressions for the S Modifier ................................................. 59
With Configuring .REGEX_CHAR and .REGEX_WILD.....................59
Predefined and Built-In Macros ..................................................................... 60
Predefined Macros: Run-Time Macros ......................................................... 60
Run-Time Macros............................................................................................. 60
Predefined Macros: General Macros ............................................................. 62
Predefined Macros: State Macros................................................................... 63
An Example Use of the State Macros ..................................................... 66
Built-in Macros ................................................................................................. 66
Compatibility with Other Make Utilities............................................... 68
3.6 Build-Script Line Prefixes ............................................................................... 69
Do Not Echo the Build-Script Line (Silent Operation)................................69
Ignore the Build-Script Line Exit Status........................................................ 70
Override the –n Command-Line Option ...................................................... 71
Select the Shell Program.................................................................................. 71
Iterate the Build Script..................................................................................... 71
Miscellaneous Prefixes .................................................................................... 72
Build Script Compatibility with Other Make Utilities................................72
3.7 Build-Script Problems: The cd and set commands...................................... 72
Using Multiple-Command Build-Script Lines............................................. 73
Using Directives: %chdir and %setenv ......................................................... 74
Contents vii

3.8 Makefile Directives...........................................................................................74
Percent Directives .............................................................................................74
Conditional Directives ..............................................................................75
Iteration Directives....................................................................................75
Other Percent Directives...........................................................................76
Conditional Directives ..............................................................................76
Conditional Directives and Continued Lines........................................77
Conditional Expressions..................................................................................78
Simple Expressions ...................................................................................78
Comparison Operators .............................................................................79
Functional Operators (Also Called Built-In Functions) .......................80
File-Test Operators ....................................................................................80
Command-Execution Operator ...............................................................81
Logical Operators ......................................................................................82
Iteration Directives ...........................................................................................82
The %foreach Directive.............................................................................82
The %while Directive................................................................................84
Effects of %foreach and %while when Using omake in a VOB..........84
Interrupting the Iteration .........................................................................85
Another Method of Iteration....................................................................85
A Sample Makefile ....................................................................................85
Other Percent Directives..................................................................................86
Compatibility with Other Make Utilities ...............................................91
Dot Directives....................................................................................................91
Compatibility with Other Make Utilities .............................................101
3.9 Target Attributes.............................................................................................101
Using Attributes..............................................................................................102
Attributes and Inference Rules .....................................................................102
List of Attributes .............................................................................................102
3.10 Special Targets ................................................................................................105
Compatibility with Other Make Utilities .............................................106
viii ClearCase OMAKE Manual

3.11 Search Directories........................................................................................... 106
Implied Location of Missing Files ........................................................ 107
Search Directory Macros ............................................................................... 107
The .PATH Macros ................................................................................. 107
The VPATH Macros................................................................................ 108
Search Directories and Run-Time Macros .................................................. 108
Search Directories and File Lookup............................................................. 108
Search Directories and Inference Rules....................................................... 109
Debugging the Search Directories ............................................................... 110
Compatibility with Other Make Utilities............................................. 110
3.12 Response Files................................................................................................. 110
Automatic Responses .................................................................................... 111
Adding Program Names........................................................................ 111
Response Class Parameters ................................................................... 111
Defining or Modifying a Response Class ............................................ 112
Disabling a Response Class ................................................................... 112
Response File Example #1...................................................................... 113
Response File Example #2...................................................................... 113
Using Automatic Responses......................................................................... 114
Generation of Automatic Responses .................................................... 114
Deletion of Automatic Responses......................................................... 114
Built-In Automatic Responses............................................................... 115
Built-in Responses for Windows NT.................................................... 115
Inline Response Files ..................................................................................... 115
Deletion of Inline Response Files.......................................................... 116
Inline Response File Example ............................................................... 116
Compatibility with Other Make Utilities............................................. 117
4. Debugging Makefiles ...................................................................................................... 119

4.1 Command-Line Options ............................................................................... 119
4.2 Read-Time Debugging .................................................................................. 120
Output Produced by –#1 ............................................................................... 121
Output Produced by –p................................................................................. 122
Contents ix

The Macro Definitions ...................................................................................122
The Search Directories ...................................................................................123
The Automatic Response Definitions ..........................................................123
The Inference Rules ........................................................................................123
The Targets and Build Scripts.......................................................................125
The Final –p Output .......................................................................................125
A. Errors and Warnings.......................................................................................................127

A.1 Reducing Message Severity ..........................................................................127
A.2 Error Messages and Explanations ................................................................128
B. Exit Status Values ...........................................................................................................135
C. Built-In Macros and Rules .............................................................................................137

C.1 Macros ..............................................................................................................137
Predefined General Macros...........................................................................137
Predefined State Macros ................................................................................138
Built-In Macros................................................................................................140
C.2 Macro Modifiers..............................................................................................141
C.3 Inference Rules................................................................................................142
Compatibility with Other Make Utilities ....................................................144
D. Compatibility and Emulation ........................................................................................145

D.1 PM/CB (Intersolv Configuration Builder and PolyMake) .......................145
System Macros ................................................................................................145
Transformation Macros .................................................................................147
Built-In Functions ...........................................................................................148
Built-In Operations (Percent Directives) .....................................................148
Directives .........................................................................................................148
Reserved Targets.............................................................................................151
Local Input Scripts..........................................................................................151
Operation-Line Modifiers..............................................................................152
PM/CB Emulation..........................................................................................152
x ClearCase OMAKE Manual

Emulation at Startup Time .................................................................... 153
Emulation After Startup Time............................................................... 153
The Command Line ................................................................................ 153
The Emulation File (BUILTINS.CB) ..................................................... 153
The Initialization File (TOOLS.INI)...................................................... 154
The BUILTINS File.................................................................................. 154
The Makefile ............................................................................................ 154
Makefile Contents ................................................................................... 154
Operation Lines (Build Scripts)............................................................. 155
Unsupported PM/CB Features .................................................................... 156
Unimplemented Directives and Reserved Targets ............................156
Iteration Groups ...................................................................................... 156
Suffix Dependencies ............................................................................... 157
Command-Line Flags ............................................................................. 157
Shared Definitions .................................................................................. 157
Operation-Line Modifiers ...................................................................... 157
Built-In Operations ................................................................................. 157
D.2 Microsoft NMAKE Compatibility ............................................................... 158
NMAKE Directives ........................................................................................ 158
NMAKE Emulation........................................................................................ 158
Emulation at Startup Time .................................................................... 158
Emulation After Start-Up Time............................................................. 159
The Command Line ................................................................................ 159
The Emulation File (BUILTINS.NM) ................................................... 159
The Initialization File (TOOLS.INI)...................................................... 159
Macros ...................................................................................................... 160
Build Scripts............................................................................................. 161
Inline Response Files .............................................................................. 162
Unsupported NMAKE Features .................................................................. 162
D.3 Opus Make v5.2x Compatibility and Emulation....................................... 162
Contents xi

D.4 Borland Make Compatibility ........................................................................163
D.5 UNIX Make Compatibility ............................................................................164
E. Regular Expressions.......................................................................................................167
E.1 Configuration of Regular Expressions ........................................................167
Regular Expression Components .................................................................167
Referencing the Matched Expression ..........................................................169
E.2 Macro Modifiers in OMAKE.........................................................................169
Regular Expressions for the M Modifier .....................................................169
Regular Expressions for the S Modifier.......................................................170
Glossary ...................................................................................................................................171
Index..........................................................................................................................................175
xii ClearCase OMAKE Manual

Tables
Table 1 Macro Modifiers................................................................................................... 27

Table 2 omake Command-Line Options ........................................................................ 42
Table 3 General Macros .................................................................................................... 62
Table 4 State Macros.......................................................................................................... 64
Table 5 Built-in Macros ..................................................................................................... 67
Table 6 Conditional Directives ........................................................................................ 75
Table 7 Iteration Directives .............................................................................................. 75
Table 8 Other Percent Directives ..................................................................................... 76
Table 9 Comparison Operators........................................................................................ 79
Table 10 Functional Operators........................................................................................... 80
Table 11 File-Test Operators .............................................................................................. 81
Table 12 Command-Execution Operator ......................................................................... 81
Table 13 Logical Operators................................................................................................. 82
Table 14 Other Percent Directives ..................................................................................... 86
Table 15 omake Directives.................................................................................................. 92
Table 16 Attributes ............................................................................................................ 103
Table 17 Special Targets.................................................................................................... 106
Table 18 Response Class Parameters .............................................................................. 112
Table 19 Error Messages ................................................................................................... 128
Table 20 Exit Status Values .............................................................................................. 135
Table 21 General Macros .................................................................................................. 137
Table 22 State Macros........................................................................................................ 139
Table 23 Built-In Macros................................................................................................... 140
Table 24 Macro Modifiers................................................................................................. 141
Table 25 PM/CB System Macros..................................................................................... 145
Table 26 PM/CB Transformation Macros...................................................................... 147
Table 27 PM/CB Directives.............................................................................................. 149
Table 28 PM/CB Operation-Line Modifiers .................................................................. 152
Table 29 Regular Expression Components .................................................................... 168
Table 30 Examples of Regular Expressions ................................................................... 170
Tables xiii
/vobs/doc/ccase/omake/cc_omakeLOT.fm — September 10, 1999 11:43 am

xiv ClearCase OMAKE Manual
/vobs/doc/ccase/omake/cc_omakeLOT.fm — September 10, 1999 11:43 am

Preface
The ClearCase omake software build program provides compatibility with Windows NT-based
make tools.
About This Manual
This manual is for users who are new to the omake tool and who may have other experience with
make tools. It provides information on setting up omake, gives an overview of building
programs using omake, and describes omake usage. The appendices contain information on
error messages, macros and rules, compatibility, and regular expressions.
If you are already familiar with the build process and makefiles, the recommended sequence for
proceeding through the documentation is:
➤ Read Chapter 1, Introduction, for configuration and usage information

➤ Read the first section of Chapter 2, Overview of Using omake
➤ Refer to Chapter 3, omake Reference, for information on omake-specific features
➤ Read the remaining chapters of this book
If you are not familiar with the build process and makefiles, the recommended sequence for
proceeding through the documentation is:
➤ Read Chapter 2, Overview of Using omake

➤ Read Chapter 1, Introduction, for configuration and usage information
Building Software with ClearCase provides additional information about using the clearmake and
omake build tools.
Preface xv
ClearCase Documentation Roadmap
Project
Development Management
Developing Managing
Software with Projects with
ClearCase ClearCase
Orientation
Introduction to
ClearCase
ClearCase and
MultiSite Release
Notes ClearCase Build
ClearCase Administration Management
Tutorials
Administering ClearCase
ClearCase OMAKE Manual
(Windows)
ClearCase
Product Family Building
Installation Notes Software with
ClearCase
ClearCase
MultiSite Manual
More Information
ClearCase Reference
Manual
ClearCase Quick
Reference Guide
ClearCase Online Help
clearcase.rational.com
xvi ClearCase OMAKE Manual

Typographical Conventions
This manual uses the following typographical conventions:
➤ ccase-home-dir represents the directory into which the ClearCase Product Family has been
installed. By default, this directory is C:\Program Files\Rational\ClearCase.
➤ attache-home-dir represents the directory into which ClearCase Attache has been installed.
By default, this directory is C:\Program Files\Rational\Attache, except on Windows 3.x,
where it is C:\RATIONAL\ATTACHE.
➤ Bold is used for names the user can enter; for example, all command names, file names, and
branch names.
➤ Italic is used for variables, document titles, glossary terms, and emphasis.
➤ A monospaced font is used for examples. Where user input needs to be distinguished
from program output, bold is used for user input.
➤ Nonprinting characters are in small caps and appear as follows: <EOF>, <NL>.
➤ Key names and key combinations are capitalized and appear as follows: F1, SHIFT,
CTRL+G.
➤ [] Brackets enclose optional items in format and syntax descriptions.
➤ {} Braces enclose a list from which you must choose an item in format and syntax
descriptions.
➤ | A vertical bar separates items in a list of choices.
➤ ... In a syntax description, an ellipsis indicates you can repeat the preceding item or line
one or more times. Otherwise, it can indicate omitted information.
NOTE: In certain contexts, ClearCase recognizes “...” within a pathname as a wildcard, similar
to “*” or “?”. See the wildcards_ccase reference page for more information.
➤ If a command or option name has a short form, a “medial dot” ( ⋅ ) character indicates the
shortest legal abbreviation. For example:
lsc·heckout
This means that you can truncate the command name to lsc or any of its intermediate
spellings (lsch, lsche, lschec, and so on).
Preface xvii
Online Documentation
The ClearCase graphical interface includes a standard Windows help system.
There are three basic ways to access the online help system: the Help menu, the Help button, or
the F1 key. Help➔Help Topics provides access to the complete set of ClearCase online
documentation. For help on a particular context, press F1. Use the Help button on various dialog
boxes to get information specific to that dialog box.
ClearCase also provides access to full “reference pages” (detailed descriptions of ClearCase
commands, utilities, and data structures) with the cleartool man subcommand. Without any
argument, cleartool man displays the cleartool overview reference page. Specifying a command
name as an argument gives information about using the specified command. For example:
> cleartool man (display the cleartool overview page)
> cleartool man man (display the cleartool man reference page)
> cleartool man checkout (display the cleartool checkout reference page)
ClearCase’s –help command option or help command displays individual subcommand syntax.
Without any argument, cleartool help displays the syntax for all cleartool commands. help
checkout and checkout –help are equivalent.
> cleartool lsprivate –help

Usage: lsprivate [-tag view-tag] [-invob vob-selector] [-long | -short]
[-size] [-age] [-co] [-do] [-other]
Additionally, the online ClearCase Tutorial provides important information on setting up a user’s
environment, along with a step-by-step tour through ClearCase’s most important features. To
start the ClearCase Tutorial, choose Tutorial from the Getting Started tab of the ClearCase Home
Base.
Technical Support
If you have any problems with the software or documentation, please contact Rational Technical
Support via telephone, fax, or electronic mail as described below. For information regarding
support hours, languages spoken, or other support information, click the Technical Support link
on the Rational Web site at www.rational.com.
xviii ClearCase OMAKE Manual

Your Location Telephone Facsimile Electronic Mail
North America 800-433-5444 408-863-4194 support@rational.com

toll free or Cupertino, CA
408-863-4000 781-676-2460
Cupertino, CA Lexington, MA
Europe, Middle +31-(0)20-4546-200 +31-(0)20-4546-201 support@europe.rational.com

East, and Africa Netherlands Netherlands
Asia Pacific 61-2-9419-0111 61-2-9419-0123 support@apac.rational.com

Australia Australia
Preface xix
xx ClearCase OMAKE Manual
1 Introduction
1
omake is a programming tool that helps you maintain programs, particularly those that are
constructed from several component files. omake controls the entire process of building your
program, whether it involves preprocessing files, compiling, linking, or other steps. It keeps
track of the project files, recompiling and relinking them only when required. Moreover, omake
maintains important information, such as compiler and linker options, in an editable text file
called a makefile. If you've ever returned to a project after a long absence, you know how hard
is to remember all the parameters that were used the last time the project was built. omake
remembers everything about the project for you.
ClearCase includes two independent build programs: clearmake and omake. Both incorporate
the major ClearCase build-related features, including configuration lookup, derived object sharing,
and configuration record maintenance. The omake program’s strength lies primarily in its support
for users who require compatibility with other PC-based build programs, including Borland®
Make, Microsoft® NMAKE, PolyMake, and Opus Make.
NOTE: omake is intended for use in dynamic views. You can use omake in a snapshot view, but none
of the features that distinguish it from ordinary make programs — build avoidance, build
auditing, derived object sharing, and so on — works in snapshot views. The rest of the
information in this manual assumes you are using omake in a dynamic view.
1.1 File Manifest

omake includes the following files, which are installed in ccase-home-dir\bin:
builtins.* PVCS Configuration Builder and NMAKE emulation rules and macros
omhelp.* PVCS Configuration Builder and NMAKE help screens
1 - Introduction 1
make.ini Sample omake initialization file
omake.exe omake executable
1.2 Configuring omake

omake reads an initialization file, ccase-home-dir\bin\make.ini, which you can configure. The
default make.ini file gives you a minimal configuration. Installing a new version of ClearCase
replaces your make.ini file; if you have customized this file and want to maintain your changes
after upgrading to a new version of ClearCase, back up the file before installation and restore it
afterward from your backup copy.
One of the major features of omake is its ability to emulate either PolyMake/PVCS
Configuration Builder or Microsoft NMAKE. The emulation is enabled with a command-line
option. See PM/CB Emulation on page 152 and NMAKE Emulation on page 158 for more
information. Throughout this document, PM/CB refers to PolyMake/Configuration Builder,
and NMAKE refers to Microsoft NMAKE.
1.3 Notes on Using omake

The ClearCase omake utility is based on Opus Config Builder 6.10, from OPUS Software, and is
compatible with most of its makefiles. It has been extended to include the fundamental
ClearCase build features. This section describes the ClearCase extensions and offers some basic
usage guidelines. Other sources of help on omake:
➤ omake reference page (in the ClearCase Reference Manual)

➤ omake –h
➤ omake –EN –h (help on NMAKE compatibility mode)
➤ omake –EP –h (help on PolyMake compatibility mode)
For more information on the ClearCase build model, see Building Software with ClearCase and the
clearmake reference page.
omake’s Configuration Management Features
The features described here constitute the core of ClearCase build configuration management.
2 ClearCase OMAKE Manual

Configuration Lookup
ClearCase omake uses a build-avoidance scheme based on configuration lookup that is more
sophisticated than the conventional make and Opus Make schemes, which are based on time
stamps of built objects. For example, the omake build-avoidance scheme guarantees correct
build behavior as C-language header files change, even if the header files are not listed as
dependencies in the makefile. By default, omake uses configuration lookup when you are
working in a view, and the target is located in the MVFS.
Derived Object Sharing
Developers working in different views can share the MVFS objects created by omake builds. See
the clearmake reference page for details.
Configuration Record Creation
omake audits the build operations, generating software bill-of-materials records, called
configuration records, that fully document a build. In the process, omake also performs automatic
dependency detection. By default, configurations records are created whenever you run omake
from a ClearCase view, and derived files are stored in the MVFS. See the clearmake and catcr
reference pages for details.
Express Builds
You can use either omake or clearmake during an express build, which is a build that creates
nonshareable derived objects. Express builds do not write DO information to the VOB, and
therefore enhance site-wide performance because they do not block other users’ access to the
VOB. For more information on express builds, see Building Software with ClearCase.
Command-Line Options
➤ The following options have been added to omake:

–L Disables configuration lookup and config record creation. See also clearmake –F.
–O Disables build script checking during configuration lookup.
–v Increases output verbosity during the build process (particularly with respect to
configuration lookup and config record creation).
NOTE: The Opus Make –V option, which prints the version string, has been
removed.
–W Disables shopping for derived objects to wink in. See also clearmake –V.
–T Examines sibling derived objects when determining whether a target object in a
VOB can be reused. See also clearmake –O.
1 - Introduction 3
Location of Temporary Files
The CCASE_AUDIT_TMPDIR environment variable controls the location of build audit temporary
files. By default, these files are stored in the directory named by the TMP environment variable.
If you do not set CCASE_AUDIT_TMPDIR, make sure that TMP is set to a valid temporary storage
directory on a FAT, NTFS, or LAN Manager file system. Note that setting CCASE_AUDIT_TMPDIR
to a location on a remote file system increases build times.
Non-MVFS Dependencies
omake supports local and remote non-MVFS dependencies:
➤ If the dependency resides on a local drive, you can use the local drive letter. However, to
share the applicable derived object, users on other hosts must access the dependency using
the same drive-specific pathname.
➤ If the dependency resides on a remote host, you must use UNC names. (If you use drive
letters, only users with identical network drive configurations can share the resultant DOs.)
Differences Between omake and Standard Make Tools

➤ $? means all dependencies, not just out-of date dependencies for targets for which
configuration lookup is used.
➤ Double-colon (::) rules. If any instance of a double-colon rule is out of date (based on
configuration lookup), all double-colon build scripts for the affected target are executed.
Conversely, if an error occurs during a build of the target of a double-colon rule, the target’s
remaining build scripts are not executed, and no config records or derived objects are
constructed. For example:
install::foo.exe
copy foo.exe \inst\bin
install::bar.exe
copy bar.exe \inst\bin
If foo.exe or bar.exe is out of date, both build scripts are executed. (Both programs are
copied.)

Running omake
Typically, you run omake from a view using the following procedure:
1. Set a view context by assigning a drive to a view (with Tools➔Map Network Drive in
Windows Explorer or the net use command) and changing to that view:
c:\> net use f: \\view\myview
c:\> f:
f:\>
2. Change to the appropriate directory and run omake:

f:\> cd myvob\src (`\myvob’ is the VOB-tag)
f:\myvob\src> omake options
You are using a view context to prevent VOB pathnames from being dependent on the view
the build occurs in. From f:, you (and your makefiles) can access versioned objects with
non-view-extended, absolute VOB pathnames such as \vob2\src\main.c in either cleartool
commands or standard operating system commands.
If you work directly on M:, in view-extended namespace, full pathnames to VOB objects
include a view-specific prefix, which can affect configuration lookup so as to prevent DO
sharing between views.
Makefiles
There are several rules to follow when constructing, or converting, makefiles for use by omake
on a Windows NT host.
Supporting Both omake and clearmake
It is possible, but not trivial, to prepare makefiles that can be used with either omake or
clearmake. The general approach is to supply omake-specific macro definitions in the makefile,
and to supply clearmake-specific macro overrides in a build options specification (BOS) file;
clearmake reads the BOS file, but omake does not. When clearmake executes, it looks for macro
definitions in two locations:
1 - Introduction 5
➤ %HOME%\.clearmake.options
➤ makefile.options, in the same directory as makefile (substitute the actual name of your
makefile, if it is not makefile)
BOS files at other locations can be passed to clearmake with the –A option.
Parallel and Distributed Build Operations
Each omake build executes serially on the local Windows NT host. Parallel and distributed build
operations are not available.
Build Scripts and the rm Command
It is common for a makefile to include a target whose build script invokes a command such as
rm to delete files. Some Windows NT installations include rm commands that do not actually
delete a file, but move it to a deleted directory instead. As a result, build script temporary files
become sibling DOs of the targets. To avoid this problem, use a remove command—del, for
example—that actually deletes files.
Pathnames in CRs
In a config record created on Windows NT, MVFS object pathnames begin with the VOB-tag and
do not include view-tag prefixes. For example:
...
---------------------------- MVFS objects: ----------------------------
\proj1\include\cmsg.h@@\main\nt3\39 <22-Jul-94.17:49:53>
\proj1\lib\fsutl.h@@\main\12 <22-Jun-94.12:07:24>
...
Pathnames in this format can be cut and copied, and applied elsewhere as is, if you are on a drive
assigned to a view (with Tools ➔ Map Network Drive in Windows Explorer or the net use
command).

1.4 Auditing 16-bit Tools
Compilers, linkers, and other tools written to run on MS-DOS or Windows (16-bit tools) require
special handling when used in audited builds with omake.
The program vdmaudit allows auditing of 16-bit tools. To use vdmaudit, you need to have
omake run vdmaudit and let it call the 16-bit tool. This involves either editing the makefile
where it calls the 16-bit tool, or if your makefile defines a macro for each 16-bit tool, redefining
the macros in the makefile or on the omake command line.
If the makefile defines a macro for each 16-bit tool, you can change the macros to call vdmaudit.
For example, if your makefile contains macros like
CPP=cl.exe
LINK=link.exe
change them as follows:
CPP=vdmaudit cl.exe
LINK=vdmaudit link.exe
You can redefine the macros on the omake command line like this:
omake –EN –f disptree.mak CPP="vdmaudit cl.exe" LINK="vdmaudit link.exe"
Call all 16-bit tools from vdmaudit; if you do not, omake does not audit all tools and the
configuration record is incomplete.
An alternative method for auditing 16-bit tools is to use the ClearCase Virtual Device Driver
(VDD). To install the VDD during ClearCase installation, select 16-bit build auditing on the
ClearCase Client Options or ClearCase Server Options page. The VDD runs any time a 16-bit
tool is run, whether during an audit or not. However, the VDD can cause 16-bit tools to fail to
display all output or to fail to clear the screen when done.
1 - Introduction 7
2 Overview of Using omake
2
This chapter provides an overview of using omake to build and maintain programs. It describes
the process omake uses to build programs, the files you use to control omake’s operation, and
some selected features of omake.
2.1 The omake Process

omake works in two distinct phases:
➤ read time, when it reads the files (called makefiles) that provide it with a project description
➤ run time, when it builds the project
A project consists of one or more targets. A target is a thing that can be made, usually a file, such
as a source file, object file, or executable file. The target is said to exist if the file is present on disk.
The pathname to a target is its location on disk. The target’s configuration record (CR) is a software
bill of materials record that fully documents the build of the target and supports omake’s ability
to rebuild. The CR includes the following information:
➤ Which MVFS file system objects (and which versions of those objects) were used by the
build as input data
➤ The commands used to perform the build
➤ Which MVFS files were created as output
The target's time stamp is the time the target was last changed. This is usually the creation or
modification time of the file.
2 - Overview of Using omake 9

Reading the makefiles gives omake dependency information. A dependency is the relationship
between a target and the things needed to make it. For example, an executable file that is linked
from a set of object files must be relinked if any of the object files change; thus, the executable file
depends on the object files. One way to think of this dependency information is as a data tree,
which may be several levels deep. For example, an executable proj.exe may depend on main.obj
and io.obj, each of which depend on source files main.c and io.c, respectively.
The things a target depends on are its dependencies. In fact, dependencies are also targets because
they can be (and are) made.
To make or build a target means to bring it up to date. A build script (sometimes referred to as a
shell line) is a command that omake executes to bring a target up to date. For example, a link
command updates an executable file. The command returns an exit status, which lets omake
know whether it succeeded. If the command succeeds, omake generates and stores a
configuration record. This configuration record is attached to each file modified or created by
executing the build script. A file produced by the build and associated with a configuration
record is called a derived object (DO).
If the target (derived object) is located in the MVFS, omake uses configuration lookup to
determine whether it is necessary to rebuild a derived object by executing a build script or by
reusing an existing instance of the derived object. omake traverses the dependency tree and
compares the configuration record of each target to the build configuration of the current view
(the set of dependency versions, the current build script, and the current build options),
rebuilding the target if necessary. For each target, the tree is traversed to the bottom and back up.
If the target is not in a VOB, omake compares the time stamp of the target to those of its
dependencies. If the target is older than any of its dependencies, omake rebuilds it.
Note that if you are working in a snapshot view, omake does not perform configuration lookup
and it does not create derived objects or configuration records. For more information about
snapshot views, see Developing Software with ClearCase.
NOTE: In this chapter, the examples assume that the files are in a VOB.
2.2 Invoking omake from the Command Line

omake is invoked from the command line with:
omake [ option | macro | target | @file ] ...

Each option is indicated by a – (dash) or / (slash) followed by a single letter, which is the option
name. Options can be grouped; that is, –k –v is the same as –kv. The case of the option name is
important. omake’s command-line options are described in Command-Line Options on page 41.
Each macro is a macro definition of the form name=[value]. Macros defined on the command line
take precedence over macros defined in the makefile. See Macros on page 21.
Each target is a target you want omake to build. If no targets are listed on the command line,
omake builds the makefile's default target.
Each @file directs omake to read file for additional options, macros, targets and @files. The @file is
treated as if the contents of file were on the command line except that each end-of-line character
in file is treated as a space.
You do not have to group options, macros, or targets on the command line; you can intermix
them. However, omake reads makefiles in order of their placement on the command line, and
builds targets in order as well. If the same macro or parameter is defined more than once on the
command line, the last definition is the one omake uses.
2.3 Initialization Files and Makefiles

omake reads its instructions from an initialization file, which holds instructions used to
customize the general operation of omake, and from a makefile, which has instructions for a
specific project.
The Initialization File
The initialization file contains instructions for all make programs and is used to customize the
operation of omake. omake reads the initialization file, ccase-home-dir\bin\make.ini, whenever
it starts up.
The initialization file contains targets, inference rules, and macro definitions that you do not
want to duplicate in every makefile, and it is used to customize the operation of omake. The
initialization file is processed like a makefile, with the following exceptions:
➤ The first target in the first makefile is the default target and is built if no command-line
targets are specified. There is no special significance to the first target in the initialization
file.

➤ Targets defined in the initialization file are redefined by targets of the same name that are
defined in the makefile. That is, these targets are available to all make programs, but they
can be redefined in the makefile.
The Makefile
The makefile has instructions for a specific project. The default name of the makefile is literally
makefile, but you can specify a different name with the –f option. The –f file command-line
option names the makefile, and several –f options can be given; each makefile is read in the order
it appears on the command line. If there are no –f command-line options, omake tries the file
named makefile. If it doesn't exist, omake tries makefile.mak.
Continued Lines
Lines in the makefile can be very long; the total makefile line length is limited only by available
memory. For easier reading, a long line can be broken up by typing \<ENTER> after part of the
line and typing the rest of the logical line on the next physical line of the makefile. For example,
the line
first_part_of_line second_part_of_line
is the same as
first_part_of_line \
second_part_of_line
To have the backslash (\) as the last character on the line, use \\<ENTER> or \<SPACE>,<ENTER>.
Comments
The simplest makefile statement is a comment, which begins with the comment character (#).
Everything on a makefile line after the # is ignored. Use \# for a literal # character. The following
large comment may appear in a makefile to describe its contents:
#
# Makefile for omake
#
# Compiler: Microsoft Visual C++ 5.0
# Linker: Microsoft(R) 32-Bit Incremental Linker Version 5.00.7022a
#

The comment character can also be used at the end of another makefile statement:
some makefile statement # a comment
If \<ENTER> appears on a commented line, the comment acts until the end of the line and the
following line is still continued. The following examples are equivalent:
line_one \
line_two # more_line_two \
line_three
line_one line_two line_three
Contents of Initialization Files and Makefiles
With a few exceptions, the initialization file holds the same kind of information as a makefile.
Both the initialization file and makefiles consist of these components:
➤ Rules
➤ Dependency lines
➤ Build scripts
➤ Macro definitions
➤ Inference rules
➤ Response files
➤ Directives
These components are described in the following sections.
2.4 Rules
A rule tells omake both when and how to make a file. For example, suppose your project involves
compiling source files main.c and io.c, and linking them to produce the executable project.exe.
The following makefile manages the task of making project.exe:

project.exe : main.obj io.obj
link /out:project.exe main.obj io.obj
main.obj : main.c
cl /c main.c
io.obj : io.c
cl /c io.c
This makefile shows three rules, one each to make project.exe, main.obj, and io.obj. These rules
are called explicit rules because they are supplied in the makefile. omake also has inference rules
that generalize the make process. Inference rules are discussed in Inference Rules on page 29.
2.5 Dependency Lines

Lines that include the colon (:) are called dependency lines. They specify a relationship between
targets and the files on which they depend. This is the general form of a dependency line:
target [ target ... ] [ attribute ... ] : [ dependency ... ]
To the left of the colon is the target of the dependency. To the right of the colon are the dependencies
needed to make the target. The target depends on the dependencies. For example, the following
line states that project.exe depends on main.obj and io.obj:
There can be one or more targets, optionally a list of target attributes, a colon, one or more space
or tab characters, and an optional list of dependencies. (Other make programs may call the
dependency a source, dependent, or prerequisite.) The first target name must start in the first
column of the makefile.
Dependency lines specify when to make the target. omake uses configuration lookup to
determine whether to build the target or reuse an existing derived object.
Explicit and Inferred Dependencies
The dependencies that are listed explicitly on dependency lines are called explicit dependencies.
For example, the line

test.exe : main.obj sub.obj
declares that test.exe depends on main.obj and sub.obj. When omake builds test.exe, it uses
configuration lookup to compare test.exe with your current build configuration to see whether
it must be updated.
A dependency line can declare that several targets have several dependencies. In the following
line, both main.obj and sub.obj depend on both system.h and io.h:
main.obj sub.obj : system.h io.h
omake also has inference rules to infer a dependency of a particular target. Sources determined
using inference rules are called inferred dependencies. For a discussion of inference rules, see
Inference Rules on page 29.
Macros in Dependency Lines
The name of the target (and its root name) can be referenced in a dependency line through the
use of macros. For example:
main.obj sub.obj : $*.h io.h
omake processes the target main.obj first; the expression $* evaluates to the root of the target
name, main. The dependency line declares that main.obj depends on main.h and io.h. Next,
sub.obj is processed, with $* evaluating to sub. Thus, this line also declares that sub.obj depends
on sub.h and io.h.
The Make Process Is Recursive
A basic feature of omake is to make a target's dependencies before the configuration lookup is
performed for the target. The following line instructs omake to make main.obj and io.obj before
performing configuration lookup for project.exe:
This line instructs omake to make main.c before performing configuration lookup for main.obj:
main.obj : main.c

Detected Dependencies
Unlike standard make variants, omake does not require you to declare source-file dependencies
in the makefile; omake detects dependencies automatically. This feature guarantees, for
example, correct build behavior as C-language header files change, even if the header files are
not listed as dependencies in the makefile. However, the list of dependencies must include
build-order dependencies, for example, object modules and libraries that must be built before
executables.
NOTE: You may want to include source-file dependencies in your makefile to ensure portability
to another group or company that is not using omake.
Wildcards in Dependency Lines
The dependency side of a dependency line can use a wildcard specification such as this:
main.exe : *.obj
omake provides this feature, but using it can cause problems if a required file is missing. For
example, the line
main.exe : *.obj
link /out:$(.TARGET) $(.SOURCES)
works fine until a required .obj file is accidentally deleted. omake works until the next update of
main.exe, when the link command isn’t called with all the required object names.
The Dependency Line Separator
The colon (:), which separates targets and dependencies on a dependency line, is also the
character used as the drive separator in Windows NT. To distinguish this colon from the drive
separator, you must put white space in front of it or put a space (target:<SPACE>), a tab
(target:<TAB>), a semicolon (target:;), another colon (target::), or nothing (target:<ENTER>) after it.
We suggest putting at least one space before and after it.

A Dependency Example
Assume the program test.exe is linked from main.obj and sub.obj. These object modules are
compiled from source files main.c and sub.c, respectively. The makefile looks like this:
D-1 test.exe : main.obj sub.obj

link /out:test.exe main.obj sub.obj
D-2 main.obj : main.c
cl /c main.c
D-3 sub.obj : sub.c
cl /c sub.c
Line D-1 declares that test.exe depends on main.obj and sub.obj. Lines D-2 and D-3 declare that
main.obj depends on main.c and sub.obj depends on sub.c.
Assume that since the last build of test.exe, main.c has been changed and sub.c has not. Running
omake causes the first target in the makefile, test.exe, to be evaluated. test.exe depends on
main.obj, which depends on main.c. main.c doesn't depend on anything. Comparing the
configuration record of main.obj with the current view’s build configuration shows that
main.obj must be rebuilt using the selected version of main.c. This is done with the build script:
cl /c main.c
Next, sub.obj, which depends on sub.c, is evaluated. Configuration lookup shows that sub.obj
does not need updating.
Comparing the configuration record of test.exe with the current build configuration shows that
test.exe must be rebuilt with the selected version of main.obj. The following build script updates
test.exe:
link /out:test.exe main.obj sub.obj
Running the command omake again causes omake to display this message:
omake: 'test.exe' is up to date.

2.6 Build Scripts
All lines that immediately follow a dependency line and begin with white space are the target’s
build script. Each build script is a command or list of commands, such as compile or link, that
omake executes to update the target. The explicit rule for a target consists of the dependency line
plus the build scripts.
Build scripts appear in the following form:
target [ target ... ] [ attribute ... ] : [ dependency ... ]

build script
.
.
.
For example, the following makefile specifies that making project.exe requires running the
program link to link main.obj and io.obj. This build script is run only if omake determines that
project.exe needs to be rebuilt.

link /out:project.exe main.obj io.obj
A build script can appear on the dependency line by separating it from the last dependency by
a semicolon. For example, the following lines are equivalent:
test.exe : test.c ; cl /c test.c
test.exe : test.c
cl /c test.c
A target can have a build script with multiple lines, listed one after the other. For example:

echo Linking project.exe
link /out:project.exe main.obj io.obj >link.out
The first line shows that Windows NT command processor commands can be executed by
omake. The second line shows redirection of output, where the output of the link program is
redirected to the link.out file.

Build Script Execution
omake executes build scripts in one of two ways:
➤ Direct execution of the shell-line command as an .exe file.
➤ By passing the build script to a shell program such as cmd.exe. This method is slower and
takes additional memory, but is necessary if the build script requires a feature that only the
shell program can provide.
Each build script runs in its own child process or subshell of omake, starting in the current
directory. This behavior has special consequences for commands that are supposed to have effect
between build scripts. For example, the build scripts
copy_to_tmp :
# Comment and blank lines are allowed
chdir some_directory
copy *.* c:\temp
copy the files that match *.* in the current directory to c:\temp. This occurs because the copy
command starts in the current directory, irrespective of the chdir command executed before it.
See Using Multiple-Command Build-Script Lines on page 73 for the correct way to do this.
Auto-Detection Mode
By default, omake detects when to use the shell program (defined by the .SHELL directive; see
Table 15 on page 92) and when to use direct execution. The shell program is used when the
command is or does one of the following:
➤ A Windows internal command
break, call, cd, chdir, cls, copy, ctty, date, del, delete, dir, echo, erase, for, if, md, mkdir, path,
pause, prompt, rd, rem, ren, rename, rmdir, set, time, type, ver, verify, vol
To change the list of internal commands, use the SHELLCOMMANDS macro. See Table 5
on page 67.
➤ A batch file (a file with a .bat or $(SHELLSUFFIX) extension
➤ A multiple-command shell line
➤ Uses redirection of input or output

Standard Execution Mode
The standard execution mode is to use the shell program for every build script. For Windows NT,
the default shell program is named in the COMSPEC environment variable. If this EV is not
defined, cmd.exe is used.
omake can execute the build script without using the shell program if the command being run
is executable and the build script does not use I/O redirection. Use the colon (:) build-script
prefix to have omake execute the build-script line without using the shell program (see Select the
Shell Program on page 71).
Build-Script Line Exit Status
After it executes each build-script line, omake checks the command exit status. The exit status is
a number the program returns and is tested by omake. At the Windows NT command line, you
can check the exit status of the last executed program by using the if errorlevel command.
By convention, programs return a zero exit status when they finish without error and a nonzero
status when an error occurs. The first build-script line that returns a nonzero exit status causes
omake to display this message:
omake: shell line exit status exit_status. Stop.
This usually means that the program being executed failed. Immediately after displaying this
message omake does any necessary deinitialization and exits. You can control this behavior with
the –i or –k command-line options (see Command-Line Options on page 41), with build-script line
prefixes (see Build-Script Line Prefixes on page 69), or with target attributes (see Target Attributes
on page 101).
Some programs return a nonzero line exit status inappropriately; you can have omake ignore the
exit status by using a build-script prefix. Prefixes are characters that appear before the program
name and modify the way omake handles the command script. For example:

– link /out main.obj io.obj
The dash (–) prefix causes omake to ignore the exit status of that build-script line. If the exit status
is nonzero, omake displays the following message:
omake: Shell line exit status exit_status (ignored)

See Build-Script Line Prefixes on page 69 for more information.
2.7 Macros
A macro is a makefile line that consists of a macro name, an equal sign ( = ), and a macro value. In
the makefile, a macro reference is an expression of the form $(name) or ${name}. The reference is
macro-expanded to produce value. When name is a single character, the ( ) or { } around name are
optional; for example, $X, $(X), and ${X} all specify the value of macro X. The macro character ($),
always instructs omake to expand the macro that follows. When you need a dollar sign in your
makefile, you must use $$.
In the following makefile, the text main.obj io.obj occurs more than once. To reduce the
amount of repeated text, you can use a macro definition to assign a symbol to the text.

link /out main.obj io.obj
main.obj : main.c
cl /c main.c
io.obj : io.c
cl /c io.c
Here is the same makefile written with the introduction of four macros:
OBJS = main.obj io.obj

CC = cl
CFLAGS = /c
project.exe : $(OBJS)
link $(OBJS) /out : project.exe
main.obj : main.c
$(CC) $(CFLAGS) main.c
io.obj : io.c
$(CC) $(CFLAGS) io.c
The value of the OBJS macro is the list of object files to be compiled. The macro definitions for
CC and CFLAGS make it easier to change the name of the C compiler and its options.

omake imports environment variables as macros, so you can refer to things like $(COMSPEC) or
$(PATH) in your makefile without having to define them in the makefile.
Macro Precedence
Macros can be defined in makefiles, defined on the command line, or predefined by omake.
omake also accesses the values of environment variables as if they were macros.
Where a macro is defined determines its precedence. To redefine an existing macro, the new
definition must have at least as high a precedence. This is the order of precedence:
1. Macros predefined by omake
2. Command-line definition
3. Environment definition (with –e command-line option)
4. makefile (and make.ini) definition
5. Environment definition (default)
By default, the lowest precedence for a macro is the environment definition. The –e
command-line option gives the environment definition a higher precedence than the makefile
definition. With this option, a macro definition in a makefile does not redefine a macro from the
environment.
Defining Macros in the Makefile
Macros are defined at read time in a makefile with macro definition lines of the form
name = [ text ] standard definition
name ?= [ text ] conditional definition
name := [ text ] expanded definition
name += [ text ] appended definition
where name is the macro name starting in the first column of the makefile. The name can include
any characters except the equal sign (=), the colon (:), and white space. By convention, the name
is composed of uppercase letters, periods (.), and underscores (_). Any macro references in name
are expanded; if you want a literal dollar sign in name you must use $$.

The text is arbitrary text and can reference the value of other macros with expressions of the form
$(other_macro). There are four forms of macro definition:
Read-Time Expansion of Macros
At read time, some parts of a makefile are macro-expanded; other parts are not expanded until
later.
For macro definitions, expansion depends on the separator between the macro name and the value:
= macro name is expanded; value is not expanded until referenced
(In NMAKE compatibility mode, value is expanded before assignments)
?= macro name is expanded; value is not expanded until referenced
+= macro name is expanded; value is not expanded until referenced
:= macro name is expanded; value is expanded
The makefile can also contain conditional directives such as these:
%if condition
%elif condition
The condition is macro-expanded and evaluated only if the previous enclosing condition is true.
For example:
%if condition_1 (expanded & evaluated)
% if condition_2 (expanded & evaluated only if condition_1 is true)
% endif
%else
% if condition_3 (expanded & evaluated only if condition_1 is false)
% endif
%endif
Standard Macro Definition: name = [ text ]
This definition defines macro name and sets its value to text. If text is not given, the value of the
macro is the null string, “”. White space before or after the = is ignored.
Conditional Macro Definition: name ?= [ text ]
This definition is like the standard macro definition, but the macro is defined only if it isn't
already defined. White space between name and ?= and between ?= and text is ignored.

Expanded Macro Definition: name := [ text ]
This definition defines macro name and sets its value to the macro-expansion of text. The text is
arbitrary text and can refer to the value of other macros with expressions of the form
$(other_macro), which are expanded. If text is not given, the value of the macro is the null string
(“”). White space between name and := and between := and text is ignored. Expanded macro
definitions are useful when the text is expensive to calculate (it may require reading a file, for
example).
Appended Macro Definition: name += [ text ]
This definition appends text to the current value of macro name. White space between name and
+= is ignored. If there is no white space between += and text, the new value is OLDVALUEtext.
Otherwise, the new value is OLDVALUE <SPACE> text (with one intervening space). If name is an
undefined macro, this definition is the same as a standard macro definition.
Case-Sensitivity of Macro Names
By default, for Windows NT omake macro names are case-insensitive. The .CASE_MACRO and
.NOCASE_MACRO directives turn case-sensitivity on and off.
The Location of Macro Definitions
By convention, macro definitions appear at the beginning of the makefile. Macros must be
defined before they are expanded; if they are not, their expanded value is the null string.
Indenting Macro Definitions
Usually name starts in the first column of the makefile line, but macro definitions can be indented
(for example, inside an %if...%endif conditional directive). Indenting is allowed only before the
first target in the makefile, or after a nonindented macro definition. This restriction is a
consequence of omake's use of indenting to indicate a target's build scripts.
Undefining Macros
Macros can be undefined with the %undef directive at read time:
%undef CFLAGS
Some macros that omake predefines cannot be undefined with this directive.

Example Macro Definitions
➤ This defines the macro OBJS as main.obj sub.obj. $(OBJS) is main.obj sub.obj.
OBJS = main.obj sub.obj
➤ This defines PROJECT as $$/Make. $$ is a literal dollar sign. This means that $(PROJECT)
is $/Make.
PROJECT= $$/Make
➤ These define the macro DEBUG as 7 and the macro CFLAGS as –Z$(DEBUG). $(DEBUG)
is 7 and $(CFLAGS) is –Z7.
DEBUG = 7 (defined as 7)
DEBUG ?= i (not redefined)
CFLAGS = –Z$(DEBUG)
➤ This shows an appended macro definition inside a conditional directive. If the OPT macro
is defined, the CFLAGS macro's value has <SPACE>–Od appended to it. The new value of
CFLAGS is –Z$(DEBUG) –Od and $(CFLAGS) is –Z7 –Od.
%if defined(OPT)
CFLAGS += –Od
%endif
➤ This defines macro OBJSLIST as project.lst and then defines OBJS. The := causes the right
side to be expanded. OBJSLIST is expanded, producing project.lst and the @ macro
modifier (see Macro Modifiers on page 52) reads the file project.lst. The contents of this file is
the value of OBJS. The expression := is useful here because reading the file is a relatively
slow process that you want done only once. If OBJS is defined with a standard macro
definition, each occurrence of $(OBJS) causes omake to read project.lst. By using :=,
project.lst is read only once, when OBJS is defined.
OBJSLIST = project.lst
OBJS := $(OBJSLIST,@)
Defining Macros on the Command Line
Macros can be defined on the command line, and the value of the command-line macro overrides
a makefile macro or environment definition with the same name. Only standard macro
definitions are allowed on the command line. A command-line macro that contains spaces must

be enclosed in double quotes. (To include a literal double quote, use \".) For example, the
command line
omake BSCFLAGS=/n "CFLAGS=-Zi -Od"
runs omake with BSCFLAGS defined with the value –n and CFLAGS defined with the value
–Zi –Od.
Dynamic Macros
omake defines some special macros whose values are dynamic. These run-time macros return
information about the current target being built. For example, the .TARGET macro is name of
the current target, the .SOURCE macro is the name of the inferred dependency (from an
inference rule) or the first of the explicit dependencies, and the .SOURCES macro is the list of all
dependencies.
Using run-time macros, the example can be written as follows:

CC = cl
CFLAGS =
link /out:$(.TARGET) $(OBJS)
main.obj : main.c
$(CC) $(CFLAGS) –c $(.SOURCE)
io.obj : io.c
As you can see, the build scripts that update main.obj and io.obj are identical when dynamic
macros are used. Dynamic macros are important for generalizing the build process with
inference rules, as shown in Inference Rules on page 29.
Macro Modifiers
Macros can be used to reduce the amount of repeated text. They are also used at run time to
generalize the build process with inference rules. You often want to start with the value of a

macro and modify it in some manner. For example, to get the list of source files from the OBJS
macro you can define this macro:
SRCS = $(OBJS,.obj=.c)
This definition uses the from=to macro modifier to replace the from text in the expansion of OBJS
with the to text. The result is that $(SRCS) is main.c io.c. In general, to modify a macro, expand
it with
$(name,modifier[,modifier ...])
Each modifier is applied in succession to the expanded value of name. Separate modifier with
comma.
File Name Components
There is a complete set of macro modifiers for accessing parts of file names. For example, with
this macro definition:
SRCS = \src\main.c parse.l
Table 1 lists some of the modifiers.
Table 1 Macro Modifiers
Modifier and description Example Value
D, the directory $(SRCS,D) \src .
E, the extension (or suffix) $(SRCS,E) .c.l
F, the file name $(SRCS,F) main.c parse.l
Tokenize
The Wstr modifier replaces white space between elements of the macro with str, a string. The str
can be a mix of regular characters and special sequences, the most important sequence being \n,
which represents a newline character (like pressing the <ENTER> key). For example:
$(OBJS,W +\n) is main.obj +

io.obj

Other Modifiers
Other modifiers include @ (include file contents), LC (lowercase), UC (uppercase), M (member),

and N (nonmember). The M and N modifiers and the S (substitute) modifier use regular
expressions for powerful and flexible pattern-matching. See Macro Modifiers on page 52 for more
information.
Environment Variables
Environment variables are placed into the environment with this command:
set name=value
NOTE: This example applies to cmd.exe; the command varies depending on the shell you use.
By default, omake preloads all environment variables as macros. The .NOENVMACROS

directive prevents this loading and hence prevents omake from accessing the value of any
environment variables.
Macro Expansion or Macro Referencing
Expanding a macro also expands any macro references recursively. For example:
CDEFS = –DDEBUG –DNT

COPTS = –Ot
CFLAGS = –Zi $(CDEFS) $(COPTS)
The expression $(CFLAGS) evaluates to –Zi –DDEBUG –DNT –Ot.
In a macro reference $(name), name can reference other macros. For example, given the definitions
SYS = NT
NTFLAGS = –DNT –UNT
the expression $($(SYS)FLAGS) evaluates to –DNT –UNT.

Run-Time Expansion of Macros
Macros in build scripts are expanded immediately before the build scripts are executed. There
are several macros whose values change dynamically at run time. They are discussed in
Predefined Macros: Run-Time Macros on page 60.
Recursive Macro Definitions
NOTE:NMAKE supports recursive macro definitions as shown in this section. When omake is
emulating NMAKE, recursive macro definitions are supported.
A macro value may reference other macros. If the value circularly references itself, omake
displays a warning message when the macro is expanded. For example, the values
A = A $B
B = B1 $A B2
%echo $A
generate this message:
omake: file (line num): Recursive macro 'A = A $B' (warning).

A B1 B2
The expression $A expands to A expansion_of_B. In turn, the expansion of B is

B1 expansion_of_A B2. When omake tries to expand a macro that is already being expanded, it
displays the warning message and ignores the recursive expansion. The file and num depend on
where the example lines were defined.
You can make this kind of recursive reference by using the expanded macro definition:
B := B1 $A B2
This type of definition expands any references in the right side before the macro definition of B
occurs.
2.8 Inference Rules

Inference rules generalize the build process, which eliminates the need to give omake an explicit
rule for each target. For example, compiling C source (.c files) into object files (.obj files) is a
common occurrence. Rather than require a statement that each .obj file depends on a like-named

.c file, omake uses an inference rule to infer that dependency. The dependency determined by an
inference rule is called the inferred dependency.
Inference rules also provide build scripts to update the target from the inferred dependency. The
target inherits these build scripts if it does not have its own.
omake predefines several inference rules, and you can change their definitions or define your
own rules.
Defining Inference Rules
Inference rules (also called metarules) are identified by the use of the rule character (%) in the
dependency line. This character is a wildcard, matching zero or more characters. For example,
here is an inference rule for building .obj files from .c files:
%.obj : %.c
This rule states that a .obj file can be built from a corresponding .c file with the build-script line
$(CC) $(CFLAGS) –c $(.SOURCE). The .c and .obj files share the same root of the file name.
When the dependency and target have the same file name except for their extensions, this rule
can be specified in an alternative way:
.c.obj :
The alternative form is compatible with other make utilities.
omake predefines the %.obj : %.c inference rule as listed above so the example now becomes
much simpler:

CC = cc
CFLAGS = /Zi
link /out:$(.TARGET) $(OBJS)

The General Inference Rule Definition
[ Tp ] % [ Ts ] [ attribute ... ] : [ Sp ] % [ Ss ]
build script
.
.
.
On the target side of the dependency line is a pattern: a target prefix Tp, a %, and a target suffix
Ts. Any target attributes (see Target Attributes on page 101) appear next. On the dependency side
of the line is a dependency prefix Sp, a %, and a dependency suffix Ss. The prefixes can contain
any character except % and, in particular, can be a directory. The suffixes can contain any
character (including %, which is taken literally).
In order for a rule to match a target, Tp and Ts must match the first and last parts of the target
name, with % matching everything in between. The inferred dependency name is Sp followed
by the characters matched by %, followed by Ss.
The Target Inherits Build Scripts and Attributes
Following the dependency line are the build-script lines that update the target from the inferred
dependency. If a target doesn't have its own build scripts, it inherits the build scripts and
attributes of the inference rule. Target attributes take precedence over rule attributes.
Alternative (Suffix-Only) Form
Inference rules can also be defined in this form:
.source_extension.target_extension :
build script
.
.
.
The source_extension is the extension of the dependency. The target_extension is the extension of
the target. This alternative form is compatible with other make utilities and is discussed in
Compatibility with Suffix Rules (.SUFFIXES) on page 37. The suffix-only form is converted by
omake to the equivalent metarule form:

%.target_extension : %.source_extension
build script
.
.
.
Automatic Use of Inference Rules
omake uses inference rules when building a target that has no build scripts. Even when targets
have build scripts, you can cause omake to use inference rules by giving the target the .INFER
target attribute.
In either case, omake first builds the target's explicit dependencies (those listed on dependency
lines), and then uses its inference rules to search for an inferred dependency. The search proceeds
by finding all rules that match the target, building each possible inferred dependency name in
turn, and checking whether the inferred dependency exists as a file. If it exists, omake proceeds
as follows:
1. If the target has no build scripts, the target inherits the inference-rule build scripts and
attributes. When this is a conflict between the attributes, the target's attributes take
precedence.
2. The inferred dependency is added to the target as a dependency.
3. The inferred dependency is built.
Inference Rules and Target Groups
Inference rules can build several targets (a target group) from a single dependency. To do this put
the targets on the target side of the rule, with a plus sign (+) between them. (There must be white
space between the + and the target names.) When omake executes the rule's build scripts to
update any target in the group, all targets are updated.
For example, here is a rule for building both .c and .h files with a yacc program, which takes a .y
file as input:

%.h + %.c : %.y
$(YACC) –d $(YFLAGS) $(.SOURCE)
copy y.tab.h $(.TARGET,1)
copy y.tab.c $(.TARGET,2)
del ytab.*
Note the use of $(.TARGET,num), which evaluates to the names of both the .h and .c files. The
num macro modifier selects the numth element of the macro. So, for example, $(.TARGET,1) is
parse.h when the inferred dependency is parse.y.
Multiple-Step Inference Rules
When omake tries to find the inferred dependency of a target, it first tries all rules that directly
produce the target. If the inferred dependency cannot be found with one rule, omake chains rules
into twos, threes, and so on. omake always chooses the smallest number of rules (the shortest
path) between a target and its inferred dependency.
If a multiple-rule path is found, omake creates the targets between the inferred dependency and
the target, and chains them. Chained targets have special properties:
➤ omake deletes chained targets as part of the build; as soon as it finishes running the build
scripts, it deletes any intermediate chained targets. A chained target is not considered a
sibling derived object.
➤ For the purposes of time stamp comparisons, a chained target is the same age as the target
that depends on it. This allows omake to compare the time stamp of the target at one end of
the chain with the time stamp of the inferred dependency at the other end and correctly
determine whether the target needs updating.
NOTE:Although the characters matched by % are the same within a single inference rule, the %
can match different characters for the different rules of a multiple-rule path.
Chained Targets Are Deleted Automatically
To illustrate how the deletion of chained targets works, here are rules for extracting a .obj file
from a .c file and for compiling the .obj file into a .exe file:
%.obj : %.c
$(CC) $(CFLAGS) -c $(.SOURCE)

%.exe : %.obj
$(LINK) /out:$(.TARGET) $(LINKFLAGS) $(.SOURCE)
Assume that omake is trying to build main.exe and that main.c exists, but main.obj does not.
omake finds the two-rule inference that uses the %.obj : %.c rule to produce main.obj, and the
%.exe : %.obj rule to produce main.exe. omake retrieves main.c and compiles it to produce
main.obj. It then links main.obj to produce main.exe. After running the %.exe and %.obj
build-script lines, omake deletes main.obj because it was marked as being chained.
To prevent this deletion, write a rule that combines the rules in the multiple-rule path but that
has a shorter path. A one-step rule between main.exe and main.c that leaves behind main.obj is
%.exe : %.c
$(LINK) /out:$(.TARGET) $(LINKFLAGS) $(.SOURCE)
A second way to prevent this deletion is to use the .PRECIOUS attribute. omake does not delete
targets with this attribute. Modify the %.obj : %.c rule as follows:
%.obj .PRECIOUS : %.c

Preventing Multiple-Step Rules
A target with the .NOCHAIN attribute instructs omake to try only the one-step paths for finding
the inferred dependency.
A rule with the .NOCHAIN attribute cannot be used in the middle of a multiple-step rule. That
is, it means this rule is terminal.
Inference Rule Search Order
To determine which inference rules to use, omake gathers all rules that can build the current
target and sorts them from best score to worst score. For each rule in this sorted list, omake
constructs the dependency name and tests whether it exists as a file. If so, omake chooses this
rule as the inference rule, and this file as the inferred dependency.
The score is defined as the number of characters in the target name that % matches; the best score
is zero. Often, many rules may have the same score (for example, all rules that build %.obj
targets), and omake puts these rules in creation order. For identical scoring rules, the rule created
first (usually built in to omake or in make.ini) is first on the list.

Overriding the Rule Ordering
Sometimes the rule ordering is inappropriate. For example, you may have both video.c and
video.cpp that can be used to build video.obj, and you want omake to use video.c. By making
video.obj depend on video.c, omake chooses the %.obj : %.c rule to update video.obj:
video.obj : video.c
If you find yourself doing this, you can prevent omake from doing the inference rule search by
supplying your own build scripts. For example:
video.obj : video.c
One problem with supplying the build scripts explicitly is that they must change if the
%.obj : %.c rule changes. To avoid this problem, use the %do directive to execute the build
scripts of the %.obj : %.c rule:
video.obj : video.c
%do "%.obj : %.c"
Rule Finding for Target Names with a Directory Component
When the target name has a directory component (for example, objs\video.obj), omake uses the
following methods to find a rule:
1. The rules with directory components in their targets are matched against the complete target
name.
2. If no rules match in Step #1, the rules without directory components in their targets are
matched against the file name of the target, by default. The .UNIXPATHS directive changes
this behavior. When you use .UNIXPATHS, the rules without directory components are
matched against the full target name.
Common Operations on Inference Rules
These operations on inference rules are performed most often:
➤ Redefining an inference rule
Inference rules can be redefined any number of times. To redefine an inference rule, provide
a new definition in one of two places:

➣ In make.ini, to give every makefile access to the rule
➣ In your makefile, if you want a local variant
➤ Changing an inference rule's attributes
To change an inference rule's attributes, use a dependency line with the new attributes and
don't specify a build script. These attributes are combined with the current attributes.
➤ Disabling an inference rule
To disable an inference rule, use a dependency line with no attributes and no build script:
%.obj : %.c
...or...
.c.obj :
➤ Disabling the search for an inferred dependency
omake searches for an inferred dependency for targets that have no build script. You can
omit the inference search with the .NOINFER attribute. For example:
all .NOINFER : import test export
The all target only drives the build of its dependencies, so no inference search is needed.
➤ Using % in nonrules
A % on the target side of a dependency line indicates that this line is an inference rule. To use
a literal % without indicating a rule, give the target the .NORULE attribute:
percent%.obj .NORULE : percent%.c
➤ Changing the rule character
The .RULE_CHAR directive sets the rule character. For example, to change it to an asterisk:
.RULE_CHAR : *
Built-In Inference Rules
omake predefines several rules. You can reject them by using the –r (reject rules) command-line
option or the .REJECT_RULES directive. The built-in rules are listed in Inference Rules on
page 142.

Compatibility with Suffix Rules (.SUFFIXES)
Many make utilities, such as PM/CB, NMAKE, and Borland Make, have a simple style of
inference rule, called suffix rules. These rules use only the suffix (extension) of the file name. Here
is a suffix rule:
.c.obj :
$(CC) $(CFLAGS) –c $<
Some make utilities also use a .SUFFIXES directive to control the search order of the suffix rules.
The .SUFFIXES directive lists the suffixes in the order that inference rules are to be attempted.
Each appearance of .SUFFIXES prepends to the current list of suffixes, but a .SUFFIXES directive
with nothing on the dependency side clears the list:
.SUFFIXES : (clears list)
.SUFFIXES : .lib (list is: .for .lib)
.SUFFIXES : .exe .obj (list is: .exe .obj .lib)
When looking for an inference rule to build a target, the list is traversed from left to right with
each extension being combined with the target extension to form the name of a suffix rule. If that
suffix rule exists, omake forms the name of the corresponding inferred dependency and, if it
exists, the inference search is complete.
Handling of Suffix Rules and .SUFFIXES
omake converts suffix rules into equivalent omake inference rules. If .SUFFIXES is used, omake
uses the suffixes order to sequence only its suffix rules. Nonsuffix rules are left in creation order.
By default, all inference rules are in creation order. Note that a .SUFFIXES directive with no
suffixes on the dependency side effectively disables all suffix rules, but nonsuffix rules remain
enabled.
2.9 Response Files

omake has two kinds of support for response files: automatic response files, where omake
determines when to build a response file, or inline response files, where statements that create
response file creating statements are written in the makefile.

Inline Response Files
Response files can also be coded inline in your makefile. For example:
project.dll:$(OBJS)
link @<<
$(OBJS)
/out:$(.TARGET)
$(.TARGET)
<<
The link program is invoked as link @response_file, where response_file is a name generated by
omake. The response_file contains:
main.obj io.obj /out:project.exe
2.10 Directives
Makefile directives control the makefile lines omake reads at read time. The following makefile
uses conditional directives (%if, %elif, %else, and %endif) to support both Borland and
Microsoft compilers. Comments have been added for documentation.
# This makefile compiles for the configuration specified

# in the CFG macro.
CFG = Release
# Configuration-dependent section
%if $(CFG) == Release
CFLAGS = /O1
LINK_FLAGS =
%elif $(CFG) == Debug
CFLAGS = /Od /Zi
LINK_FLAGS = /debug
%else
%abort Unsupported configuration $(CFG)
%endif

project.exe: $(OBJS)
link /out:$(.TARGET) $(LINK_FLAGS) $(OBJS)
%.obj: %.c
cl $(CFLAGS) /c $(.SOURCE)
The layout of this makefile is fairly traditional; macros are defined first, and the primary target
follows the macros.
This example also uses the %abort directive to abort omake if the makefile does not support a
particular compiler. Directives can also be used at run time, to control the build scripts omake
executes. For more information, see Makefile Directives on page 74.
2.11 The Keep-Working Mode

omake usually stops at the first build-script line that returns an error. This is the right behavior
for interactive work, but not for a long, unattended build such as an overnight build of your
entire system. The solution is to use the keep-working mode. This mode allows omake to do the
maximum amount of work consistent with any errors that occur during the build.
In the keep-working mode, a build script error while omake is updating a target causes omake
to stop working on the current target. The make process continues, with omake noting that the
current target was incompletely built. Future targets are updated only if they do not depend on
any incomplete targets.
Because omake does as much work as possible, you only have to fix the problem targets and start
omake again. The keep-working mode does the maximum amount of safe and correct work.
This mode is enabled with the .KEEPWORKING makefile directive (see Dot Directives on
page 91) or –k command-line option (see Command-Line Options on page 41).

3 omake Reference
3
This chapter contains reference information for omake.
3.1 Command-Line Options

Command-line options or flags are indicated by a dash (–) or slash (/) followed by a one-letter
option name. Some options take an argument and that argument follows the option name with
or without intervening space. Several options can be grouped together after a single dash, but
options that take an argument can only be at the end of the group. To signal the end of options,
a double dash (--) can be used in case a command-line target or macro has a dash as the first
character. Here are some examples:
omake –n –d –f project.mak –p
omake –ndpf project.mak
omake –f ../makefile -- –readme–
The first and second example are identical and show how options can be grouped. The third
example shows the use of the double dash to indicate the end of options. This is required to
prevent omake from interpreting –readme– as more options.
Initial Command-Line Parameters: OMAKEOPTS
Before parsing the command line, omake looks for the OMAKEOPTS environment variable and, if
it is found, parses its value. After reading the initialization file, omake also parses the value of
the MFLAGS macro for additional command-line parameters. MFLAGS can be defined on the
3 - omake Reference 41
command line, in make.ini or in the environment (but an MFLAGS definition in the makefile is
ineffective). Using the –z command-line option causes omake to ignore MFLAGS. Setting
MFLAGS in make.ini is strongly discouraged, because directives can be used instead. After the
OMAKEOPTS environment variable is parsed, the command line itself is parsed.
The Command-Line Options
The options in Table 2 are parameters for omake’s native mode. Many parameters are different
for NMAKE and PolyMake emulation modes.
Table 2 omake Command-Line Options (Part 1 of 4)
Option Meaning
–a Updates all targets whether or not they are out of date. If no targets are specified
on the command line, the default target is updated.
–A Uses automatic dependencies. This option is enabled only if you are not using
configuration lookup (because you are processing non-MVFS files or using the
–W option).
–b file Names file as the initialization (built-ins) file (The default is make.ini.) If file is
the empty string, no initialization file is read. You can specify this on the
command line with "–b<SPACE>", –b"<SPACE>" or –b<SPACE>"".
The value of the BUILTINS macro is set to the full pathname of file.
–d Run-time debugging mode. A run-time trace of omake operation displays

targets being made and rules being tried. See also the –p command-line option.
–D Keep-directory mode. The first access of the current directory or any search
directory to look for a file reads the directory into memory. Subsequent accesses
to the directory use the in-memory version and are much quicker.
–e Environment macro precedence. Macros defined from environment variables

have a higher precedence than makefile macros.
–EN Sets omake's emulation mode to NMAKE. Selecting this emulation causes the
rest of the command line to be interpreted as an NMAKE command line. See
Microsoft NMAKE Compatibility on page 158 for more information.
–EO Default emulation mode; that is, no emulation.

Option Meaning
–EP Sets omake's emulation mode to PM/CB. Selecting this mode causes the rest of
the command line to be interpreted as a PM/CB command line. See PM/CB
(Intersolv Configuration Builder and PolyMake) on page 145 for more information.
–E2 Sets omake's emulation to Opus Make v5.2x. See Opus Make v5.2x Compatibility
and Emulation on page 162 for more information.
–f file Specifies file as the makefile. More than one –f option may appear, and the files
are read in order. omake first tries to read file, but if file does not exist, omake
tries file.mak. If file is con or – the console is read for the makefile and
<CTRL+Z>,<ENTER> (<CTRL+D> for UNIX) as the last input finishes the input.
If no –f options are specified, omake tries makefile and then makefile.mak.
The value of the MAKEFILE macro is the file of the first –f option.
–G Restricts dependency checking to makefile dependencies only—those

dependencies declared explicitly in the makefile or inferred from an inference
rule. All detected dependencies are ignored. For safety, this option disables winkin
of DOs from other views; it is quite likely that other views select different
versions of detected dependencies.
For example, a derived object in your view may be reused even if it was built
with a different version of a header file than your view currently selects. This
option is mutually exclusive with –W.
–h Displays a help screen showing the command-line syntax and command-line

options. The current emulation mode determines the help that is displayed.
–i Ignores exit status, which causes the nonzero exit status from any build scripts
to be ignored. This option can be used to collect all errors into a single file
without stopping omake. See also the –k option.
Many compilers send their error messages to standard error. This
command-line option redirects standard error either into a file or to the
standard output. With it, the command
omake –i –x – > errs
collects all messages into the errs file. See the –x command-line option.
Option Meaning
–k Keep-working mode. Any errors when updating a target cause work on that
target to be stopped, but the make process continues. Because the target was
incompletely made, any other targets that depend on it are prevented from
being updated.
This mode is handy for long, unattended builds because it maximizes the
amount of making without completely ignoring the exit status as the –i
command-option does.
–L Disables configuration lookup and config record creation. See also clearmake
–F.
–M Makes makefile. omake makes the makefile before reading it. See the section
Makefile Directives on page 74.
–n No execute. Displays, but does not execute, the build scripts that update the
targets. This option is useful for debugging makefiles and for showing the work
that will be done. To override this option for a recursive make, use the .MAKE
target attribute (see Target Attributes on page 101) or & shell-line prefix (see
Build-Script Line Prefixes on page 69).
–O Disables build script checking during configuration lookup.
–p Prints debugging information to screen. Macros are listed with their location of
definition and value; search directories are indicated; inference rules are
displayed and targets, build scripts and attributes are printed. The
interpretation of the output is discussed in Chapter 4, Debugging Makefiles.
–r Rejects inference rules. omake ignores inference rules that are built in and that
are defined in the initialization file. Only inference rules defined in the makefile
are used.
–s Silent mode. Build scripts are usually displayed before they are executed. The
–s option prevents this display.
–T Examines sibling derived objects when determining whether a target object in a

VOB can be reused (is up to date). By default, when determining whether a
target can be reused, omake ignores modifications to objects created by the
same build rule that created the target (sibling derived objects). –T directs
omake to consider a target out of date if its siblings have been modified or
deleted. –T is equivalent to clearmake –R.

Option Meaning
–v Increases output verbosity during the build process (particularly with respect to
configuration lookup and config record creation).
–W Disables shopping for derived objects to wink in. –W is equivalent to clearmake

–V.
–x file Redirects error messages into file. As a special case, if file is –, the error messages
are redirected to the standard output. For example:
omake –x make.err (errors go into make.err)
omake –x – >> make.err (errors are appended to make.err)
–z Ignores MFLAGS. The MFLAGS macro is not examined for additional options.
–#1 Debugging option—read-time debugging mode. A read-time trace of omake

operation displays makefiles being read and conditional directives being
interpreted. See Chapter 4, Debugging Makefiles for an interpretation of the –#1
output.
–#2 Debugging option—warns about undefined macros. When omake tries to

expand the value of a undefined macro, it displays a warning message. The
default action is to silently ignore undefined macros.
–#4 Debugging option—warns about unrecognized makefile lines. When omake

reads a line in a makefile it can't understand, it displays a warning message. The
default action is to silently ignore unrecognizable makefile lines.
–#8 Debugging option—leaves behind generated response files and batch files.
omake uses several temporary files, which are usually deleted before omake
quits. This option causes omake to leave them behind so they can be examined.
Debugging features can be specified as a sum or as a comma-separated list. For example, both
–# 7 and –# 1,2,4 display makefile lines, warn about undefined macros, and warn about
unrecognized makefile lines). See also the description of the .DEBUG directive in the section Dot
Directives on page 91.
The Current Options
After reading its initialization file, omake exports all command-line options except –b and –f into
the MFLAGS macro (unless you are using omake in an emulation mode). Also, the state of each
command-line option is kept in the macros listed in the section Predefined Macros: State Macros on
page 63 and in Appendix C, Built-In Macros and Rules.
3.2 Locating the Initialization File

omake locates the initialization file in the following manner:
1. If the –b file command-line option is used, omake uses file.
2. If the OMAKECFG environment variable exists, omake uses its value.
3. Otherwise, omake searches for make.ini in the following locations:
a. The current directory
b. The directory of the omake.exe file
c. Directories specified in the INIT environment variable
Here, we assume the initialization file is named make.ini.
When the initialization file is found, omake searches its directory for its ancillary files (omhelp.*
and builtins.*).
Disabling the Initialization File
omake does not look for the initialization file if you use the command-line option –b"<SPACE>"
or if you set the OMAKECFG environment variable to the empty string. To set the environment
variable:
set OMAKECFG=<SPACE><ENTER>

3.3 The Make Process
This section describes the events that occur when you invoke omake.
Read Time
At read time, omake parses the command line, reads the initialization file, and reads one or more
makefiles. Steps in the start-up process:
1. If the OMAKEOPTS environment variable exists, it is parsed for command-line parameters.
2. The command line is parsed.
3. The initialization file is found and read.
4. If the MFLAGS macro is defined and the –z command-line option was not specified, the
macro is expanded and parsed as a command line.
5. Each makefile is read in the order it appears on the command line.
omake reads the initialization file and makefiles, collecting macro definitions, dependency lines,
rules, and build scripts. Dependency lines are built into a set of dependency trees, with each
target node connected by a dependency arc to each of its dependency nodes. Because each
dependency is itself a target, it is connected to its own dependencies. At the leaf nodes of the trees
are targets without any dependencies.
Run Time
At run time, omake makes the command-line targets or, if none are specified, the first normal
target in the first makefile. In this context, “normal” means not a directive, special target,
attribute, or inference rule. This first normal target is called the default target. Briefly, these are
the steps involved in making a target:
1. Locate the target.
Look for the target as a file on disk and, if found, record its location.
2. Make its dependencies.
Make the target's explicit dependencies.
3. Locate its inferred dependency.
If the target has no build scripts, use inference rules to look for an inferred dependency. If the
inferred dependency is found, add it as a dependency of the target and add the rule's build
scripts to the target. Then, make the inferred dependency.
4. Compare and update the target.
omake's build avoidance scheme includes automatic dependency detection. By default,

omake uses configuration lookup to compare the configuration record of the target to your
build configuration and determine whether a build is required. If the targets are not in a
VOB, omake compares time stamps to determine whether a build is necessary. If the time
stamp of any target dependency is more recent than that of the target itself, the target is
updated by executing its build script. After the build scripts have been executed, the target
is made.
If the target was found in Step #1, omake assumes its location stays the same. If the target was
not found, omake makes a second attempt to locate it on disk. In either case, omake also updates
its internal copy of the target's time stamp.
The consequence of Step #2 and Step #3 is that omake makes a target's dependencies recursively.
omake goes as far as it can down the dependency tree and works back up, using configuration
record matching and build avoidance, as described in Step #4. If the target is not in a VOB, omake
checks the time stamp of the target against its dependencies. For targets that have at least one
newer dependency, the target's build scripts are executed to update the target.
The utility of Step #3 is that it generalizes the make process. Inference rules are akin to statements
such as “all .obj files can depend on like-named .c or .cpp files.”
An extensive example that shows the read-time steps in the make process is in Chapter 4,
Debugging Makefiles.
Run-Time Initialization and Deinitialization
Before making any targets, omake executes the build scripts of the .BEFORE special target if it is
defined.
Immediately before exiting, omake executes the build scripts of the .AFTER special target if it is
defined. omake then removes any temporary files it has created and exits.

Updating the Time Stamp
If the target is not in a VOB, omake updates its internal copy of a target's time stamp after the
target has been made. The value the time stamp is set to is determined as follows:
➤ If the target has build scripts, the time stamp is set to the current system time after all build
scripts have been executed.
➤ If the target has no dependencies, the time stamp is set to the current system time.
➤ Otherwise, the time stamp is set to the most recent of the target's current time stamp and
those of all its dependencies.
3.4 Targets
This section provides general information about targets.
Case-Sensitivity of Target Names
By default, for Windows NT omake target names are case-insensitive. The .CASE_TARGET and
.NOCASE_TARGET directives turn case-sensitivity on and off.
A Target Has a Name and a Pathname
It is an important fact that a target has both a name (from the makefile) and a pathname (its location
on disk). Each name in the makefile specifies a unique target. The pathname is determined when
omake searches for the target on disk.
If name does not have a path component in it, omake may search multiple directories to locate
the target and the name and pathname may differ. If name has a path component, or if omake is
not instructed to search multiple directories, the name and pathname are the same. See Search
Directories on page 106.
Using the Path Separator in Names
On Windows NT, you can use either a slash ( / ) or backslash ( \ ) as the path separator inside of
omake makefiles. Windows NT accepts either character as the path separator when Make looks
in the file system directory.
We suggest you use the slash in your makefiles because the backslash can be used to modify the
meaning of the character that follows it. For example, \<ENTER> continues makefile lines and \#
is a literal #. If you use the backslash as the path separator, you must be aware of how omake
interprets this expression:
\the-following-character
The First Target in the Makefile Is the Default Target
If no targets are specified on the omake command line, the first normal target in the makefile is
made. In this context, a normal target is not a directive, special target, attribute, or inference rule.
It is a common practice to have the default target depend on other targets to be built. Consider
this makefile:
all : a.exe b.exe
a.exe : dependencies
build scripts
b.exe : dependencies
build scripts
Running omake without a command-line target causes all (a.exe and b.exe) to be made. The
target all is sometimes called a dummy target because it is used to drive the build of other (real)
targets.
Targets May Appear on Several Dependency Lines
A target may appear on the target side of more than one dependency line. The target depends on
all dependencies on all dependency lines that the target appears on, but the build script that
updates the target can be listed after only one of the dependency lines.

Double-Colon Dependency Lines
When a double-colon (::) separates the target and its dependencies, build scripts can be specified
after each double-colon dependency line. If in the MVFS, all build scripts are executed if any
dependency on any dependency line is newer than the target. Double-colon dependencies are
rarely used. Here is a simple example:
backup :: a.exe
copy $(.SOURCE) c:\backup
backup :: b.exe
copy $(.SOURCE) c:\backup
If you were to execute the omake backup command, a.exe is built and then b.exe . Each target is
copied to c:\backup after it is built.
Mixing Single-Colon and Double-Colon Dependency Lines
A target cannot appear on the target side of both double- and single-colon dependency lines. If
it does, omake displays a warning and ignores the offending dependency line.
Targets Without Dependencies
A target that has no dependencies (neither explicit nor inferred) is up to date if it exists as a file;
if it doesn't exist, it is out of date and must be updated.
Target Groups
Several targets may be made simultaneously from a single dependency. This is called a target
group. Updating any target of the group (by running the target's build scripts) updates all the
targets.
To indicate a target group, put a plus sign (+) between each target on the dependency line. For
example, here is a rule to build parse.h and parse.c simultaneously from parse.y using the
program called yacc:
parse.h + parse.c : parse.y
yacc –d parse.y # produce ytab.h & ytab.c
copy ytab.h parse.h # copy ytab.h to parse.h
copy ytab.c parse.c # copy ytab.c to parse.c
del ytab.* # remove the ytab files
Assume that parse.y was changed recently and that some other target depended on parse.h.
Making this other target causes parse.h to be made, which executes the build scripts. If parse.c
is evaluated later in the make process, omake recognizes that it is a member of the same target
group as parse.h, which has already been updated, and does not rerun the build scripts.
3.5 Macros
A macro definition associates a name and a value. The macro expansion of a macro name returns
the value. Macros are used at read time to organize names of files, compiler options, and so on.
At run time, macros provide information about the current target being built.
Macro Modifiers
When a macro is referenced, the expanded value can be modified through the use of macro
modifiers. To modify a macro, expand it with:
Everything between $( and ) is expanded, the value of name is expanded, and then each modifier
is applied in succession to the expanded value. The separator between name and modifier can be
a comma or colon, but subsequent modifiers must follow a comma. A literal comma is available
by using \,. Some modifiers use regular expressions. The treatment of the backslash, can make it
awkward to use a modifier when a backslash also terminates its argument. You can work around
this by using two backslashes before a comma separator(\\,) or
if possible, move the awkward modifier to the end. If this not possible, use two modifications.
For example,
OBJS = $(SRCS,<obj\,R,>.obj)
does not work, but this pair of macro definitions does:
OBJS = $(SRCS,<obj\)
OBJS := $(OBJS,R,>.obj)

Considering the macro value as a list of macro elements separated by white space, the first
modifier is applied to all elements, then the next modifier is applied, and the next, and so on. The
following list of modifiers is organized into functional groups. Examples of each modifier use the
following macro definition:
SRCS = \src\main.c \sys\sub.cpp io.cpp
Filename Components
B The base name part of the element. This is the file name without an extension.
$(SRCS,B) is main sub io
D The directory part of the element. For X:\file, X:file, \file, or dir\file, the directory part is
X:\, X:, \, or dir, respectively. If there are no directory separators the directory part is
the current directory ( . ).
$(SRCS,D) is \src \sys .
E The extension (or suffix) of the element. This is “” if there is no extension.

$(SRCS,E) is .c .cpp .cpp
F The file name part of the element. The part after the last directory separator.
$(SRCS,F) is main.c sub.cpp io.cpp
P The path part of the element. This is similar to the D modifier, but includes the last
directory separator. So for X:\file, X:file, \file, dir\file, or file the path part is X:\, X:, \,
dir\ or “”, respectively.
$(SRCS,P) is \src \sys
R The root part of the element. The same as its full name minus its extension.
$(SRCS,R) is \src\main \sys\sub io
Z The drive letter part of the element (or the current drive), terminated by a colon
$(SRCS,Z) is V: V: V: (if “V” is the current drive)
Absolute Pathname
A[sep] Convert the element to an absolute pathname. If the element is not already an
absolute path, this modifier prepends the current drive and working directory.
Then, the sep separator replaces the slash and backslash. Finally, .. and . directory
parts are removed. For example, if the current working directory is v:\stage,
then:
$(SRCS,A) is v:\src\main.c v:\sys\sub.cpp v:\stage\io.cpp
$(SRCS,A/) is v:/src/main.c v:/sys/sub.cpp v:/stage/io.cpp
If sep is missing, the backslash is used.
Append and Prepend Strings

>string Appends string to each element of the macro.
$(SRCS,B,>.obj) is main.obj sub.obj io.obj
<string Prepends string to each element of the macro.
Change Case
LC Lowercase. Change the letters of the element to lowercase.
$(SRCS,Z,LC) is d: c: c:
UC Uppercase. Change the letters of the element to uppercase.

$(SRCS,B,>.obj,UC) is MAIN.OBJ SUB.OBJ IO.OBJ
Expand Pathnames
X This modifier expands elements into their pathnames. As an example:
.PATH.obj = objs
OBJS = 1.obj 2.obj
link $(.TARGET), $(OBJS,X);
The .PATH.obj macro defines a search directory where .obj files are located. The X
modifier causes omake to search for each element in its search directories. Assuming
1.obj and 2.obj are in the objs directory:
$(OBJS) is 1.obj 2.obj
$(OBJS,X) is objs\1.obj objs\2.obj
Include File
@ If the macro's value names a file, the value of the macro modification is the contents of
the file with spaces, tabs, and newlines collapsed to single spaces.
This modifier is best used with the := expanded macro definition so the file or project is
read only once, when the macro definition occurs. For example:
FILE = link.rsp (response file)
OBJS := $(FILE,@) (read contents of link.rsp)

Include File with Regular Expression Matching
@d regex d subst d
Include file, with regular-expression matching. [A regular expression .
matches any single character; a regular expression \. is a literal period (or
dot). Several regular expression characters that omake uses are poor choices
when the regular expression is used to match file names. omake allows you
to redefine these characters with the .REGEX_CHAR and .REGEX_WILD
directives.]
The macro's value is the name of a file to be read. Each line in the file is
examined with the regular expression regex. If regex matches the line, the
matched part of the line is replaced with the substitution string subst. If regex
does not match, the line is skipped.
The d is a single-character delimiter that cannot appear in regex or subst and
cannot be a comma (usually a slash (/) or single quote ( ' ) is used).
Member and Nonmember

Mregex The member selector. This selects elements matching regular expression
regex.
$(SRCS,M\.c$$) is \src\main.c
After macro expansion, \.c$$ is \.c$, a regular expression that matches .c at
the end of the element. This matches .c but not .cpp.
M"spec" The member selector, where spec is enclosed in double quotes and is a
Windows NT file specification, not a regular expression:
$(SRCS,M"\src\main.c") is \src\main.c
$(SRCS,M"*.cpp") is io.cpp
One use of this modifier is to determine whether the current target belongs to
some special list of names. For example:
SPEC_OBJS = objs\spawn.obj objs\sync.obj
%.obj : %.c
if '$(SPEC_OBJS,M"$(.TARGET)")'if $(.TARGET) is in SPEC_OBJS
special commands to handle special objects
endif
Nregex The nonmember selector. It selects elements that do not match regex.
$(SRCS,N\.c) is io.cpp
N"spec" The nonmember selector, where spec is a Windows NT file specification, not a
regular expression:
$(SRCS,N"\src\main.c") is \sys\sub.cpp io.cpp
$(SRCS,F,N"*.c") is \sys\sub.cpp io.cpp
Select a Particular Element
number Selects the numberth element of this macro value. If the macro value doesn't
have this numberth element, the modified value is the empty string. This
modifier is particularly useful for inference rules that build multiple targets
from a single dependency.
$(SRCS,2) is \sub.cpp
String Substitution
Sd regex d subst d
The substitution modifier. The d is a single-character delimiter that is neither in regular
expression regex nor in substitution string subst and isn't a comma. If regex matches some
part of a macro element, the matched part of the element is replaced with subst. The
element is not changed if regex does not match it.
If regex is omitted, the last regular expression is reused. This is useful in combination
with the member modifier, M. For example, given that CFLAGS has the value
–AX –Ifoo –Ibar –DX=–IT:
$(CFLAGS,M^–I,S'^–I'') is foo bar

$(CFLAGS,M^–I,S''') is foo bar
The M modifier selects elements that start with –I, and the S operator substitutes the –I
with nothing. This is quite different from
$(CFLAGS,S'^–I'') is –AX foo bar –DX=–IT
Notice how the elements unmatched by regex ^–I are not changed.
from=to
This modifier replaces occurrences of from string with to string. If from does not appear
in a macro element, the element is not changed.
from and to can be simple strings or can use the percent sign (%) as a wildcard. If % is
used, the search for from is anchored to the end of the element.
$(SRCS,%.c=%.obj) is \src\main.obj \sys\sub.cpp io.cpp

$(SRCS,.c=.obj) is \src\main.\sys\obj sub.cpp io.cpp
The use of % anchors the search for %.c so it matches main.c but not sub.cpp. The simple
string substitution .c=.obj replaces any occurrence of .c with .obj so matches and
changes sub.cpp to sub.objpp.

The delimiter between strings is the equal sign, but it is treated literally when prefixed
with a backslash (\=). If fromString ends with a backslash, you must use two (\\=) to
mean literal backslash followed by a delimiter. (Under PM/CB or NMAKE emulation,
the delimiter is not quotable. This means that \= is taken literally, and there is no need
to double the backslash).
For historical reasons, when fromString and toString are simple strings, this modifier
operates on the entire macro value rather than on its elements, and can be used to replace
the white space between elements. However, the W modifier is better suited for this task.
Tokenize
Wstr
This modifier replaces the white space that separates macro elements with str, where
str is treated literally except for the following special sequences:
\n A newline character
\r A return character
\t A tab character
\\ A backslash
\ddd An octal character ddd (1 to 3 digits)
\xdd A hexadecimal character dd (1 or 2 digits)
This example can be used to prepare an inline response file with the OBJS definition
of the X modifier listed above:
$(OBJS,X,W+\n) is objs\1.obj+
objs\2.obj
project.exe;
Wildcard Expand File and Directory Names

* or *F The macro's value is a wildcard file specification used to match file names. The
value of the macro modification is the list of files that matches the specification.
This list includes any directory components. For example:
SPEC = obj\*.obj (set up file spec)
OBJFILES := $(SPEC,*F) (get list of obj\*.obj files)
*D Like the *F modifier but matches directory names. For example:
SPEC = * (set up directory spec)
SUBDIRS := $(SPEC,*D) (get list of subdirectories)
Regular Expressions
The M, N, and S modifiers use pattern-matching strings known as regular expressions. The
regular expression (or regex) matches characters in strings, both literally and with wildcards, and
the match is case-sensitive.
Configuring Regular Expressions
Two directives control the special characters that appear in regular expressions:
➤ .REGEX_CHAR sets the character used to indicate special character sequences. The default
character is a backslash (\).
➤ .REGEX_WILD sets the wildcard character that matches any single character. The default
character is a period ( . ).
The backslash and period are standard characters in UNIX-styled regular expressions, but are
awkward to use in Windows NT because the backslash is the directory separator and the period
is the file extension separator. The .REGEX_CHAR and .REGEX_WILD directives can be used
to change these characters to some other characters.
The default characters are most problematic for the M and N modifiers when you are trying to
match a file name. For this case, use the M"spec" and N"spec" modifiers because the spec is a file
specification rather than a regular expression.
Regular Expressions for the M Modifier
Assume the following macro definitions:
SRCS = main.c sub.cpp io.cpp

CFLAGS = –AX –Ifoo –Ibar /Ibaz –DX=–IT xI.c yi.c
To select files whose names include .c:
$(SRCS,M\.c) is main.c sub.cpp
To select files that end in .c, the search can be anchored to the end with the regex character $. To
get $ to the regular expression, you need to use $$ in the makefile:
$(SRCS,M\.c$$) is main.c
You can also select .c files with the M"spec" modifier:

$(SRCS,M"*.c") is main.c
Analogous to the $ anchor, the ^ regex character anchors the search to the front of the macro
element:
$(CFLAGS,M–I) is –Ifoo –Ibar –DX=–IT

$(CFLAGS,M^–I) is –Ifoo –Ibar
The [set] regex characters indicate a set of characters, where the set can be single characters ([aA@]
matches a, A or @), a range of characters ([a-z] matches a through z), or a mixture. For example:
$(CFLAGS,M^[–/]I) is –Ifoo –Ibar /Ibaz
Regular Expressions for the S Modifier
One powerful feature of regular expressions is that when used in substitutions, they can access
the matched parts of the string. The down side of the more powerful regular expressions is that
the expressions can be hard to read. For example, when DIR = NT_L, the expression
$(DIR,S/$.*$_.*/\1/) is NT.
The $, $ pair surround a part of the regular expression that is referenced later. Inside the pair
is .*, which matches any character (.) repeated zero or more times (*). Taken together they instruct
omake to match any character, zero or more times, and attach a tag to it. The rest of the regular
expression is _ , which matches _, and .*, which matches any character repeated zero or more
times.
The expression \1 is the stuff matched in the first pair of , so the substitution evaluates to
NT. Simple regular expressions are easy to read, but if you write more complicated expressions,
be sure to provide thorough comments.
With Configuring .REGEX_CHAR and .REGEX_WILD
If omake is configured for alternative regular expression characters, the following directives
appear in your make.ini:
.REGEX_CHAR : ~
.REGEX_WILD : ?
These examples are then easier to read:
$(SRCS,M.c) is main.c sub.cpp
$(SRCS,M.c$$) is main.c
$(DIR,S/~(?*~)_?*/~1/) is NT
Predefined and Built-In Macros
omake defines several macros before it reads the initialization file and makefiles. These macros
come in two general types:
➤ Predefined macros cannot be changed or undefined. Important subsets of the predefined

macros are run-time macros, which change with the target being built, and state macros,
which hold the state of the command-line options and read-time directives.
➤ Built-in macros have the same precedence as normal makefile macros, and so can be
redefined in your makefiles or on the command line.
Predefined Macros: Run-Time Macros
The run-time macros are changed dynamically by omake according to the current target being
built. A key feature of the run-time macros is that their values are pathnames. A target's pathname
is the location the target was found on disk. The run-time macros evaluate to pathnames, which
is important for use on build scripts where the executed command needs the location of the
targets.
All run-time macros have names of the form .name, and several have one- or two-letter aliases
that you are likely to see in makefiles of other make utilities.
Run-Time Macros
.NEWSOURCES (alias is ?)
This run-time macro evaluates to the pathnames of all dependencies newer than the
current target. When configuration lookup is enabled, it evaluates to the pathnames of
all dependencies for the current target, unless that behavior is modified with the
.INCREMENTAL_TARGET directive (see the description in the section Dot Directives

on page 91), in which case it evaluates to the pathnames of all dependencies different
from the previously recorded versions.
.SOURCE (alias is <)

This run-time macro evaluates to the pathname of the inferred dependency that satisfied
the inference rule, or the pathname of the first explicit dependency if no inferred
dependency was found. Here is a sample inference rule for updating an object file with
the Borland C compiler:
%.obj : %.c
cl –c $(.SOURCE)
Using an alias, the second line becomes
cl –c $<
.SOURCES (aliases are ^ or **)

This run-time macro evaluates to the pathnames of all dependencies of the current target.
Here is an example that updates an executable with all object files that are its
dependencies:

link /out :$(.TARGET), $(.SOURCES);
link /out : $@, $**;
NOTE: The ** macro is the exception to the rule that multicharacter macros need
parentheses or braces around them. The expression $** is the same as $(**).
.TARGET (alias is @)
This run-time macro evaluates to the pathname of the target currently being made. For
example:
%.obj : %.c
cl –o$(.TARGET) –c $(.SOURCE)
cl –o$@ –c $<
This macro can also be used at read time on a dependency line and its value is the name
of the target currently getting dependencies. For example:
main.obj io.obj : $(.TARGET,R).c io.h
Using an alias, the line becomes
main.obj io.obj : $(@,R).c io.h
The .TARGET macro value is set to the name of each target, in turn. The dependencies
are then macro-expanded and added to the target's list of dependencies. The R macro
modifier selects the root of the target name. The above dependency line is equivalent to
the two lines:
main.obj : main.c io.h

io.obj : io.c io.h
For compatibility with other Make utilities. omake treats $$@ on the dependency side of
a dependency line just like $(.TARGET).
.TARGETROOT (alias is *)
This run-time macro evaluates to the pathname of the target currently being made, minus
its extension. This macro can also be used at read time on the dependency side of a
dependency line. Here is the read-time example from the .TARGET macro rewritten
using this macro:
main.obj io.obj : $(.TARGETROOT).c io.h
Using an alias, the line becomes
main.obj io.obj : $*.c io.h
Predefined Macros: General Macros
The predefined macros in Table 3 give you information about the current make process. One of
the most important predefined macros is the status macro at the bottom of the table.
Table 3 General Macros
General Macro Value
$ When you need a literal dollar sign, you must use $$.
BUILTINS The name of the initialization (built-ins) file.

Table 3 General Macros
General Macro Value
CWD The current working directory (same as MAKEDIR).
FIRSTTARGET The first command-line target or, if one isn't given, the default
makefile target.
INPUTFILE The name of the current makefile. omake can read multiple
makefiles, either with multiple –f command-line options or
with %include directives. The INPUTFILE macro's value is the
current makefile being read.
MAKEARGS The command line with which omake was started, including
command-line options, macros, and targets.
MAKEDIR The current working directory (same as CWD).
MAKEMACROS The command-line macros with which omake was started. Any
macros that have spaces are enclosed in double quotes so they
can be used on a command line again.
MAKESTATUS The exit status of omake. This can be used in the .AFTER
special target to determine whether omake is exiting with an
error.
MAKETARGETS The list of command-line targets passed to omake.
MAKEVERSION The version number for omake. Its format is X.Y, where X and
Y are the major and minor release numbers.
OPUS Defined to the value 1. This can be used to test if you are
running omake or some other Make.
status The exit status of the last build script. It is used with conditional
directives to execute other build scripts. For historical reasons,
the name is lowercase.
Predefined Macros: State Macros
The state macros provide information about the state of command-line options and read-time
directives. Most command-line options have an equivalent directive, and the value of the state
macro is the same regardless of whether the option or directive was used.
In the list of macros in Table 4, the macro values that are the state of a directive or command-line
option mean the value is 1 if the directive or command-line option was used; otherwise the value
is 0.
NOTE: In terms of implementation, .ALWAYS, .IGNORE, and .SILENT are not directives; they
are target attributes. They look like directives when they appear in a makefile on the target side
of a dependency line when there are no dependencies. For example,
.IGNORE :
sets the .IGNORE attribute for every target created after the appearance of this line. The
.ALWAYS, .IGNORE, and .SILENT macros have the correct value, as if these attributes were
directives.
NOTE: Exactly one of $(.MS_NMAKE), $(.omake) or $(.POLY_MAKE) is 1.
Table 4 State Macros (Part 1 of 3)
State Macro Value
.ALWAYS The state of the .ALWAYS directive and –a command-line option.

[In terms of implementation, .ALWAYS, .IGNORE, and .SILENT
are not directives but target attributes. They look like directives
when they appear in a makefile on the target side of a dependency
line when there are no dependencies. For example,
.ALWAYS :
sets the .ALWAYS attribute for every target created after the
appearance of this line. The .ALWAYS, .IGNORE, and .SILENT
macros have the correct value, as if these attributes were directives.]
.CASE_MACRO The state of the .CASE_MACRO directive.
.CASE_TARGET The state of the .CASE_TARGET directive.
.DEBUG The debug options as set by the .DEBUG directive and –#

command-line option.
.DEBUG_PRINT The state of the .DEBUG_PRINT directive and –p command-line

option.
.DEBUG_RUN The state of the .DEBUG_RUN directive and –d command-line

option.
.ENVMACROS The state of the .ENVMACROS directive.

State Macro Value
.ENV_OVERRIDE The state of the .ENV_OVERRIDE directive and –e command-line

option.
.IGNORE The state of the .IGNORE directive and –i command-line option.
.IGNORE_MFLAGS The state of the –z command-line option.
.KEEPDIR The state of the .KEEPDIR directive and –D command-line option.
.KEEPWORKING The state of the .KEEPWORKING directive and –k command-line

option.
.MAKE_MAKEFILE The state of the .MAKE_MAKEFILE directive and –M

.MS_NMAKE The state of the .MS_NMAKE directive (exactly one of

$(.MS_NMAKE), $(.omake) or $(.POLY_MAKE) is 1) and –EN
.OPUS_52X The list of Opus Make v5.2x compatibility features, as set by the –E2
.omake The state of the .omake directive and –EO command-line option.
.POLY_MAKE The state of the .POLY_MAKE directive and –EP command-line

option.
.QUERY The state of the –q command-line option.
.REGEX_BACK The regular expression literal backslash (\). This is provided for
writing regular expressions that are independent of the value of
.REGEX_CHAR. If .REGEX_CHAR is \,its value is \\; otherwise,
its value is \.
.REGEX_CHAR The regular expression escape character set by the .REGEX_CHAR

directive.
.REGEX_DOT The regular expression literal dot (or period). If .REGEX_WILD is .

its value is ${REGEX_CHAR}.; otherwise its value is .
.REGEX_WILD The regular-expression wildcard character set by the

.REGEX_WILD directive.
State Macro Value
.REJECT_RULES The state of the .REJECT_RULES directive and –r command-line

option.
.REREAD The state of the .REREAD directive.
.RULE_CHAR The regular expression character as set by the .RULE_CHAR

directive.
.SHELL The command to execute the shell program as set by the .SHELL
directive.
.SILENT The state of the .SILENT directive or –s command-line option.
.SUFFIXES The list of suffixes as set by the .SUFFIXES directive.
.UNIXPATHS The state of the .UNIXPATHS directive.
An Example Use of the State Macros
Here is an example of how these macros can be used. Suppose you want to debug the contents
of your makefile with the –#1 command-line option. However, this also produces output that
comes from your make.ini file, which probably doesn't need debugging. Modify your make.ini
file with the following:
_OLD_DEBUG := $(.DEBUG) (get current state)
.NODEBUG : 1 (turn off “-#1” if set)
[Original make.ini goes here]
.DEBUG : $(_OLD_DEBUG) (restore state)
Built-in Macros
omake defines the built-in macros with a default value, but they have the same precedence as
normal makefile macros, so you can be redefine them.

Table 5 Built-in Macros (Part 1 of 2)
Built-in Macro Definition
CC Used in the %.obj : %.c rule, the name of the C compiler. The default
value is cl, the name of the Microsoft C Compiler.
FC Used in the %.obj : %.for rule, the name of the FORTRAN compiler.
The default value is f77l, the name of the Lahey FORTRAN
Compiler.
LIBEXE Used in the %.lib : %.obj rule, the name of the object librarian. The
default value is lib, the name of the Microsoft Librarian.
LINK Used in the %.exe : %.obj rule, the name of the object linker. The
default value is link, the name of the Microsoft Linker.
MAKE The value is the full pathname of the omake executable.

The MAKE macro is special in that its appearance on a build script
overrides the –n (no execute) command-line option for that line.
This can be used in a recursive make, to have omake do the
recursion.
MAKE_TMP The name of the directory omake uses for temporary files (response
files under Windows NT).
Initially MAKE_TMP is undefined and omake uses the current
directory for temporary files. If you define MAKE_TMP its value
must be an absolute directory (such as D:\ or D:\tmp, preferably
on a RAM disk. As an example, the environment variable TMP
names the temporary directory some compiler vendors use. You
can put the following in make.ini:
MAKE_TMP = $(TMP)
Table 5 Built-in Macros (Part 2 of 2)
Built-in Macro Definition
MFLAGS After all makefiles have been read, omake defines MFLAGS with
all the command-line options. MFLAGS is useful for invoking
omake recursively, as shown in the definition of the MAKE macro.
NOTE: MFLAGS is a built-in macro only if you are using native
omake mode (that is, if you are not using an emulation mode). In
the emulation modes, it is not predefined by omake; however, you
can redefine it.
A second usage of MFLAGS is to pass initial options to omake. If
MFLAGS is defined in the environment or in make.ini its value
specifies additional command-line options. For example, the
following macro definition placed in make.ini turns on the
keep-directory mode (the –D command-line option):
MFLAGS = –D
Because there are directives for all command-line options, this
second usage of MFLAGS is discouraged, and you should use the
directives.
OS The operating system. Its value is NT.
RC Used in the %.res : %.rc rule, the name of the resource compiler. The
default value is rc, the name of the Microsoft Resource Compiler.
SHELLCOMMANDS Alphabetical list of commands known to need execution by the

shell program. If you define this macro, omake uses it to detect
when to use the shell program. This macro is not defined initially;
omake uses an internal list of commands. See Auto-Detection Mode
on page 19.
SHELLSUFFIX Suffix used by omake for the batch files it generates. The default
value is .bat.
Compatibility with Other Make Utilities
Through emulation, omake supports all PM/CB and NMAKE macros.

3.6 Build-Script Line Prefixes
Build-script line prefixes control some aspects of a build script's execution. The prefixes occur on
a build script before the program name. The first nonprefix character ends the prefixes. The bar
(|) prefix can also be used to mark the end explicitly. White space is allowed between prefixes,
and between the prefixes and the program name. Build-script lines are macro-expanded before
prefixes are detected, so you can use macros to define prefixes.
Do Not Echo the Build-Script Line (Silent Operation)
@
Usually omake echoes (displays) the build-script line to the screen before executing it.
The @ prefix prevents this display except when running omake –n.
Here is an example of its use:
ERRS = make.out
$(OBJS) : $(.TARGETROOT).c
%do CCquiet
CCquiet :
@ %echo –n Compiling $(.SOURCE) ...
@ $(CC) $(CFLAGS) –c $(.SOURCE) >>$(ERRS) 2>&1
@ %echo done.
Redirection by >>$(ERRS) 2>&1 means error messages from the compiler go into the
$(ERRS) file. The net effect is that a series of messages of the form:
Compiling file ... done.
are displayed and all output from the compiler is redirected into make.out.
See also the .SILENT target attribute to turn on silent mode for this target and the –s
command-line option to enable silent mode for all targets.
@@
The @@ prefix prevents display of the build script before execution, even when running
omake –n.
Ignore the Build-Script Line Exit Status
–[num]
Normally a build-script line that returns a nonzero exit status causes omake to terminate
with this message:
omake: Shell line exit status exit_status. Stop
The dash (–) prefix causes omake to ignore the exit status returned from the build-script
line and to continue. The message now becomes
omake: Shell line exit status exit_status (ignored)
If num is given, omake ignores the exit status if it is no greater than num. For example,
the prefix –4 tells omake to ignore an exit status of 1, 2, 3, or 4. There must be at least one
space after num.
Importantly, the status macro is set to the exit status of the build-script line so it can be
tested later. See the -- prefix for an example.
See also the .IGNORE target attribute to turn on ignore mode for this target and the –i
command-line option to turn on ignore mode for all targets.
--[num]
Ignores the exit status and does not display the warning message.
If num is given, omake ignores the exit status if it is less than or equal to num.
For example, the CCquiet target of the previous example can be changed to this:
CCquiet :
@ %echo –n Compiling $(.SOURCE) ...
@– – $(CC) $(CFLAGS) –c $(.SOURCE) >>$(ERRS) 2>&1
%if $(status) != 0
% abort FAILED! See $(ERRS) for errors.
%endif
@ %echo done.
Now a failed compile generates this message:
Compiling file ... FAILED! See make.out for errors.

~[num]
Ignores the exit status and displays the warning, but the status macro retains its current
value and is not set to the exit status of this build script.
~~[num]
Ignores the exit status, does not display the warning, and retains status.
Override the –n Command-Line Option
&
Use & to override the –n command-line option for this build script. This prefix is useful
if you are using omake to call itself recursively.
It is usually better to use the .MAKE target attribute instead because it overrides the –q
and –t command-line options as well as –n.
Select the Shell Program
omake executes all build-script lines with the shell program. This default can be changed with
the .SHELL directive, which selects use of the shell program, and the .NOSHELL directive,
which selects direct execution.
The following prefixes override the general execution mode for the current build script:
: Executes the build script directly.
+ Executes build scripts by the shell program.
Iterate the Build Script
!
Iterates the build-script line for each element of $? (i.e. $(.NEWSOURCES)) or $** (that
is, $(.SOURCES)), on the basis of which appears first on the build-script line. The
build-script line is executed once for each element of the original $? ($**) with the value
of $? ($**) taking on successive elements of the original $? ($**).
If neither $? nor $** appears on the build-script line, the build-script line is iterated once
for each element of $**.
omake performs iteration explicitly over any number of build scripts with its %foreach
directive, and the use of the ! prefix is discouraged.
Miscellaneous Prefixes
|
The bar (|) prefix can be used to mark the end of the prefixes. This is useful when the
command to be executed starts with one of the prefixes.
>
The right-angle bracket (>) prefix inserts an extra carriage return/linefeed after the build
script is executed. This prefix is supplied for PM/CB compatibility.
Build Script Compatibility with Other Make Utilities
For PM/CB compatibility, omake supports named prefixes such as (Silent) and (Ignore).
3.7 Build-Script Problems: The cd and set commands

Each build-script line is executed in a separate shell and each build-script line starts in omake's
current directory. This causes special problems for some commands. For example, suppose you
want to change directory to subdir and run omake recursively. If you use this line
recursion :
chdir subdir
omake $(MFLAGS)
omake $(MFLAGS) is executed from the current directory.
Similarly, setting environment variables does not work as expected because the environment
variable is set in one shell, the shell exits back to omake, and the next command is executed in a
new shell that doesn't have the variable set. For example, this line
setenv :
set USER=dgk (for UNIX: setenv USER dgk)
echo %USER% (for UNIX: echo $$USER)

does not echo dgk.
NOTE: In NMAKE emulation mode, omake handles internally any commands that start by
setting an environment variable, rather than calling the shell.
Both of these problems can be avoided by using a multiple-command build-script line.
Using Multiple-Command Build-Script Lines
The syntax for executing more than one command in the same shell is
( command [ & command ] ... )
The build-script line is enclosed in parentheses, and the commands are separated by a semicolon
(;). (Use \; to specify a literal semicolon). In terms of implementation, omake writes each
command to a batch file and uses the shell program named by the .SHELL directive to execute
the batch file.
The recursion examples now become
recursion :
( chdir subdir & omake $(MFLAGS) )
Simultaneous support of all operating systems is possible with a conditional macro definition
such as this:
%if $(OS) == NT
; = &$; is ”&”
%else
; = ;$; is “;”
%endif
The multiple-command build-script line is now written as

( chdir subdir $; omake $(MFLAGS) ) (for all OSes)
After the macros are expanded, the build-script line becomes

( chdir subdir ; omake (if OS is not Windows NT)
command_line_flags )
( chdir subdir & omake (for Windows NT)
command_line_flags )
Using Directives: %chdir and %setenv
Rather than use a multiple-command build script to work around the change-directory problem,
you can use a directive. The %chdir directory directive causes omake to change to directory, where
it stays until the next %chdir directive or until omake exits. The recursion example becomes
recursion :
%chdir subdir (change to the subdirectory)
omake $(MFLAGS) (do the recursive Make)
%chdir $(MAKEDIR) (change the directory back)
Likewise, to work around the environment-variable problem you can use a directive. The
%setenv name val directive sets environment variable NAME to the value val. Like %chdir,
%setenv has effect until the next %setenv or until omake exits. The setenv example becomes
setenv :
%setenv USER dgk
echo %USER% (for unix: echo $$USER)
For more information and recommendations on using the %setenv directive, see the %setenv
entry in Table 14 on page 90
3.8 Makefile Directives

Makefile directives provide additional control over omake. There are two directive types:
➤ Percent directives, which use a percent sign (%) as the first non-white-space character on
the makefile line.
➤ Dot directives, which look like dependency lines but use special target names.
Percent Directives
Percent directives (usually called directives) have names of the form %name, where the directive
character is the first non-white-space character on the makefile line. White space is allowed
between % and name.

These directives work as read-time directives or run-time directives with indentation
determining whether they are interpreted at read time or at run time. If % is in the leftmost
column of the line, the directive is interpreted at read time. If white space precedes the %, the
directive is interpreted at run time.
Conditional Directives
The directives in Table 6 control the flow of the make process at read time and at run time.
Conditional and iteration directives can be nested up to 31 levels deep.
Table 6 Conditional Directives
Conditional Directives Action Applicable Time
% if condition Start a conditional block read/run
% ifdef macro Start a conditional block read/run
% ifndef macro Start a conditional block read/run
% elif / elseif condition Continue the conditional block read/run
% else Start the default conditional block read/run
% endif End the conditional block read/run
Iteration Directives
The directives in Table 7 provide iteration control. Iteration and conditional directives can be
nested up to 31 levels deep.
Table 7 Iteration Directives
Iteration Directives Action Applicable Time
% foreach name [ in ] list Loop for each name in list read/run
% while condition Loop while condition is true run
% end End the loop read/run
% break Interrupt and quit innermost loop run
% continue Interrupt and restart innermost loop run
Other Percent Directives
Other directives behave like commands built into omake. At run time, these directives can be
preceded with build script prefixes, which must appear before the %. See Table 8.
Table 8 Other Percent Directives
Other Directives Action Applicable Time
% abort [ status ] [ message ] Display message, exit with status read/run
% chdir directory Change current directory run
% do [ target ] [ macros ] Execute build scripts of target run
% echo [ –n ] message Display message read/run
% error [ status ] [ message ] Display message, return exit status read/run
% exec command Execute command line read
% include [ <" ] file [ >" ] Include contents of makefile read
% restart [ flags ] Start omake again read/run
% set name [+]= value Set macro name with value run
% setenv name [ = ] value Set environment variable NAME read/run
% undef name Undefine macro name read/run
Conditional Directives
Conditional directives control the makefile lines that omake reads at read time and control the
build scripts that are executed at run time. These directives take this form:

%if condition
[ makefile line ]
.
.
.
[ %elif condition ]
[ makefile line ]
.
.
.
[ %else ]
[ makefile line ]
.
.
.
%endif
Each %if must be matched with an %endif. There can be several %elif or %elseif clauses and at
most one %else. Lines between %if or %elif and the next %elif, %else, or %endif are a block. If
condition is true, the makefile lines are read or the build-script lines are executed. If none of the
conditions is true and there is a %else clause, the block between %else and %endif is read or
executed.
If the % of these directives is in the leftmost column of the makefile, the directive is evaluated at
read time; otherwise, the directive is evaluated at run time. There can be white space between %
and the name of the directive. No shell-line prefixes are allowed before the %.
There are two specialized version of %if:
%ifdef name
%ifndef name
%ifdef is true if macro name is defined. %ifndef is true if macro name is not defined.
Conditional Directives and Continued Lines
Directives can be used to conditionally select the continued makefile text. For example:
OBJS = main.obj parse.obj \ (the continuing line)
%ifdef Debugging
version.obj \ (continues here if Debugging defined)
mymalloc.obj
%endif
a blank line (continues here if Debugging undefined)
If the Debugging macro is not defined, the OBJS macro value is main.obj parse.obj. If the macro
is defined, the value is main.obj parse.obj version.obj mymalloc.obj.
The way this example is written makes the blank line after the %endif necessary because the
continuing line is continued on the first line after the %endif if Debugging is undefined. Using
an %else clause instead is done like this:
OBJS = main.obj parse.obj \ (the continuing line)
%ifdef Debugging
version.obj \ (continues here if Debugging defined)
mymalloc.obj
%else
(continues here if Debugging undefined)
%endif
Conditional Expressions
The %if, %elif, and %while directives use the same conditional expressions. These expressions
compare strings and numbers and combine the comparisons with logical operations. As well, the
status of macros, command-line targets, and files can be determined. All macro references in the
conditional expression are expanded before the expression is evaluated.
The conditional expressions here are organized by type. Some examples follow each type. For the
examples, the following macros are assumed to be defined:
DEBUG = 0
MODEL = S
NONE =
OBJS = 1.obj 2.obj
Simple Expressions
You can use any of the following simple expressions:
value
'value'
"value"

If value is zero, the condition is false; all other values indicate true. Single or double quotes must
be placed around value if it contains spaces or is null.
Some examples of simple expressions:
%if ASTRING value is true

%if 12 value is true
%if 0 value is false
%if $(MODEL) value is true
%if $(DEBUG) value is false
%if $(NONE) ERROR! $(NONE) is null
%if "$(NONE)" value is false
Comparison Operators
The operators in Table 9 make a numerical or alphabetical comparison of two values, returning
1 if the comparison is true and 0 if false.
Table 9 Comparison Operators
Comparison Operators Value
value1 = = value2 True if value1 is equal to value2.
value1 != value2 True if value1 is not equal to value2.
value1 < value2 True if value1 is less than value2.
value1 <= value2 True if value1 is less than or equal to value2.
value1 > value2 True if value1 is greater than value2.
value1 >= value2 True if value1 is greater than or equal to value2.
If both values start with a digit, the comparison is done numerically; otherwise, it is done
alphabetically. If either value contains spaces or is null, it must be enclosed in quotes. The
case-sensitivity of the string comparison is the same as for target names and can be set with the
.CASE_TARGET and .NOCASE_TARGET directives.
Some examples of comparison operators:
%if $(MODEL) == S true
%if ABC > DEF false
%if $(DEBUG) != 0 false
%if $(XYZ) == 1 error! $(XYZ) is null
%if $(XYZ)x == 1x false
%if $(OBJS)x != x error! $(OBJS) has spaces
%if "$(OBJS)" != "" true
Functional Operators (Also Called Built-In Functions)
All function operators take one or more arguments and return a value that is false (0), true (1) or
some other number. See Table 10.
Table 10 Functional Operators
Functional Operators Value
%defined(name) True if macro name is defined.
%dir(name) True if name is a directory.
%exists(name) True if name is a file or directory.
%file(name) True if name is a file.
%length(name) The number of characters in $(name).
%make(name) True if name is a command-line target.
%member(name, list) True if name appears exactly as an element of list.
%null(name) True if macro name is undefined or if its expansion is null.
%time(name) The on-disk time stamp of file name.
%writable(name) True if file name is writable
File-Test Operators
The file-test operators return the state of files and directories. Most of these operators have been
made obsolete by the equivalent functional operators. See Table 11.

Table 11 File-Test Operators
File-Test Operators Value
–d name Same as %dir(name).
–e name Same as %exists(name).
–f name Same as %file(name).
–r name Same as %file(name).
–w name Same as %writable(name).
–z name True if name is a zero-length file.
One example of the file-test operators:

%if –e builtins.mak (true if builtins.mak exists)
Command-Execution Operator
The command-execution operator executes a command using the shell program. See Table 12.
Table 12 Command-Execution Operator
Execution Operator Value
[ command ] The exit status of the command. By convention,

where the brackets are literal. commands return 0 if they succeed.
Here is an example that runs a hypothetical mkproto program that freshens the prototype files
for its source files:
%if [ mkproto *.c ]

% abort MKPROTO could not freshen the prototypes.
%endif
You can ignore the exit status of the executed command by using an %if ... %endif pair that
doesn't encapsulate anything. The following idiom is seen in VC++ makefiles:
!if [ if exist MSVC.BND del MSVC.BND ]
!endif
Logical Operators
The logical operators are used to combine other expressions. Each expression is evaluated from
left to right, but parentheses can be used to order the evaluation. See Table 13.
Table 13 Logical Operators
Logical Operators Value
exp1 && exp2 True if both exp1 and exp2 are true.
exp1 | | exp2 True if either exp1 or exp2 are true.
! exp True if exp is false. False if exp is true.
( exp ) The same state as exp.
Unlike the C programming language, the logical expressions do not short-circuit. That is, exp1
and exp2 are always evaluated. For example:
%if %defined(NONE) && %null(NONE) (true)
%if %defined(XYZ) && %null(XYZ) (false)
%if ! ( ! –f file || –z file ) (false if file is absent or is zero length)
Iteration Directives
The %while and %foreach directives provide iteration capability. At run time, they allow build
scripts to be executed multiple times. %foreach can also be used at read time to replicate makefile
lines.
The %foreach Directive
The form of the %foreach directive:

%foreach name [ in ] list_of_names
[ makefile lines ]
.
.
.
%end
The list_of_names is a set of names separated by white space. The value of macro name is set to the
first name in the list_of_names, and the makefile lines after this %foreach and before %end are
read (at read time) or executed (at run time).
If the % of these directives is in the leftmost column of the makefile, the directive is evaluated at
read time; otherwise, the directive is evaluated at run time. No shell-line prefixes are allowed
before %foreach or %end.
When %end is reached, name is set to the next name in list_of_names, and the loop is restarted.
When there are no more names, the loop is done and name returns to its previous value. For
PM/CB compatibility, the optional keyword in is supported and %endfor can be used in place
of %end.
If %foreach is used at read time, the makefile lines between %foreach and %end are parsed
multiple times with name taking on successive values from the list_of_names. As the lines are
parsed, all macro references to name are expanded.
The following example illustrates the use of %foreach at read time:
%foreach var in main sub io

macro_$(var) = $(value_$(var)) $(other_macro)
$(var).obj : $(var).c
cl –c $(.SOURCE)
%endfor
All $(var) references are replaced with the current value of var. omake treats the previous
%foreach loop as if it has read the following makefile lines:
macro_main = $(value_main) $(other_macro)

main.obj : main.c
cl –c $(.SOURCE)
macro_sub = $(value_sub) $(other_macro)

sub.obj : sub.c
cl –c $(.SOURCE)
macro_io = $(value_io) $(other_macro)
io.obj : io.c
cl –c $(.SOURCE)
Here is an example run-time use of %foreach to build a linker response file using the inline
response file syntax (see Inline Response Files on page 115):
link @<< (link @response_file)
%foreach x in $(.SOURCES)
$x (adds each .obj +)
%end
(adds a blank line)
$(.TARGET); (add the executable name)
<< (end the response file)
The %while Directive
The form of the %while directive:
%while condition
[ build script and directives ]
.
.
.
%end
While condition is true, the build script and directives are executed. When %end is reached the
loop is restarted at the top, condition is reexpanded (because it may contain macros) and retested.
This is repeated until condition is false. No shell-line prefixes are allowed before %while or %end.
Effects of %foreach and %while when Using omake in a VOB
When you use omake in a VOB, there is a side effect with the %foreach and %while directives:
the build script in the CR does not expand loop macros. For a makefile like this one,
all:
%foreach platform in i386,alpha
$(MAKE) -f makefile.$(platform)
%endfor
you may expect the configuration record to include the commands that were executed. For
example:

omake -f makefile.i386
omake -f makefile.alpha
Instead, you get a copy of the build script:
%foreach platform in i386,alpha

omake.exe -f makefile.
%endfor
Note also that the platform macro has a value only while inside the loop, so the macro in the
configuration record is expanded to nothing.
Interrupting the Iteration
%break and %continue interrupt the iteration. %break stops the iteration immediately.
%continue restarts the iteration at the top. For %while, the condition is reexpanded and retested.
For %foreach the name is advanced to the next in the list_of_names.
Another Method of Iteration
The ! shell-line prefix iterates the build script for elements of either the .SOURCES (or **) or
.NEWSOURCES (or ?) macro, depending on which appears first on the build script. During
iteration, ${.SOURCES} (or $**) or ${.NEWSOURCES} (or $?) evaluates to each element, in turn,
of the macro value. We discourage the use of ! for iteration, preferring the explicit %foreach
directive.
A Sample Makefile
Here is a sample makefile that uses conditional and iteration directives:

%if defined(CV) (Read-time conditional)
CFLAGS = –Od –Zi
%else
CFLAGS = –Ox
%endif
io.obj : io.c
%while 1 (Loop forever)
––$(CC) $(CFLAGS) –c io.c >io.err (Compile and gather errors)
2>&1
%if $(status) == 0 (Run-time conditional)
%break (Quit if no compiler errors)
%endif
notepad –mnext_error io.c (Else edit the file)
%if %time(io.c) < %time(io.err) (Was io.c written?)
%error Compile of $(.SOURCE) failed (No, exit with an error)
%endif
%end
erase io.err (Remove the error file)
This example shows a compile-edit loop. The file io.c is compiled with all compiler messages
redirected into file io.err. The 2>&1 redirection is compatible with Windows NT and UNIX. It
means that standard error is redirected to standard output. The previous > io.err means standard
output is redirected into io.err. Together, they mean that standard error and standard output both
are redirected to io.err.
If the compile is successful, the %break directive breaks out of the %while 1 loop and erase the
io.err file.
Otherwise, notepad (the Notepad editor) is started and told to jump to the first error. After
Notepad finishes, check the time of the io.c file to see whether it is more recent than io.err,
indicating it has been changed. If it hasn't, omake breaks out of the compile-edit loop by calling
the %error directive. Otherwise, the loop restarts and the file is recompiled.
Other Percent Directives
Table 14 lists other percent directives, the times at which they are applicable, and their
descriptions.
Table 14 Other Percent Directives (Part 1 of 6)
Appl.
Directive Description
Time
%abort [ status ] [ message ] read, At read time and run time, terminates omake with the
run user-supplied exit status (1 if status is not supplied). If message is
supplied, omake prints it on the error output (standard error)
before terminating; otherwise, the termination is silent.
When you use %abort while running omake in a VOB, no
configuration records are written. Any output files created by the
build script before it executes %abort will not be derived objects.

Appl.
Time
%chdir directory run Changes the current directory to directory. This is the directory in
which each subsequent build script starts. For example:
copyit:
%chdir subdir (change into subdir)
omake $(MFLAGS) (do recursive Make)
%chdir .. (change back to parent)
The MAKEDIR macro is not affected by %chdir and its value is

always the directory omake started in.
%do [ target ] [ macro[+]=value ... ] run Executes the build scripts of the named target. The exit status of the
%do is the exit status of the last build script of target. If nothing
appears after the %do, this directive does nothing. If target does
not exist, omake issues a warning unless %do is preceded by the
@ shell-line prefix. During the %do, the attributes of target are the
combination of the attributes of the current target and of target,
with the current target having precedence.
macro=value is a macro definition that is in effect during this %do.
macro+=value appends value to the current value of macro. For
either form of redefinition, the old value is restored after the %do
is finished.
Spaces in target or between the start of macro and the end of value
must be enclosed in double quotes. Up to 10 macro definitions are
allowed per %do, separating the definitions with spaces. For
example, the directive
sub.obj : sub.c
%do COMPILE.c CFLAGS="–Od –Zi" OUT= –Fo$(.TARGET)
COMPILE.c :
%echo Compiling ==== $(.SOURCE) =====
cl $(CFLAGS) $(OUT) –c $(.SOURCE)
%echo Done ==== $(.SOURCE) =====
produces the following build scripts when sub.obj is updated:
Compiling ==== sub.c =====
cl –Od –Zi –Fosub.obj –c sub.c
Done ==== sub.c =====
Appl.
Time
When you use %do while running omake in a VOB, the

commands from the other build script are not included in the
configuration record. For the following makefile, any changes to
the rules to build MakeCFile do not cause derived objects built
previously to be out of date even though the executed commands
have changed:
.c.obj: $<
%do MakeCFile
MakeCFile:
@echo Building an object from a .C source file
$(CC) /c $<
%echo [ –n ] message [ >[>] file ] read, Prints message either to standard output, or to file if redirected. The
run message is followed by a line feed unless –n is specified.
%error [ status ] [ message ] read, At read time, omake terminates with the user-supplied exit status
run (1 if status is not supplied). At run time, this returns the exit status
which is treated like any other build script exit status. If message is
supplied, omake prints this message to the error output (standard
error) before terminating or returning the exit status.
%exec command_line read Executes the command_line and sets the status macro to the exit
status.

Appl.
Time
%include [ < or " ]name[ > or " ] read Reads the contents of name into this makefile. omake looks in
different places depending on whether the file is specified as name,
"name", or <name>.
If name is an absolute pathname, it is only looked for there.
Otherwise, "name" is looked for in the directory of the including
file, then in the directory of make.ini. For Windows NT, omake
looks in the directory of omake.exe, then in directories of the INIT
environment variable, and finally in directories of the INCLUDE
environment variable.
<name> is looked for in the same manner as is "name", except the
directory of the including file is not used.
For example:
%include c:\home\make.ini (use the absolute path)
%include <c:\home\make.ini> (use the absolute path)
%include make.sub (use dir. of including file)
%include "make.tpl" (use the search scheme)
%restart [ command_line ] read, Restarts the make process. If command_line is not supplied , omake
run is restarted with the original command line. This calls omake
recursively; it can be done at read time, and you don't have to keep
track of the command line.
When you use %restart while running omake in a VOB, %restart
writes a configuration record before restarting. Note that if a target
with %restart in the build script is ever winked in, the %restart
doesn’t occur. Therefore, if your build script has a rule to rebuild
the makefile and then has %restart to re-read the updated makefile
and continue building, this won’t have the desired effect if the
makefile is winked in. omake winks in the makefile, but no restart
occurs, and the old makefile is used for the rest of the build. In this
case, use –M or .MAKE_MAKEFILES to build the makefile before
reading it. In all other cases, mark any target with %restart in it as
.NOWINK_IN.
Appl.
Time
%set name [+]= value read, Sets macro name to value at run time, although it can be used at read
run time. Use += to append to the current definition. As an example,
the build scripts of the following set_debug_flags target sets the
LDFLAGS value to /CO and appends –Od –Zi to the CFLAGS
value:
set_debug_flags :
%set LDFLAGS = /CO
%set CFLAGS += –Od –Zi
Any white space between = and value is ignored. Any white space
between += and value is condensed to a single space.
%setenv name [=] value read, Sets environment variable NAME (converted to uppercase on
run Windows NT) to value. The variable is available to omake and to
the build scripts executed by omake. If NAME is also a macro, the
macro value is also updated to value. After omake terminates, the
variable reverts to its previous value.
The main use of this directive is in the .BEFORE special target to
set up the environment for the makefile. For example:
.BEFORE .MAKE :
%setenv INIT=$(MAKEDIR);$(INIT)
We recommend that you not use this command with configuration
records at run time. It can cause evaluation of environment
variables to differ from run to run and prevents shopping from
finding an appropriate derived object to wink in. For example, in
one build, target A is built and changes the value of an EV; target
B is then built and references the value of that EV. The next time
the build is run, B may be built first and because A hasn’t modified
the environment variable yet, it runs differently than it did the first
time.

Appl.
Time
%undef name read, Undefines the macro named name.

run When you use %undef while running omake in a VOB, the
configuration record is stored as if the %undef were not executed.
The %undef runs when the build script is executed, but if the
macro is evaluated after the %undef, the macro is expanded in the
configuration record with the value of the macro when the target
build began, even though the build script ran with the macro
expanding to nothing after the %undef.
For PM/CB compatibility, omake supports the %exit directive and %status functional operators.
For Microsoft NMAKE compatibility, omake supports the !CMDSWITCHES and !message
directives. See Appendix D, Compatibility and Emulation, for details.
Dot Directives
Dot directives (usually called directives, also) appear on a dependency line where the target
name is a period followed by uppercase letters. Dot directives only work at read time.
At read time, dot directives modify the operation of omake from the point the directive is
encountered in the makefile. Most command-line options have an equivalent dot directive.
The current state of each dot directive is kept by omake in a state macro with the same name as
the directive. For example, the .DEBUG directive sets the debugging mode and the $(.DEBUG)
is the value of the debugging mode. (The state macros are described in Predefined Macros: State
Macros on page 63 and listed in condensed form in Table 22 on page 139.)
Directives that are effective on only some operating systems are noted.
Some directives accept lists of patterns as their dependents. Pattern lists may contain the pattern
character %. When evaluating whether a name matches a pattern, the tail of the prefix of the
name (subtracting directory names as appropriate) must match the part of the pattern before the
%; the suffix of the name must match the part of the pattern after the %. For example,
/dir/subdir/x.o matches the patterns %.o, x.o, subdir/%.o, and subdir/x.o, but does not match
/dir/subdir/otherdir/x.o.
The following omake directives accept pattern lists:
.DEPENDENCY_IGNORED_FOR_REUSE:
.DO_FOR_SIBLING:
.INCREMENTAL_REPOSITORY_SIBLING:
.INCREMENTAL_TARGET:
.SIBLING_IGNORED_FOR_REUSE:
.WINK_IN:
Table 15 lists omake directives, their equivalent command-line options (if applicable), and the
actions that the directives perform.
Table 15 omake Directives (Part 1 of 10)
Directive Flag Action
.CASE_MACRO : By default, omake treats macro names in a

case-insensitive fashion. For this reason, the makefile
lines
Case = mixed
CASE = upper
define only one macro. That is, both $(Case) and
$(CASE) evaluate to upper. The .CASE_MACRO
directive makes macro names case-sensitive. Now the
lines above define two macros, Case and CASE.
The .NOCASE_MACRO directive makes macro
names case-insensitive.
.CASE_TARGET : By default, omake treats target names as

case-insensitive, so main.obj and MAIN.OBJ are the
same target. The .CASE_TARGET directive causes the
case of target names to be considered.
The .NOCASE_TARGET directive makes target
names case-insensitive.

.DEBUG : value –# Selects debugging actions, as listed below:

0 Turn off all warnings
1 Display makefile lines as they are read
2 Warn about undefined macros when they get
expanded
4 Warn about unrecognized lines in the makefile
8 Leave behind automatic response and batch files
The values can be summed to combine message types.
Also, each .DEBUG directive adds its value to the
current value, which is initially zero. The .NODEBUG
directive turns off listed values. Here are some makefile
examples:
.DEBUG : 4 (warn about unknown lines)
.DEBUG : 0 (no warnings)
.DEBUG : 6 (warn about unknown lines and undef’d macros)
.DEBUG : 1 (and display makefile lines)
.NODEBUG : 1 (now don't display makefile lines)
.DEBUG_GRAPHICS : Run-time debugging uses line-drawing characters.
.DEBUG_PRINT : –p Selects the print debugging information mode.
.DEBUG_RUN : –d Selects the run-time debugging mode.
.DEPENDENCY_IGNORED_FOR_REUSE : Ignores the files when omake determines whether a

file ... target object in a VOB can be reused (is up to date). By
default, omake considers that a target cannot be reused
if its dependencies have been modified or deleted since
it was built. This directive applies only to reuse, not to
winkin. Also, it applies only to detected dependencies,
which are dependencies that do not appear in the
makefile.
You can specify the list of files with a tail-matching
pattern; for example, subdir/%.module.
Unlike the files listed in most directives, the files on this
list refer to the names of dependencies, not to the
names of targets. As such, the directive may apply to
the dependencies of many targets at once. This
directive is most useful when identifying a class of
dependencies found in a particular toolset for which
common behavior is desired across all targets that have
that dependency.
.DO_FOR_SIBLING : file ... This directive is intended to be used in its negative

form (.NODO_FOR_SIBLING).
.NODO_FOR_SIBLING disables the creation of a
derived object for any file listed if that file is created as
a sibling derived object (an object created by the same
build rule that created the target). These sibling
derived objects are left as view-private files.
pattern, for example, %.lock.
list refer to the names of sibling objects, not to the
the siblings of many targets at once. This directive is
most useful when identifying a class of siblings found
in a particular toolset for which common behavior is
desired across all targets that have that sibling.

.ENV_OVERRIDE : –e Causes macro definitions from the environment to take

precedence over macros defined in the makefile. The
negation is .NOENV_OVERRIDE.
.ENVMACROS : Causes macro definitions to be made of each

environment variable. The negation is
.NOENVMACROS.
.INCLUDE : file ... The files are included at this point in the makefile. Each
file is treated as if %include file happened at this point.
.INCREMENTAL_REPOSITORY_SIBLING : The sibling files listed are incremental repository files

file ... created as siblings of a primary target. They may
contain incomplete configuration information, and
should prevent omake from winking in the primary
target. This is a very special-purpose directive, useful
when a toolset creates an incremental sibling object,
and the user wants more manual control over that
object.
pattern, for example, %.pdb.
list refer to the names of sibling objects and not the
.INCREMENTAL_TARGET : tgt ... Merges configuration record incrementally for the

listed targets. In other words, this directive combines
information from instances of this target generated
previously with the current build of this target. This
directive is most useful when building library archives,
because typically only some of the objects going into a
library are read each time the library is updated.
pattern, for example, %.a.
.KEEPDIR : –D Enables keep-directory mode. The first access of the

current directory or any search directory (see Search
Directories on page 106) to look for a file results in the
directory being read into memory and kept.
Subsequent accesses to the directory use the
in-memory version and occur much quicker. The
negation is .NOKEEPDIR.
.KEEPWORKING : –k Enables the keep-working mode. Any errors when

updating a target cause work on that target to be
stopped, but the make process continues. Because the
target was incompletely made, any other targets that
depend on it are prevented from being updated.
This mode is handy for long, unattended builds
because it maximizes the amount of making without
ignoring the exit status as the –i command-option
does. The negation is .NOKEEPWORKING.
.MACRO_CHAR : char Selects char as the new macro character. It is best to put
this directive in make.ini. For example, to change from
the default $ to + use the following:
.MACRO_CHAR : +
These characters cannot be the macro character:
(){},:=#'@*<&

.MAKE_MAKEFILE : –M Tells omake to make each makefile before trying to

read it.
In terms of implementation of this feature, omake
reads make.ini and checks whether
.MAKE_MAKEFILE is selected. If it is, omake makes
the first makefile and reads it. It then makes the next
makefile (if any) and reads it.
The make.ini file must supply the rule for making the
makefile. The .MAKE_MAKEFILE directive turns on
makefile making. You can put this rule in make.ini:
.MAKE_MAKEFILE
makefile:
imake -I \im\rules
Doing an omake target uses imake to create the
makefile. You can also define different search
directories for different extensions with the .PATH.ext
macro. See Search Directories on page 106 for more
information.
NOTE: Be aware that the makefile may contain
information (such as the imake directory) needed to
extract the makefile itself. You must provide some
means of determining the imake directory from
make.ini or you must give the imake directory to
omake with a command-line macro.
.MS_NMAKE : –EN Turns on fullest Microsoft NMAKE emulation. A

discussion of the extensive NMAKE emulation
capability is in the section Microsoft NMAKE
Compatibility on page 158.
The .MS_NMAKE, .omake, and .POLY_MAKE are
exclusive modes and only one of them is true. That is,
only one of $(.MS_NMAKE), $(.omake) or
$(.POLY_MAKE) is 1 at any one instance.
.NOCMP_SCRIPT : tgt ... Builds the specified targets as if the –O option were
specified; build scripts are not compared during
configuration lookup. This is useful when different
makefiles (and, hence, different build scripts) are
regularly used to build the same target.
The list of targets may be specified with a tail-matching
pattern, for example, %.obj.
.omake : –EO omake is set to its native emulation mode.
.POLY_MAKE : –EP Selects PM/CB emulation. A discussion of the

extensive PM/CB emulation capability is in the section
PM/CB (Intersolv Configuration Builder and PolyMake) on
page 145.
.REGEX_CHAR : char Selects char as the character that indicates special

regular expression character sequences. It is best to put
this directive in make.ini. For example, to change from
\ (the default) to ~ use this directive:
.REGEX_CHAR : ~
.REGEX_WILD : char Selects char as the regular expression wildcard

character that matches any single character. For
example, to change from . (the default) to ? use this
directive:
.REGEX_WILD : ?
.REJECT_RULES : –r Rejects all rules defined prior to this directive's

appearance. Use this directive at the top of make.ini to
reject omake's predefined rules. Use it at the top of
your makefile to reject all rules defined prior to the
makefile.
The –r command-line option is equivalent to a
.REJECT_RULES directive at the end of make.ini.
.RESPONSE.XXX : [ response definition ] Controls automatic response files (see Response Files on
page 110).

.RULE_CHAR : char Selects char as the new inference rule character. It is

best to put this directive in make.ini. To change from
the default of % to * use this directive:
.RULE_CHAR : *
.SHELL : [ .AUTO | .NOMULTI Names both the shell program and its flags and specifies
| .NOREDIR ] ... [ program flags ] that the shell program be used to execute every build
script. The .NOSHELL directive specifies that all build
scripts are executed directly, without using the shell
program. However, the + (use shell) and : (suppress
shell) shell-line prefixes override any .SHELL and
.NOSHELL directives.
The .AUTO keyword tells omake to determine when
to use the shell program. Without this keyword, the
.SHELL directive specifies that the shell program is
used for every shell line.
The .NOMULTI keyword tells omake not to do special
processing for multiple-command shell lines. By
default, omake turns the multiple-command shell line
into a batch file, which is executed.
The .NOREDIR keyword tells omake not to handle
redirection of I/O on the shell line. By default, omake
handles redirection except |. With this keyword, only
the shell program handles redirection. If .NOREDIR is
used with .AUTO, any redirection on the shell line
causes the command to be executed by the shell
program.
A .SHELL directive that is not given a program reverts
to the remembered shell program:
.SHELL :
to execute every build script. The initial value of
.SHELL:
.SHELL : $(COMSPEC) /C
To use the MKS shell, use the following syntax:
.SHELL : “$(ROOTDIR)\mksnt\sh”
.SIBLING_IGNORED_FOR_REUSE : file ... Ignores files when omake determines whether a target
object in a VOB can be reused (is up to date). This is the
default behavior, but this directive can be useful in
conjunction with the .SIBLINGS_AFFECT_REUSE
directive or –t command-line option. This directive
applies only to reuse, not to winkin.
pattern, for example, %.lock.
list refer to the names of sibling objects, not to the
.SIBLINGS_AFFECT_REUSE : –t Examines sibling derived objects when determining

whether a target object in a VOB can be reused (is up to
date). By default, omake ignores modifications to
objects created by the same build rule that created the
target (sibling derived objects). This directive tells
omake to consider a target out of date if its siblings
have been modified or deleted.
.SUFFIXES : [ extension ... ] Limits and orders the inference rules. .SUFFIXES is
provided for compatibility with other make utilities
and its use is discussed in the section Compatibility with
Suffix Rules (.SUFFIXES) on page 37.

.UNIXPATHS : Determines where omake looks for the inferred

dependency when the current target name has a
directory component. omake first tries matching the
pathed target name against pathed inference rules.
When the .UNIXPATHS directive is used, the second
step is to match the full target name against unpathed
inference rules, effectively causing omake to look for
the inferred dependency in the same directory as the
target.
The default behavior, and the behavior when the
.NOUNIXPATHS directive is used, is that the second
step matches the file name of the target name against
unpathed inference rules, causing omake to look for
the inferred dependency in the current directory and in
the search directories.
.WINK_IN : tgt ... –W This directive is intended to be used in its negative

form (.NOWINK_IN). .NOWINK_IN specifies that
the configuration lookup is restricted to the current
view for the listed targets.
The list of targets may be specified with a tail-matching
pattern; for example, %.obj.
omake supports many other directives for NMAKE, PM/CB, and Borland Make compatibility.
See Appendix D, Compatibility and Emulation for details.
3.9 Target Attributes

Target attributes are properties given to targets. Attributes can be either positive or negative, for
example, .ATTRIBUTE and .NOATTRIBUTE.

Using Attributes
Attributes are given to targets on dependency lines, but there are two forms:
target [ ... ] [ attribute ... ] : [ dependencies ... ]

attribute [ ... ] : [ target ... ]
Each form assigns the attributes to the indicated targets. The first form places the attributes after
the targets and before the colon, and each target is given all attributes.
The second form has attributes only to the left of the colon and targets to the right. Each target is
given all attributes. If no targets are listed, the attributes are given to all targets defined in the
makefile after this line. This is very useful. To give all targets in a makefile an attribute, put a line
of this form before any other dependency lines. This example that gives all targets defined after
this line the .PRECIOUS attribute:
.PRECIOUS : (near the top of the makefile)
Attributes and Inference Rules
Inference rules can have attributes, and the target being made with the inference rule inherits the
additional attributes of the rule.
A target's attributes have a higher precedence than a rule's attributes. If a target and a rule specify
an inconsistent attribute, the target's attribute is accepted.
List of Attributes
Table 16 lists attributes and their definitions.

Table 16 Attributes (Part 1 of 3)
Attribute Definition
.ALWAYS Always rebuilds this target, regardless of the results

of configuration lookup or the time stamps of its
dependencies. The –a command-line option is
equivalent to an .ALWAYS attribute for each target
in the makefile.
.CHAIN Enables chaining of inference rules. (See

Multiple-Step Inference Rules on page 33.)
.NOCHAIN disables chaining, which may increase
processing speed slightly.
.DEFAULT When omake is run without specifying any targets

on the command line, the first target (the default
target) is built. The lack of user control of the default
target makes it difficult to write generalized
makefiles that can be included by other makefiles.
The .NODEFAULT attribute indicates that this
target is not the default target. Use .NODEFAULT
for all non-rule targets in included, generalized
makefiles. That way a makefile can include the
generalized makefile without having the first target
of the generalized makefile be the default target.
In the generalized makefile you can put
.NODEFAULT as the attribute of each target, or put
.NODEFAULT :
at the top of the generalized makefile and
.DEFAULT :
at the bottom.
Note that targets defined in make.ini or in makefiles
included from make.ini are never the default target.

.IGNORE When making a target, a build script that returns a

nonzero status causes omake to terminate unless the
target has the .IGNORE attribute. The –i
command-line option is equivalent to a .IGNORE
attribute for all targets. The status of individual
build scripts can be ignored with the dash (–)
shell-line prefix.
.INFER omake uses inference rules to look for the inferred

dependency for targets that have no build scripts. To
omit, the inference rule check for targets without
build scripts, give them the .NOINFER attribute. To
force the inference rule check for targets with build
scripts, give them the .INFER attribute.
.MAKE Overrides the –n and –q command-line options. It is

useful when executing omake recursively. For
example:
nt .MAKE :
( cd msdos.dir ; omake $(MFLAGS) )
If you execute omake –n nt, omake changes
directory into nt.dir and executes omake –n.
Without this attribute, the result is to display
( cd nt.dir ; omake $(MFLAGS) )
Similarly, omake –q nt changes directory and
execute of omake –q.
The .MAKE attribute differs from the & shell-line
prefix, which overrides only the –n command-line
option. The appearance of $(MAKE) on the build
script also overrides only –n.
.NOCMP_NON_MF_DEPS : tgt ... Builds the specified targets as if the –G option were
specified; for each specified target, any dependency
not explicitly declared in the makefile is not used in
configuration lookup.

.NOCONFIG_REC : tgt ... Builds the specified targets as if the –L option were
specified; modification time is used for build
avoidance, and no CRs or derived objects are
created.
.PRECIOUS When a build script returns a nonzero status, omake

checks whether the current target has been written.
If it has, omake deletes the target, which prevents
corrupted files from being used. This attribute
prevents the deletion of the target itself and of
chained targets.
.RULE Is set when the percent sign appears on a

dependency line. You can use the .NORULE
attribute to allow a target with % in its name.
On rare occasions, you may want to use .RULE to
specify an inference rule that does not use %.
Because % is a wildcard character, a rule without it
matches exactly one target name.
.SILENT A target's build scripts are displayed before being

executed unless the target has the .SILENT attribute.
The –s command-line option is equivalent to a
.SILENT attribute for every target in the makefile.
The @ shell-line prefix also prevents display of the
build script.
3.10 Special Targets

omake uses some special targets at special times. The build scripts associated with these special
targets are used at specific times as described in Table 17.

Table 17 Special Targets
Special Target Usage
.AFTER : [ source ... ] After omake builds its last target and immediately before it
exits, any sources to .AFTER are built, and the build scripts of
.AFTER are executed. To execute the build scripts even when
using omake –n, give .AFTER the .MAKE attribute.
.BEFORE : [ source ... ] Before omake builds its first target, any sources to .BEFORE
are built, and then the build scripts of .BEFORE are executed.
To execute the build scripts even when using omake –n, give
.BEFORE the .MAKE attribute.
.DEFAULT[.ext] : If omake needs to update a target that has no build scripts,

omake looks for an inference rule that matches the target
name and uses the build scripts of the matched inference
rule. Each .DEFAULT[.ext] target is converted into an %.ext :
inference rule.
For PM/CB compatibility, we support .DEINIT, .EPILOG, .INIT, and .PROLOG. See
Appendix D, Compatibility and Emulation.
3.11 Search Directories

When omake looks for a file that has no path component in its name, omake’s default behavior
is to search only the current directory. The search can be tailored to include other directories, a
useful feature when your project is spread over multiple directories.
With omake’s search directory support you write dependency lines, such as
main.obj : main.c io.h
and have omake figure out where main.c and io.h are actually located.

Implied Location of Missing Files
omake uses the search directories to locate files. It is clear that if the file exists, the location of the
file is the directory it was found in. What happens if the file is missing? In this case omake
assumes that the missing file is located in the first directory of the appropriate search directory,
or . (the current directory) if none is appropriate.
Search Directory Macros
Search directories are set up with macro definitions. omake supports .PATH macros and VPATH
macros, but issues a warning if you use both.
The .PATH Macros
The .PATH[.ext] macros define the directories omake uses to find files that don't have a path
component. The optional .ext makes the .PATH.ext macro extension-specific. That is, .PATH.ext
defines the search directories only for files with extension .ext. The .PATH macro (with no
extension) controls the search for all files that aren't handled by a specific .PATH.ext.
The value of the .PATH macros is a list of directory names separated by semicolons (;). Here are
two examples:
.PATH = .;..
.PATH.obj = ..\obj
The first definition tells omake that all files can be found in . (the current directory) or .. (the
parent directory). The second definition tells omake that files with extension .obj can be found
in directory ..\obj (and in no other directory). Note that we have defined both a nonspecific
.PATH and an extension-specific .PATH.obj. omake uses directories defined in .PATH.obj to
search for files with the .obj extension, and uses .PATH for all other files.
Here is an example of a makefile that uses search directories:
OBJS = main.obj sub.obj io.obj

.PATH.c = ..
link $(.SOURCES), $(.TARGET), $(.TARGET,B);
The .c files are located only in the parent directory (..). Because no .PATH macro is defined,
omake searches for all other files only in the current directory.

The VPATH Macros
The VPATH macros are similar to the .PATH macros with one major exception: the VPATH
macros specify search directories in addition to the current directory. That is, the VPATH macros
always have . as the first directory.
Search Directories and Run-Time Macros
Macros in omake usually serve as a simple text replacement. However, the run-time macros
(.TARGET, .SOURCE, .SOURCES, and so on) include the location the target or source was
found. For example,
.PATH.obj = objs
main.obj : main.c
%echo $(.TARGET)
displays objs\main.obj because the main.obj target is located in the objs directory, and the
value of the .TARGET macro is the pathname of the target.
Here is a %.obj : %.c rule for Borland C that uses two run-time macros:
%.obj : %.c
$(CC) $(CFLAGS) –o$(.TARGET) –c $(.SOURCE)
The –o option names the output object file. $(.TARGET) is the name of the .obj file to be created,
including its path component, according to where omake found the file. If the .obj file was not
found, .TARGET is the implied location, either the first directory in .PATH.obj, if defined.
Otherwise, it is the first directory in .PATH, if defined, or finally, in the current directory.
Search Directories and File Lookup
When omake searches for a file that has no directory component, it looks in the appropriate
search directories. After the file has been located, omake assumes the file's location is permanent.
Occasionally, this behavior is in error. For example, suppose a project's current .c and .obj files
reside in a remote directory that .PATH references. You want to change a local copy of the
project's main.c and then compile and link the resulting local main.obj with the remote .obj files,
to produce a local .exe file.

Here is your makefile:
%.obj : %.c
OBJS = main.obj io.obj keyboard.obj

.PATH = .;c:\remote
link $(.SOURCES), $(.TARGET);
You change the local copy of main.c. If there is a remote main.obj but no local main.obj, when
you run omake, these commands are executed:
cl –c main.c
link c:\remote\main.obj c:\remote\io.obj c:\remote\keyboard.obj, project.exe;
The compilation uses the local main.c to produce a local main.obj, but the link uses
c:\remote\main.obj rather than the local main.obj. This is because there was no main.obj in the
current directory when omake started, and omake found main.obj in the c:\remote directory.
This example illustrates a general problem with omake. For reasons of speed, omake looks for a
file (target) one time only. After that, omake assumes the location of the target is constant. As you
have seen, this assumption can be wrong.
You can use the .REREAD target attribute to change this behavior. After executing the build
scripts that update the target, omake searches again for any targets with the .REREAD attribute
on disk. In the example, the make works correctly if the inference rule is given the .REREAD
attribute:
%.obj .REREAD : %.c

The inference rule has the .REREAD attribute because main.obj inherits the build scripts and
attributes of the matched inference rule.
Search Directories and Inference Rules
When omake tries to find an inferred dependency for an inference rule, it constructs a particular
file name. If the file name has no path component, omake tries the file name in each search
directory in order. Because each search directory is searched for each possible inferred

dependency, omake runs more slowly with a large number of search directories. The –D
command-line option (keep directory) speeds up this search.
Debugging the Search Directories
To debug your search directory choices, use the –p command-line option to print a section titled
Search directories.
For PM/CB compatibility, we support the .SOURCE directive, which is an additional way to
specify search directories. The .SOURCE directive can specify the search directory for specific
files, rather than for file extensions only.
For NMAKE compatibility, we support their search paths for dependents. See Microsoft NMAKE
Compatibility on page 158.
3.12 Response Files

Many programs (for example, Microsoft link and lib, Gimpel lint, and omake) can receive their
input from a response file. A response file is needed when the length of the build script exceeds
the Windows NT command-line limit of 1024 bytes.
omake supports:
➤ Automatic response files and variables, where it generates a response file or places long
command lines into an environment variable.
➤ Inline response files, where you place response file syntax around your build scripts.
The advantage of automatic responses (both files and variables) is that you don't worry about the
command-line limit. When the build script exceeds the limit, a response file or variable is
generated and the command is executed, using the response file or variable. You set up
automatic responses once, for your linker, librarian, and so on. Thereafter, the build script for
those programs can be arbitrarily long.
Inline response files are supported by a variety of other make utility vendors, and we support
the syntax they use.

Automatic Responses
A response class is a generalization that describes how and when to generate a response file or
variable. When a response class has been defined, you tell omake which program names accept
that class of response. We predefine most popular response classes, so you need only give omake
the names of programs that accept these predefined responses.
To tell omake which programs accept response files or use environment variables, use the
.RESPONSE.xxx directive, where xxx is a name of a response class. For example, we predefine a
LINK response class that describes the response file acceptable by Microsoft Link.
The .RESPONSE.xxx directive is used to define or modify a response class and to add program
names to an existing response class:
.RESPONSE.xxx : [ parameter ... ] [ program ... ]
The parameters describe the response class. Each program is either a base name (having neither
path nor extension) or a pathname. Normally, you use a base name only, but for special
circumstances you can use a pathname. If you use a pathname, you can have different response
classes based on the literal name of the shell-line program, the pathname to the shell-line
program, or the base name of the shell-line program.
Adding Program Names
To add program names to an existing response class, list them to the right of the directive. For
example, adding LINK-style automatic response file support for SLR Systems optlink and
Borland tlink is done as follows:
.RESPONSE.LINK : optlink tlink
The response classes are searched for program names in order from most recently defined to first
defined, so your use of a program name overrides any predefined usage.
Response Class Parameters
Table 18 shows the parameters, the meaning of each parameter and its default value, and
whether the parameter is used in response files and/or response variables.

Table 18 Response Class Parameters
Used in Used in
Parameter Meaning and Default
Files? Variables?
pre=ppp ppp is the prefix before the response file. If ppp yes yes
contains white space it must be enclosed in
double quotes.
suf=uuu uuu is the suffix (extension) of the response file. yes
env=eee 1 eee is the name of an environment variable. yes
in=num num is the shell-line length at which a response yes yes

file or variable should be generated. The default
is 1024 bytes.
sep=s s is the character that separates the program's yes

logical lines on the command line.
con=c c is the character that connects physical lines in yes

the response file into a single logical line.
out=num num is the output line length of lines in the yes

response file. The default value is 76.
1 The env parameter determines whether a response variable or response file is used. If env is
defined, an environment variable is used; otherwise, a response file is used.
Defining or Modifying a Response Class
To define a new response class or to modify an existing response class, use a .RESPONSE.XXX
directive with only parameters. One of pre, suf, or env is required when defining or modifying
a response type.
If env is defined, the response uses an environment variable, otherwise a file is used.
Disabling a Response Class
A directive with neither parameters nor program names removes support for this response class.
For example, you can turn off all LINK response files with this directive:
.RESPONSE.LINK :

Response File Example #1
This directive defines response file support for Microsoft link:
.RESPONSE.LINK : pre=@ suf=.rsp sep=, con=+ link
This declares the LINK response class has a prefix of @, a suffix of .rsp, logical lines on the
command line are separated with commas, and physical lines in the response file are connected
into one logical line with +<ENTER>. This directive also declares that the program named link
accepts LINK-style response files. With this definition, the build script
link module1.obj module2.obj ... moduleN.obj, test.exe;
is executed as
link @tempfile.rsp
The prefix @ tells link that what follows is the name of a response file. tempfile is a unique name
of the form tempdir\MAKEnum, where num is a 4-digit number, and the response file suffix .rsp
appears as the extension of the name. The response file contents:
module1.obj module2.obj ... +

... +
... moduleN.obj
test.exe;
The first logical line module1.obj ... moduleN.obj is broken up in to several physical lines in the
response file, each line ending with +<ENTER>. The next logical line test.exe; appears as the next
physical line in the response file.
Response File Example #2
This directive adds response variable support for Microsoft CL:
.RESPONSE.CL : env=CL e:\c6\bin\cl.exe
The CL response class uses environment variable CL, and e:\c6\bin\cl.exe accepts this kind of
response. Because a pathname is used, the command line
cl /DRELEASE="Release 1.0" /DOS=nt /Ic:\local\include ...
becomes the command line
cl

with environment variable CL having the value
/DRELEASE="Release 1.0" /DOS=nt /Ic:\local\include ..., but only if cl is found on the PATH
as e:\c6\bin\cl.exe!
Using Automatic Responses
Write the build script as if the length of the command line were unlimited. When the build script
gets too long, the automatic response is generated, a modified build script is executed, and the
automatic response is removed.
Generation of Automatic Responses
To generate an automatic response, omake determines whether the program name given on the
build script matches a .RESPONSE.xxx program name (the response names). To see whether
there is a match, omake does the following:
➤ If any response names have path components, it first compares the literal shell-line program
name with all response names. If there are no matches, it looks on disk for the pathname
that corresponds to the shell-line program name and compares the pathname to all
response names.
➤ If no match has been found yet, it compares the base name of the program and with the
response names.
The first match found determines the response class.
When the response class specifies a response file, the response file is generated if the build script
is longer than the in=num parameter.
When the response class specifies an environment variable and the build script is longer than the
in=num parameter, omake places the contents of the build script in the response environment
variable, appending to the value of the environment variable if it exists.
Deletion of Automatic Responses
Automatic response files are deleted after the build script that generated the response file is
executed, unless the –#8 command-line option was used. Each automatic response variable is
restored to its previous value after the build script that generated it executes.

Built-In Automatic Responses
omake has built-in automatic response file support for Gimpel lint and Microsoft cl, lib & link.
Built-in Responses for Windows NT
.RESPONSE.STD: pre=@ suf=.rsp omake link link32 lib lib32 cl cl386
.RESPONSE.LINT : suf=.lnt lint
In addition to automatic response files, omake also supports response files coded inline in the
makefile. Here is the syntax for an inline response file:
target :
command [ prolog ] << [ response_file ]
[ line copied to response file verbatim ]
.
.
.
<< [ epilog ]
The first << introduces the response file; the last << terminates it. response_file names the file. If
the name is omitted, omake generates a unique name of the form tempdir\MAKEnum.rsp, where
num is a unique number. Everything between the pair of << is placed in the response file and the
command is invoked as
command prolog response_file epilog
The prolog and epilog are optional text. Usually, prolog is used to indicate that the following
argument is a response file, and @ is the most common prolog.
The epilog can be used for redirection or other text. There are three special words that can appear
in the epilog:
➤ KEEP specifies that the response file is not to be deleted
➤ NOKEEP pecifies that it is be deleted.
➤ ECHO specifies that the contents of the response file be displayed. omake also shows the
contents of the response file when the –n command-line option is used.

Other build scripts can appear both before and after the inline response.
Deletion of Inline Response Files
Inline response files are deleted unless the –#8 command-line option is used, or the KEEP
keyword appears in the epilog. Response files named by omake are deleted after the build script
is executed. User-named response files are deleted immediately before omake exits.
Inline Response File Example
Here is an example of an inline response file for Microsoft LINK:
program.exe : $(OBJS)
link @<< $(MAKE_TMP)\link.rsp
$(OBJS,W+\n)
$(.TARGET)
$(.TARGET,B,>.map)
$(LIBS,W+\n)
$(.TARGET,B,>.def);
<< KEEP ECHO
Here, $(OBJS) and $(LIBS) are assumed to be strings separated by white space. The W macro
modifier replaces the white space with +<ENTER>, which is the appropriate line continuation for
Microsoft link. If OBJS has the value 1.obj 2.obj, and LIBS has the value 3.lib, these build scripts
evaluate to
link @$(MAKE_TMP)\link.rsp
where the contents of the response file are
1.obj+
2.obj
program.exe
program.map
3.lib
program.def;
The KEEP keyword has omake leave behind the response file. Otherwise, omake deletes it after
the build script finishes. The ECHO keyword tells omake to display the contents of the response
file after the link @$(MAKE_TMP)\link.rsp line is displayed. The default behavior is to display
the contents only when doing omake –n.

omake supports PM/CB local input scripts, NMAKE inline files and Borland Make's &&
redirection operator. See Appendix D, Compatibility and Emulation, for details.

4 Debugging Makefiles
4
This chapter describes the process of debugging omake makefiles.
4.1 Command-Line Options

The following options are useful for debugging makefiles that do not work correctly:
➤ The –#1 option displays each makefile line as it is parsed. Blank and comment lines are
removed, and conditional directives are shown in normal, expanded, and evaluated forms.
➤ The –p option displays omake's internal information. With this option, you can examine the
following:
➣ The values of all macros, to see where they were defined. Make sure they have the correct
values.
➣ All targets, their attributes, and their build scripts, to see where they were defined. This
list also shows which target is the default target.
➣ The inference rules, to see the rules known to omake.
➤ The –d option prints a run-time trace of omake. With this option, you can do the following:
➣ Ensure that the correct initialization file and makefile are being read.
➣ Watch omake use configuration lookup to compare the target to each dependency.
4 - Debugging Makefiles 119

➣ Watch omake search for inferred dependencies by using inference rules. Ensure that
omake attempts the inference rules you have defined. Cross-check with the –p flag.
For additional help with macros, the .DEBUG : 2 directive (–#2 option) causes omake to print a
warning if a macro is used without being defined.
Finally, the .DEBUG : 4 directive (–#4 option) warns about makefile lines that omake does not
know how to handle. This can catch misspellings of directives that you may have a hard time
detecting. This directive is especially useful if you are trying to use a makefile from some other
make utility.
4.2 Read-Time Debugging

To illustrate the process of debugging a makefile, the following command was executed:
omake –ndf demo –p –#1 CV=
This is the content of makefile demo:
# List of modules, and target name.

#
TARGET = project.exe
OBJS = a.obj b.obj
# If CV is defined, compile for debug
#
%if defined(CV)
CFLAGS = –Od –Z7 # compile for debug
LINKFLAGS = -debug -debugtype:both
%else
CFLAGS = –Ox # compile for size
LINKFLAGS = # no special link flags
%endif
# The default target in the makefile

#
$(TARGET) : $(OBJS)

# Additional dependency informationyou think they should
#
b.obj : $(TARGET,B).h
PATH.h = .;C:\SRC\H
Output Produced by –#1
For the purposes of this example, an empty make.ini file was created. Here is the output, with
annotations in italics.
First, the initialization file make.ini is read:

*** Read make.ini *** (Start reading the make.ini file)
*** Done make.ini *** (Done reading the make.ini file)
Next, the makefile demo is read:

*** Read demo *** (Start reading the demo file)
+ ("+" if more than one blank or comment line was
skipped)
3: TARGET = project.exe ("num: " identifies current line number)

4: OBJS = a.obj b.obj
+
8: %if defined(CV) (Conditional directive is shown
:---> true and evaluated)
9: CFLAGS = –Od –Z7 (Conditional was true, so this line and
10: LINKFLAGS = -debug -debugtype:both this line are processed)
11: %else (End of first true conditional block)
14: %endif (Lines are skipped until the "%endif")
+
18: $(TARGET) : $(OBJS) (The default makefile target)
+
21: b.obj : $(TARGET,B).h (An additional dependency)
+
23: .PATH.h = .;C:\SRC\H
*** Done demo *** (Done reading the "demo" file)

Output Produced by –p
Following output by the –#1 option, –p prints out information internal to omake, including facts
about macros, targets, search directories, and inference rules. The output begins with this line:
*** Begin print out *** (Starts the -p printout)
The Macro Definitions
The first block of output is the macro definitions including the names and values of macros and
the location they were defined. The location is any of the following:
built in Defined by omake, but changeable by you
predefined Defined by omake, but cannot be changed by you
command line Defined by you on the omake command line
file:number Defined by you in makefile file on line number
For brevity, most omake state macros have been omitted from this list:
.DEBUG_PRINT = 1 # predefined (state macro: the
–p flag)
.omake = 1 # predefined (state macro:
emulation)
BUILTINS = C:\Program Files\Rational\ClearCase\bin\ # predefined (name of the
make.ini built-ins file)
CC = cl # built in (name of the C
compiler)
CFLAGS = –Od –Z7 # demo:9 (C compiler flags)
CV = # command line (command-line
macro)
IMPLIB = lib # built in
LINK = link # built in (name of linker)
LINKFLAGS = -debug -debugtype:both # demo:10 (linker flags)
MAKE = C:\Program Files\Rational\ClearCase\bin\ # built in (make location)
omake.EXE
MAKEARGS = -ndf demo -p -#1 CV= # predefined (make command
line)
MAKEDIR = C:\SRC # predefined (make starting
directory)
MAKEFILE = demo # command line (name of the
makefile)
MAKEFLAGS = dnp -#1 # predefined

MAKEMACROS = CV=“ “ # predefined
MAKEVERSION = 200 # predefined (make version)
MFLAGS = –dnp -#1 # predefined (command-line
flags)
OBJS = a.obj b.obj # demo:4
OPUS = 1 # predefined (you know this is
OPUS)
OS = NT # built in (name of OS)
TARGET = project.exe # demo:3
The Search Directories
The search directories output lists both the extension-specific and nonspecific search directories.
The extension-specific directories are listed first:
*** Search directories ***

for .h : .\ C:\SRC\H\
all other files : .\
A .PATH macro is not defined, so all other files are searched for only in the current directory.
The Automatic Response Definitions
Following the search directories are all automatic response file definitions. Each line of output of
this section appears in exactly the form needed as if it were input to omake.
*** Automatic responses ***

.RESPONSE.WCC386: env=WCC386 pre=-u in=1024 wcc386
.RESPONSE.LINT: suf=.lnt out=76 in=1024 lint
.RESPONSE.STD: pre=@ suf=.rsp out=76 in=1024 wpp386 cl386 cl lib32 lib link32
The Inference Rules
Following the automatic response definitions is the list of inference rules:

*** Inference rules ***
* Suffix rules *
%.obj : %.c
defined in: internal
%.obj : %.cpp
$(CPP) $(CPPFLAGS) –c $(.SOURCE)
%.obj : %.asm
$(AS) $(AFLAGS) $(.SOURCE);
%.obj : %.for
$(FC) $(.SOURCE) $(FFLAGS)
%.res : %.rc
$(RC) $(RFLAGS) -r $(.SOURCE);
%.exe : %.obj
%do %.exe
* Meta rules *
%.lib :
%if ! %null(.NEWSOURCES)
%if %exists(${.TARGET})
$(IMPLIB) -OUT:$(.TARGET) $(LIBFLAGS) $(.NEWSOURCES) $(.TARGET)
%else
$(IMPLIB) -OUT:$(.TARGET) $(LIBFLAGS) $(NEWSOURCES)
%endif
%endif
%.exe :
$(LINK) -OUT:$(.TARGET) $(LINKFLAGS) $(.SOURCES) $(LINKLIBS)
The *Suffix rules* are rules that can be of the form .fromExt.toExt. The *Meta rules* are all
other inference rules.

The Targets and Build Scripts
A list of the targets follows the inference rules. The default target is listed first.
*** Targets and commands ***
>>> default target <<< (The first target in the makefile)
project.exe : a.obj b.obj (project.exe depends on a.obj & b.obj)
b.obj : project.h (b.obj depends on project.h)
The Final –p Output
Finally, after all –p output has appeared, you see this message:
*** Done print out ***

A Errors and Warnings
A
When omake encounters a problem, it produces an error, warning, or report:
➤ Errors are the most severe, causing omake to display a message followed by the word Stop.
omake does some deinitialization and quits.
➤ Warnings are less severe. A message followed by the text string (warning) is displayed and
omake continues.
➤ Reports are the least severe. A message is displayed and omake continues.
The diagnostic message uses either of these formats:
OMAKE: message.
OMAKE: file (line number): message.
In the second, number indicates the line in makefile file that produced the diagnostic. omake
displays the additional file and line information when possible.
A.1 Reducing Message Severity

When you use the –i command-line flag or dash (-) build-script prefix to ignore the exit status,
some error messages are downgraded to report messages. For example, if a build script returns
an exit status of 4, omake displays the following error message and quits:
OMAKE: Shell line exit status 4. Stop.
However, if the build script is prefixed with the dash, the report message is
A - Errors and Warnings 127

OMAKE: Shell line exit status 4 (ignored)
and omake continues.
A second common mode is omake's keep-working mode, specified with the –k command-line
flag or .KEEPWORKING directive. If omake is in this mode and encounters a problem when
executing a target's build scripts, it stops updating the target immediately. In this mode some
error messages are downgraded to report messages. For example, the previous error message is
now this:
OMAKE: Shell line exit status 4 (keep working)
A.2 Error Messages and Explanations

In Table 19, the diagnostic messages that are followed with the text
[ (ignored) | (keep working) ] are error messages that are downgraded to report messages if
either the ignore exit status or keep working mode is in effect.
NOTE: All messages that start with Test: are errors in the conditional expression tester.
Table 19 Error Messages (Part 1 of 6)
Message Severity Explanation
Bad foreach 'line' error The foreach line has an incorrect syntax. The correct
syntax is
%foreach name [ in ] list_of_values
Bad transformation macro error The PM/CB transformation macro had an incorrect
specification.
break/continue without while/foreach error A %break or %continue has occurred without a

preceding %while or %foreach.
Can't @include 'name' warning File cannot be read with the @ macro modifier.
Can't create lock in 'directory' error omake creates a lock file in the directory and bases the
name of its response, swap, and batch files on the name
of the lock file. The directory is either the value of the
MAKE_TMP macro (if defined) or the current
directory.

Can't %do 'name' warning The %do directive specified a nonexistent target name.
Use @ as a shell-line prefix before the %do to inhibit this
warning.
Can't find 'program' on PATH error or omake cannot execute the current build script because
[ (ignored) | (keep working) ] report the program was not found in any of the directories
specified in the PATH environment variable.
Can't have : and :: dependencies for 'target' warning The target was given on the left side of a regular (single
colon) dependency and also on the left side of a
double-colon dependency. These cannot be mixed.
Can't have both .PATH.ext and VPATH.ext error Both .PATH.ext and VPATH.ext macros were specified,
but omake does not know which macro to use to
determine the search directories.
Can't open file report The version control or library file can’t be opened for
or error reading or writing.
Can't open batch file 'file' error When executing multiple commands (see Using
Multiple-Command Build-Script Lines on page 73) the
commands are written to a batch file and the batch file
is then executed. The probable cause of this error is that
the MAKE_TMP macro has not been set properly.
Can't open inline response file 'name' warning The inline response file name can’t be opened for
writing.
Can't open response file 'name' warning The automatic response file name can’t be opened for
writing.
Can't read makefile 'name' warning The makefile name can’t be read.
Can't read response file 'name' warning The response file indicated on the command line by
@name can’t be read.
Can't redirect 'file' error The redirection for file was specified incorrectly.
Can't undefine macro 'name' warning The macro name can’t be undefined because it is a
predefined macro.

COMSPEC not found error The COMSPEC environment variable, used to initialize
the name and flags of the shell program, was not found
in the environment.
Dependency-line wildcard 'spec' is null warning The wildcard specification spec didn’t expand to any file
names.
Device error error or Execution of the build script caused a device error.
[ (ignored) | (keep working) ] report
Directive failed error or The directive on this build script failed.

Directive nesting too deep. Max 31 levels. error Only 31 levels of foreach, if, and while nesting are
supported, and your makefile has more than this.
%do macro missing '=' error A %do macro was missing the equal sign that is needed
for its definition. The most probable cause is that there
are spaces around it. If you need spaces, enclose them
(or the macro definition) in double quotes.
Don't know how to make 'target' error or omake has exhausted all means of making target. If
[ (keep working) ] report omake is in keep-working mode, it marks this target as
unusable and abandons work on it. Use the –d (debug)
and –p (print info) options to determine the cause.
else without if error An %else directive encountered without a preceding

%if.
elseif without if error An %elseif directive encountered without a preceding

%if.
end without while/foreach error An %end directive encountered without a preceding

%while or %foreach.
endif without if error An %endif directive encountered without a preceding

%if.
Exec 'command' failed error The %exec command directive cannot be executed.
Inline failed warning The current inline response file couldn’t be generated
correctly.

Internal error: number error Report the internal error number to Customer Support.
Invalid executable 'program' error or The specified program is not a valid executable.
Invalid SHELL 'program' warning The specified shell program was not found.
Macro 'name' has no value warning A run-time macro is used before its value was set.
Macro 'name' not found warning The macro called name is undefined. This message
appears only when the .DEBUG directive or –#
command line flag has been set to 2. When debugging
your makefiles, this helps catch misspellings.
Maketmp failed error The routine that generates temporary file names failed,
probably because the directory for the temporary files is
full.
Max 8 response files for 'cmd' error The multiple-command build script generated more
than eight (the maximum) response files.
Missing character in 'name' error omake was looking for character to mark the end of
macro name but came to the end of the line instead.
No targets on dependency line error The dependency line being processed had no targets on
the left side.
Not enough memory to exec 'name' error or There is not enough memory to execute the specified
[ (ignored) | (keep working) ] report program name.
Nothing to make error No target was specified on the command line, and no
default target exists in the makefiles.
Option 'letter' needs value error A required argument with the command-line option
–letter was not specified.
Out of memory error omake ran out of memory. Your makefile may be too
large, or omake may have installed a TSR.
Recursive macro 'name=value' warning The macro called name has a recursive value; the value of
the macro refers to itself. omake ignores the recursion.

Regex: message warning The regular expression cannot be interpreted correctly;

message describes the reason.
Removing target report The target is removed because it was modified when an
error occurred in one of the build scripts that updated
it. Removal prevents partially written files (such as
object files) from being left behind when a compilation
fails. Targets with the .PRECIOUS attribute are not
removed.
Shell line exit status number error or The most recently executed build script had a nonzero
[ (ignored) | (keep working) ] report exit status of number.
Shell line too long error or The build script exceeded the command-line limit. Use
[ (ignored) | (keep working) ] report a response file. See Response Files on page 110.
Shell line without target error There is no target to which this build script belongs.
Build scripts are indented from the left column by a tab
or space character.
Test: bad expression "exp" error The expression exp cannot be parsed.
Test: bad first operand "op" error In a comparison, the first argument op cannot be parsed.
Test: bad operator "op" error The op is not one of the acceptable logical operators:
&&, | |, and = =.
Test: bad second operand "op" error In a comparison, the second argument op cannot be
parsed.
Test: no operand before "text" error There was no argument in a previous functional
operator.
Test: unexpected "text" error The expression is finished logically, but there was more
text on the line.
Test: unmatched quote "text" error Pairs of single or double quotes delimit strings that
contain white space. A quote in text was not matched.
Test: unknown function "name" error An expression of the form name() has an unrecognized
name.
Test: unknown operator "op" error The operator op is not recognized.

Test: ')' expected; got "text" error A right parenthesis is expected, but text was found
instead.
Too many %do macros. Max of 10. error More than 10 macro definitions were specified for each
%do directive.
Too many shell lines for 'target' warning A target on a regular (single colon) dependency line can
be given build scripts one time only. The additional
build scripts are ignored.
Unexpected what error The what is either end or endif, and there is no previous
%foreach, %while, or %if directive.
Unknown option 'letter' Only the command-line options listed by omake –h are
acceptable.
Unknown status error or The build script returned an unknown status.

Unmade 'target' due to source errors warning The –k flag was used to do the maximum useful work.
One of target's dependencies was not made because of
an error, so the target cannot be made.
Unrecognized .RESPONSE keyword warning The .RESPONSE directive does not accept the keyword
'word' word.
Unrecognized line 'line' warning omake cannot parse this line.
Unterminated inline from line number error The inline response file was not terminated before the
end of the makefile.
Unterminated what from line number error The what of if, elif, or else that started on line number
was not balanced before the end of the makefile was
reached.
User interrupt error The user typed <CTRL-BREAK> or <CTRL-C> while omake
was executing a build script.

B Exit Status Values
B
omake controls the execution of other programs, and the memory that omake itself uses is
unavailable to these programs while omake is controlling them. When omake exits, it returns an
exit status indicating the termination reason. The exit status can be tested by the command shell
or, if you are doing a recursive make, by omake itself. The exit status can also be accessed in the
.AFTER special target with the MAKESTATUS macro. For example:
.AFTER :
%if $(MAKESTATUS) == 3
%echo omake: The final build script exited with status: $(status)
%endif
Table 20 lists exit values and their meanings.
Table 20 Exit Status Values (Part 1 of 2)
Exit Status Meaning
0 Normal exit with no errors.
1 General purpose error if no other explicit error is known.
2 There was an error in the makefile.
3 A build script had a nonzero status.
4 omake ran out of memory.
5 The program specified on the build script was not executable.
6 The build script was longer than the command processor allowed.
B - Exit Status Values 135

Table 20 Exit Status Values (Part 2 of 2)
Exit Status Meaning
7 The program specified on the build script cannot be found.
8 There was not enough memory to execute the build script.
9 The build script produced a device error.
10 The program specified on the build script became resident.
11 The build script produced an unknown error.
16 The user typed <CTRL-C> or <CTRL-BREAK>.

C Built-In Macros and Rules
C
This chapter describes the macros and rules that omake uses.
C.1 Macros
This section describes general macros, state macros, and built-in macros.
Predefined General Macros
Table 21 lists predefined general macros; these macros cannot be redefined.
Table 21 General Macros (Part 1 of 2)
Predefined Macro Value
.NEWSOURCES The list of target dependencies newer than the target; all
dependencies when configuration lookup is enabled and the target is
not marked as .INCREMENTAL_TARGET
.SOURCE The inferred dependency or, if none, the first explicit dependency
.SOURCES The complete list of dependencies for a target
.TARGET The name of the target being made
.TARGETROOT The root name of the target being made
C - Built-In Macros and Rules 137

Table 21 General Macros (Part 2 of 2)
Predefined Macro Value
BUILTINS The pathname of the built-ins file
CWD The current working directory (the directory in which omake starts)
FIRSTTARGET The first command-line target or the first makefile target
INPUTFILE The current makefile being processed
MAKEARGS All command-line arguments
MAKEDIR The directory in which omake starts (same as CWD)

MAKEMACROS All command-line macros
MAKESTATUS The exit status with which omake exits
MAKETARGETS All command-line targets
MAKEVERSION The version of this omake executable
status The exit status of the last build script executed
Predefined State Macros
These predefined macros return the state of omake's command-line flags and directives. In
Table 22, when yes appears in the Directive column, the value of the named macro is the state of
the like-named directive. The Flag column shows the command-line flag that is the equivalent of
the directive (if any).
NOTE: .ALWAYS, .IGNORE, and .SILENT are actually target attributes. They look like directives
when they appear on the target side of a dependency line without any dependencies.
Nevertheless, the .ALWAYS, .IGNORE, and .SILENT macros have the correct value, as if these
attributes were directives.

State Macro Directive Flag Macro Value
.ALWAYS Yes –a 0 or 1
.CASE_MACRO Yes 0 or 1
.CASE_TARGET Yes 0 or 1
.DEBUG Yes –# The current debug options
.DEBUG_PRINT Yes –p 0 or 1
.DEBUG_RUN Yes –d 0 or 1
.ENV_OVERRIDE Yes –e 0 or 1
.ENVMACROS Yes 0 or 1
.GLOBAL_PATH Yes 0 or 1
.IGNORE Yes –i 0 or 1
.IGNORE_MFLAGS –z 0 or 1
.KEEPDIR Yes –D 0 or 1
.KEEPWORKING Yes –k 0 or 1
.MAKE_MAKEFILE Yes –M 0 or 1
.MS_NMAKE Yes 1 –EN 0 or 1
.OMAKE Yes 1 –EO 0 or 1
.OPUS_52X Yes –E2 The list of compatibility features
.POLY_MAKE Yes 1 –EP 0 or 1
.REGEX_BACK The regex literal backslash
.REGEX_CHAR Yes The regex escape character
.REGEX_DOT The regex literal dot
.REGEX_WILD Yes The regex “match any character”
.REJECT_RULES Yes –r 0 or 1

State Macro Directive Flag Macro Value
.RULE_CHAR Yes The rule character
.SHELL Yes The shell program and shell flags
.SILENT Yes –s 0 or 1
.SUFFIXES Yes The list of suffixes
.UNIXPATHS Yes 0 or 1
1. Exactly one of $(.MS_NMAKE), $(.OMAKE), or $(.POLY_MAKE) is 1.
Built-In Macros
Table 23 lists macros that are defined by omake, but can be changed.
Table 23 Built-In Macros
Built-In Macro Definition Default Value
AS Assembler masm
CC C compiler cl
FC FORTRAN compiler f77l
IMPLIB Windows NT object librarian lib
LINK Object linker link
MAKE Pathname to command-line name
MAKEFILE First makefile read
OS The operating system NT
RC Resource compiler program rc
SHELLCOMMANDS List of internal shell commands Not predefined

C.2 Macro Modifiers
When a macro is referenced, the value can be modified through the use of macro modifiers. To
modify a macro, reference it as follows:
name is expanded, and each modifier is applied in order to the elements of the value.
Table 24 lists macro modifiers and the actions they perform.
Table 24 Macro Modifiers (Part 1 of 2)
Modifier Action
number Selects the numberth element of the value.
>string Appends string to each element.
<string Prepends string to each element.
from=to Substitutes occurrences of from with to. If from does not appear in an element,
the element is not changed.
*F or * Each element is a wildcard spec evaluating to a list of files.
*D Each element is a wildcard spec evaluating to a list of directories.
@ Includes element as a text file.
@/from/to/ Includes lines in text file, matching lines with regex from and replacing the
matched part of the line with regex to.
A/ Converts element to an absolute file name using path separator /.
A\ Converts element to an absolute file name using path separator \.
A Converts element to an absolute file name using the default path separator \.
B Selects the base part of the element.
D Selects the directory part of the element.
E Selects the extension part of the element.

F Selects the file part of the element.

Table 24 Macro Modifiers (Part 2 of 2)
Modifier Action
LC Converts the element to lowercase.
Mregex Chooses elements that match regular expression regex.
M"spec" Chooses elements that match file specification spec.
Nregex Chooses elements that do not match regular expression regex.

N"spec" Chooses elements that do not match file specification spec.
P Selects the path part of the element.
R Selects the root part of the element
S/from/to/ Substitutes from (a regular expression) to to. If from does not match an
element, the element is not changed
UC Converts the element to uppercase.
Wstr Replaces white space between macro elements with str.
X Replaces element names with pathnames using the search directories.
Z Selects the drive part of the element.
C.3 Inference Rules

The following list of rules have been predefined in omake and are available unless you use the
–r command-line flag or .REJECT_RULES directive.
Notice how the rules make use of macros. For example, the %.obj : %.c rule invokes the program
$(CC) with the flags $(CFLAGS). In your makefile, set CFLAGS to the flags to be passed to the
compiler.
C source to object file, using the Microsoft C/C++ compiler:

CC = cl Windows NT
%.obj : %.c

C++ source to object file, using the Microsoft C/C++ compiler:
CPP = cl Windows NT
%.obj : %.cpp
$(CPP) $(CPPFLAGS) –c $(.SOURCE)
Assembler source to object file, using Microsoft MASM:
AS = masm
.%.obj : %.asm
$(AS) $(AFLAGS) $(.SOURCE);
FORTRAN source to object file, using Lahey FORTRAN:
FC = f77l
%.obj : %.for
$(FC) $(.SOURCE) $(FFLAGS)
Resource compiler script to resource file, using the Microsoft Resource Compiler:
RC = rc
%.res : %.rc
$(RC) $(RCFLAGS) –r $(.SOURCE)
Update an executable, using Microsoft link32 for Windows NT:
LINK = link
%.exe :
$(LINK) –OUT:$(.TARGET) $(LINKFLAGS) $(.SOURCES) $(LINKLIBS)
Object file to executable, using the %.exe rule:
%.exe : %.obj
%do %.exe
Update a library, using Microsoft lib for Windows NT:

IMPLIB = lib
%.lib :
%if ! %null(.NEWSOURCES)
% if %file(${.TARGET})
$(IMPLIB) –OUT:$(.TARGET) $(LIBFLAGS) $(.TARGET) $(.NEWSOURCES)
% else
$(IMPLIB) –OUT:$(.TARGET) $(LIBFLAGS) $(.NEWSOURCES)
% endif
%endif
The ccase-home-dir\builtins.cb and ccase-home-dir\builtins.nm files define the inference rules

used by PM/CB and NMAKE, respectively. omake reads these files when PM/CB or NMAKE
emulation is chosen; it searches for these files in the same manner in which it searches for the
make.ini file (see Locating the Initialization File on page 46).

D Compatibility and Emulation
D
omake achieves a great deal of compatibility with other make utilities, in that it understands the
makefiles and features of other makes. When a feature or operation of omake differs from that
of the other utilities, we provide emulation, so that omake operates like the other vendor’s make
utility.
D.1 PM/CB (Intersolv Configuration Builder and PolyMake)

omake is highly compatible with PolyMake up to v4.0PM/CB Compatibility and with Intersolv
PVCS Configuration Builder, to v5.1. omake supports all PM/CB macros and transformation
macros, library object modules, local input scripts, and most directives.
System Macros
omake supports the PolyMake/Configuration Builder system macros listed in Table 25:
Table 25 PM/CB System Macros (Part 1 of 2)
System Macro Value omake Macro
_Arguments The command-line arguments. MAKEARGS
_Cwd The current working directory. MAKEDIR
_Directory The current working directory. MAKEDIR
D - Compatibility and Emulation 145

Table 25 PM/CB System Macros (Part 2 of 2)
System Macro Value omake Macro
_Exe The pathname of the omake program. MAKE
_FirstTarget The first command-line target or, if none, the FIRSTTARGET

first makefile target.
_Flags The command-line flags. MAKEFLAGS
_FlagsMacros The command-line flags and macros.
_InputFile The current makefile. INPUTFILE
_Macros The command-line macros. MAKEMACROS
_PctStatus The status of last operation line. status
_Script The default or first-named makefile. MAKEFILE
_Source The source for current target. .SOURCE
_SourceRev Version control version of current target. .VERSION
_Sources The sources for current target. .SOURCES
_SysVer Operating system version: major.minor.
_TargRoot The root part of the current target name. .TARGETROOT
_Version The PM/CB version number. omake reports

v5.1.
M_ARGS See _MakeArgs above. MAKEARGS
MAKEARGS See _MakeArgs above. MAKEARGS
MAKEVER See _Version above.
PCTSTATUS See _PctStatus above. status

Transformation Macros
omake supports PM/CB v5.x, both long and one-letter names. When an omake equivalent
exists, it is listed in the right column of Table 26. Note that PM/CB macros work on text; omake
modifiers work on a macro, which is expanded into text and modified. To use the omake
equivalent, you need to define macro name with value text.
Table 26 PM/CB Transformation Macros (Part 1 of 2)
Trans. Macro Long name Result omake Equiv.
$[@,text] Include Include contents of file text. $(name,@)
$[c,str,begin,end] Clip Clip string str between begin

and end.
$[d,text] Directory Directory part of text. $(name,D)
$[e,text] Extension Extension part of text. $(name,E,.=)
$[f,path,list,ext] Filename Build file name from path, list

and ext.
$[l,text] Lower Convert text to lowercase. $(name,LC)
$[m,spec,text] Match Elements in text that match file $(name,M"spec")

spec.
$[n,text] Normalize Normalize text as absolute file $(name,A)

name.
$[p,text] Path Pathname of text. $(name,P)
$[r,text] Base Base name of text. $(name,B)
$[s,separator,text] Separators Replace text separators with $(name,Wsep) 1

separator.
$[t,t1,t2,list] Translate Translate list mapping letters
t1 to t2.
$[u,text] Upper Convert text to uppercase. $(name,UC)
$[v,text] Drive Drive label of text.
$[w,text] FileList Wildcard expand file spec. text. $(name,*F)

Table 26 PM/CB Transformation Macros (Part 2 of 2)
Trans. Macro Long name Result omake Equiv.
$[x,text] DirList Wildcard expand directory $(name,*D)

spec. text.
none $[Root,text] Root name of text. $(name,R)

1. The separator can be enclosed in double quotes, which means that the separator must be
parsed for special character sequences. The sep is never enclosed in double quotes and
is always parsed for special characters.
Built-In Functions
In conditional expressions omake accepts:
%status
This is the exit status of last build script. It is the same as $(status) in omake.
Built-In Operations (Percent Directives)
%exit [ status ]
This directive terminates the make process with exit status (0 if status isn't given). Before
terminating, the .DEINIT and .EPILOG special targets are run if they are defined.
.DEINIT is run only if .INIT was run.
Directives
omake supports most PM/CB directives. For read-time interpretation, the directive starts in the
first column of the makefile. For run-time interpretation, PM/CB emulation must be chosen and
the directive must be indented.
Table 27 provides short descriptions of the supported directives.

Table 27 PM/CB Directives (Part 1 of 3)
Applicable
Time
.ExtraLine read time Causes an additional carriage return/linefeed to be

output after each build script is executed. The negation
is .NoExtraLine.
.Emulate [ Builder | NMAKE ] read time Sets the emulation mode to either Builder (PM/CB) or
NMAKE.
.Ignore read time Causes nonzero build-script status to be ignored.
.Include [ = ] file ... read time Reads each file. If PM/CB emulation is chosen, omake
looks for relative file names in the current directory. If
PM/CB emulation is not chosen, omake treats
.INCLUDE file the same as %include(file).
.KeepDir read time Same as the omake .KEEPDIR directive.

run time
.KeepIntermediate read time Prevents intermediate files from being deleted.

run time
.KeepWorking read time Same as the omake .KEEPWORKING directive.

run time
.Keep_Lis read time Prevents local input scripts (inline response files) from
run time being deleted. The negative of this directive is
.NoKeep_Lis.
.KeepTemp read time Same as the .Keep_Lis directive.

run time
.Lis_Comments read time Causes # in local input scripts to be considered literally,

rather than as a comment character. You can also use \#
to mean a literal #.
.Lis_File [ = ] [ filename ] read time Names the local input script file. Unlike PM/CB,
run time omake allows a blank filename to reenable automatic
generation of the local input script name.
.Logfile files read time Handled by omake as .PVCS_STORAGE : files

Applicable
Time
.Ms_Nmake read time Same as the omake .MS_NMAKE directive.
.NoEnvMacros read time Same as the omake .NOENVMACROS directive.
.Order read time Same as the PM/CB .Suffixes : directive.
.Path.xxx [ = ] dir-1[;dir-2]... read time Treated as an omake .PATH.xxx = dir-1[;dir-2]... macro

definition.
.PermitComments read time Same as .Lis_Comments.
.Poly_Make read time Same as the omake .POLY_MAKE directive.

run time
.Precious read time Same as the PM/CB .KeepIntermediate directive.
.RejectInitFile read time Same as the omake .REJECT_RULES directive.
.Remake read time Targets are fully made each time they are encountered
as sources.
.Shell [ shell_program ] read time Similar to the omake .SHELL directive, but without
run time automatic detection of when to use the shell program.
.Silent read time Build scripts are not displayed on the screen before
execution.
.Source [ = ] dir_list file_list read time Gives search directories to files. The dir_list is a
semicolon-separated list of directories. The file_list is a
space-separated list of filenames or file extensions.
When omake looks for a file, it sees whether the file
name or its extension is on any file_list; if it is, omake
searches for the file on the dir_list.
.Source [ .ext ] : [ dir-1 dir-2 ... ] read time This PM/CB .SOURCE dependency is identical to a
.PATH macro and is treated as an
.PATH.xxx = dir-1[;dir-2]... macro definition.
.Suffixes [ : ] read time Without a colon, this directive acts like omake's
.REJECT_RULES directive. With a colon, it acts like
the omake .SUFFIXES directive.

Applicable
Time
.VolatileTargs read time Same as the PM/CB .Remake directive.
Reserved Targets
omake supports the following PM/CB reserved targets:
.DEINIT [ : ]
If the .INIT special target was used, the build scripts of the .DEINIT target are executed
immediately before the build scripts of .EPILOG are executed.
.EPILOG [ : ]
The same as the omake .AFTER special target. If emulating PM/CB, omake looks only
for .EPILOG. Otherwise, it looks for .AFTER, and then .EPILOG.
.INIT [ : ]
This target's build scripts are executed immediately before any other build scripts.
.PROLOG [ : ]
The same as the omake .BEFORE special target. If emulating PM/CB, omake looks only
for .PROLOG. Otherwise, it looks for .BEFORE, and then .PROLOG.
Local Input Scripts
omake accepts the PM/CB response file syntax:
target :
command [ prolog ] <X<[ text ]
build script
.
.
.
< [ epilog ]

where X is a single character, usually @. If text is given, it is copied to the response file. Each build
script is then copied to the response file. The command syntax is
command prolog Xtempfile epilog
where tempfile is a temporary file with a name of the form makenum.rsp
Operation-Line Modifiers
omake supports all PM/CB operation-line modifiers, which are listed in Table 28.
Table 28 PM/CB Operation-Line Modifiers
PM/CB operation-line modifier omake shell-line prefix
(Always) &
(ExtraLine) >
(Ignore)[ status ] - [ status ]
(Iterate) !
(NoShell) : (* if emulating PM/CB)
(Shell) +
(Silent) @
(TrackErrors) ~
PM/CB Emulation
omake is highly compatible with PM/CB, but there are differences in how they read makefiles
and in how they run. In PM/CB emulation mode,omake operates like PM/CB.

Emulation at Startup Time
If PM/CB emulation mode is selected at startup time, omake emulates the PM/CB command
line and selection of the built-ins file. Selection of PM/CB emulation at startup time is done with
the –EP flag, either on the command line or in the OMAKEOPTS environment variable.
To determine the startup emulation mode, omake examines the OMAKEOPTS environment
variable for –Ex flags. It then examines the command-line for –Ex flags. If the last –Ex flag is –EP,
omake starts up emulating PM/CB.
Emulation After Startup Time
The .POLY_MAKE directive turns on PM/CB emulation mode from the point it appears in the
initialization file or any makefile.
The Command Line
First the OMAKEOPTS environment variable is parsed for options. Then, if omake is emulating
PM/CB at startup, the MAKEOPTS and BUILD environment variables are parsed for options. Then
the command line is parsed. Parsing entails the following:
➤ Handling the case-insensitivity of the PM/CB command line.
➤ Mapping options into omake equivalents. The help screen switches to the options available
with our emulation of PolyMake v4.0 and Configuration Builder v5.x. The help screen's
contents are stored in the file omhelp.cb.
➤ Handling the long-named Configuration Builder v5.x options.
➤ Warning about unconvertible command-line options.
Without emulation at startup, the command line is as documented in the section Command-Line
Options on page 41.
The Emulation File (BUILTINS.CB)
If omake is emulating PM/CB at startup, omake reads its internal rules and macros. It then looks
for builtins.cb first in the directory of make.ini, in the directory of omake.exe, and along
directories of the INIT environment variable, in that order. If builtins.cb is found, it is read for
macros and rules that give more complete PM/CB emulation.

The Initialization File (TOOLS.INI)
PM/CB distinguishes between an initialization file and a built-ins file. Both contain initialization
information. The initialization file is almost always the file named tools.ini. If omake is
emulating PM/CB at startup, omake searches for the initialization file and reads the first one it
finds :
1. Named by the –Init command-line option
2. Named tools.ini in the current directory
3. Named tools.ini in a directory named by the INIT environment variable
If omake finds the initialization file, it reads information in the file starting with the section
heading [PVCS.
The BUILTINS File
If omake is emulating PM/CB at startup, omake searches for the built-in file, and reads the first
one it finds:
1. Named by the –b command-line flag
2. Named builtins in the current directory
3. Named builtins.mak in the current directory
4. Named by the BUILTINS environment variable
If omake is not emulating PM/CB at startup, it uses the method documented in the section
Locating the Initialization File on page 46 to locate the initialization (built-ins) file.
The Makefile
If omake is emulating PM/CB at start up, omake looks for the default makefile in this order:
script.bld, script, makefile, makefile.mak. The first file found is read. When trying to read a
makefile file that doesn't have an extension, omake tries file, file.bld, and file.mak, in that order.
Makefile Contents
If omake is emulating PM/CB:
➤ The line continuation character sequence \<ENTER> is removed from the input.

➤ The ^ character is used for quoting and produces the following effects:
Character Effect
^^ literal ^
^<ENTER> literal newline
^$ literal $
^\ literal \
^# literal #
➤ The PM/CB-compatible directives listed in the previous section can be used at run time as
well as read time.
➤ The %end directive is the same as %endif. Without emulation, %end terminates a
%foreach or %while directive.
➤ .INCLUDE file searches only the current directory.
➤ Makefile macro names are case-sensitive.
➤ The =+ macro (prepend) definition is supported.
➤ Environment variables override built-in macro definitions.
➤ The MFLAGS macro is the same as _FlagsMacros.
➤ The TMP, TEMP, and WORK environment variables are tried, in order, for the location of the
directory where temporary files are created. Without emulation, omake uses the value of
the MAKE_TMP macro.
➤ Duplicate entries in a target's dependencies are allowed. Without emulation, omake

removes duplicate dependencies.
Operation Lines (Build Scripts)
If omake is emulating PM/CB:
➤ The shell program is used for executing every build script. The * and (NoShell) prefixes
suppress execution of the shell program for this build script.
➤ The build scripts cd dir and chdir dir both change to directory dir and stay there until
changed back explicitly. Otherwise, omake starts each build script from the directory that
omake started in the $(MAKEDIR) directory.

➤ The : modifier means swap out of memory and * means suppress shell. Without emulation,
* means swap out and : means suppress shell.
➤ The - shell-line prefix does not print the Error code ... (ignored) message.
Unsupported PM/CB Features
When running PM/CB makefiles with omake, be aware of the lack of support for some PM/CB
features. The list here provides a workaround, when it is available.
Unimplemented Directives and Reserved Targets

.ArcFile Tells PM/CB about ARC compression files.
.ExamineCmt (also Checks comments for line continuation. Although this directive
known as .Examine_Cmt) is unsupported, omake handles line continuation inside
comments.
.Error[.xxx] Reserved target; supplies auxiliary instructions for building
targets when a build script returns a nonzero exit status. Use
conditional directives instead.
.FootPrint Directive; controls foot printing targets with an internal
comment record.
.Ms_Make Selects Microsoft MAKE emulation.
.NoEnvInherit Prevents the environment from being passed to child processes.
.Rebuild Directive; rebuilds previous versions of applications.
.ZipFile Directive; tells PM/CB about ZIP compression files.
Iteration Groups
An iteration group does an implicit iteration over the $? macro. For example:
test.lib : test.obj chart.obj input.obj

{
lib contract.lib –add $? omake m2 noask
}
The %foreach directive can be used in its place:
test.lib : test.obj chart.obj input.obj

%foreach file in $?
lib contract.lib –add $(file) omake m2 noask
%endfor

Suffix Dependencies
Suffix dependencies allow the specification of a set of suffixes to be tried when building a
particular target. This is not supported by omake.
Command-Line Flags
The following PM/CB command-line flags are not supported: –Batch, –C, –Compile,
–NoEnvInherit, or –Rebuild.
Makefile Contents
The ~ suffix on archive extensions to handle like-named files in different directories is not
supported. Instead, use a pattern-matching rule , such as this:
%.c : c:/apps/archives/%.c
get –q –w $(_SourceRev) $(_Source)($(_Target))
The _DefaultSuffixes system macro is not supported.
Shared Definitions
omake does not parse the PVCS Version Manager configuration file automatically.
Operation-Line Modifiers
For PM/CB, the (Always) operation-line modifier (Shell-Line Modifiers) overrides the –Touch
flag. This is not supported in omake. You can use the .MAKE attribute instead. For example:
# PM/CB
recursive :
(Always)$(MAKE) $(MAKEFLAGS)
# Omake
recursive .MAKE :
$(MAKE) $(MAKEFLAGS)
Built-In Operations
The built-in operations that handle foot printing are not supported. These operations are
%EAStamp, %ExeStamp, and %ObjStamp.

D.2 Microsoft NMAKE Compatibility
omake is highly compatible with Microsoft NMAKE up to version 1.3 (the version supplied with
Visual C++). omake supports all NMAKE directives, macros, paths, and rules.
NMAKE Directives
omake supports the following NMAKE directives:
!CMDSWITCHES {+|– } opt
This read-time directive turns on (+) or off (–) one or more options, opt.
!message message
This read-time directive is the same as the %echo message directive.
NMAKE Emulation
omake is highly compatible with NMAKE, but there are differences in how they read makefiles
and in how they run. omake's NMAKE emulation mode causes omake to operate like NMAKE.
Emulation at Startup Time
If NMAKE emulation mode is selected at startup time, omake emulates the NMAKE command
line and selection of the initialization file. Selection of NMAKE emulation at startup time is done
with the –EN flag, either on the command line or in the OMAKEOPTS environment variable.
To determine the startup emulation mode, omake first examines the OMAKEOPTS environment
variable for –Ex flags, and then the command-line for –Ex flags. If the last –Ex flag is –EN, omake
starts up emulating NMAKE.
NOTE: If you want submakes (recursive invocations of omake) to inherit NMAKE emulation
mode, you must specify the –EN flag in the OMAKEOPTS environment variable.

Emulation After Start-Up Time
The .MS_NMAKE directive turns on NMAKE emulation mode from the point it appears in the
initialization file or any makefile.
The Command Line
The OMAKEOPTS environment variable is parsed for options, and then the command line is
parsed. Parsing entails the following:
➤ Handling the case-insensitivity of the NMAKE command line.
➤ Handling all NMAKE command-line options.
➤ Switching the help screen to the options available with NMAKE v1.3. The help screen's
contents are stored in the file omhelp.nm.
Without emulation at startup, the command line is as documented in the section Command-Line
Options on page 41.
The Emulation File (BUILTINS.NM)
If omake is emulating NMAKE at startup, omake reads its internal rules and macros, and then
looks for builtins.nm in the directory of make.ini, in the directory of the omake.exe file, and
along directories of the INIT environment variable, in that order. If builtins.nm is found, it is read
for macros and rules that give more complete NMAKE emulation.
The Initialization File (TOOLS.INI)
If omake is emulating NMAKE at startup, omake reads the tools.ini file found in one of these
directories:
➤ In the current directory
➤ In a directory named by the INIT environment variable
If omake finds the initialization file, it reads information in the file starting with the section
heading [NMAKE.
Makefile Contents
If omake is emulating NMAKE:

➤ The ^ character is used for quoting and produces the following effects:
Character Effect
^^ literal ^
^<ENTER> literal <ENTER>
^$ literal $
^\ literal \
^# literal #
➤ The !else if, !else ifdef, and !else ifndef directives are supported.
➤ Macro definitions of the form
VAR = ... $(VAR) ...
are supported and cause $(VAR) to be expanded before VAR is redefined. Other macro
references in the macro value are not expanded.
➤ Macro definitions of the form
ENVVAR = value
where ENVVAR is an environment variable, assign to the ENVVAR macro the value, which is
also exported to the environment. If the –E command-line flag is used, makefile macros
cannot override environment macros, and this macro redefinition is ignored.
➤ Search paths for dependents are supported. These look like this:
forward.exe : {\src\alpha;d:\proj}pass.obj ...
➤ Inference rule search paths are supported. These look like this:
{fromdir}.fromext{todir}.toext : (build script for inference rule)
➤ Inference rules for targets that have a directory component look for the inferred source in
the directory of the target.
➤ The !include file directive is supported. If file is an absolute pathname, it is used; otherwise,
omake looks for file in the current directory, and then in the directory of the including file. If
file appears inside angle brackets (for example, <file>), omake then looks along the
directories in the INCLUDE environment variable.
Macros
If omake is emulating NMAKE, the following macro features are supported:

➤ Command-line macros are exported to the environment, and become available to recursive
makes.
➤ Environment variables override built-in macro definitions (see the section Macros on
page 52).
➤ The MAKEFLAGS macro is exported to the environment.
➤ Non-run-time macros are expanded at parse time.
Most other make programs expand macros when a build script is run, but NMAKE expands
macros when it parses the makefile. When the rule is parsed, it is defined with the expanded
value of the macro, rather than with a reference to the macro. (The reference would be
expanded at run time).
EXCEPTION: NMAKE evaluates run-time macros at run time, because the values to which
they are set are based on the target.
Therefore, the following makefile is evaluated differently by NMAKE and clearmake:
FOO=A
rule:
@echo $(FOO)
FOO=B
In NMAKE, $(FOO) is evaluated at parse time, so building rule echoes A. In clearmake,

$(FOO) is evaluated at run-time and its value after parsing is B, so building rule echoes B.
Build Scripts
If omake is emulating NMAKE, the following build-script features are supported:
➤ The shell program is used for executing every build script.
➤ The build scripts cd dir and chdir dir both change to directory dir and stay there until
changed back explicitly.
➤ A set var=val build script is treated specially and sets the environment variable VAR to the
value val. That is, this build script is treated exactly like the omake directive %setenv
var=val.
➤ The filename-parts syntax is supported. You can use this to get at the components of the
name of the first source file. The complete file name is represented with the %s syntax. Parts
of the file name are represented with this syntax: %| [parts] F , where parts is zero or more of

the following letters: none, the complete name; d, drive; p, path; f, file base name; e, file
extension.
➤ Shell-line prefixes are limited to @, -, and !. The ^ prefix stops processing of prefixes.
➤ The # character is not treated as a comment in build scripts.
omake supports the NMAKE response file syntax, except that multiple inline response files on a
single build script are not support.
NMAKE and omake allow the keywords KEEP or NOKEEP in the epilog of the inline response
file. KEEP causes the response file to be kept (that is, not deleted). The default is NOKEEP.
Unsupported NMAKE Features
Multiple inline response files on a single build script are not supported.
Arithmetic operators are not supported in preprocessing expressions.
D.3 Opus Make v5.2x Compatibility and Emulation

omake is not entirely backward compatible with Opus Make 5.2x.
The –E2 command-line flag makes omake emulate Opus Make 5.2x.
omake has the following compatibility and emulation features:

comment v5.2x treated line continuation before comment detection, so a comment
character (#) in a continued line causes the rest of the continued line to be
ignored. omake treats # as a comment only until the end of the current physical
line.
do In v5.2x, macro definitions on the %do line were separated by a comma (,). In
omake, they are separated by white space and must be enclosed in double
quotes if they contain white space.

infer v5.2x searched for the inferred source unless explicitly told not to with the
.NOINFER attribute. omake searches for the inferred source only for targets
without build scripts but the .INFER attribute can force it to search.
noiterate The shell-line prefix ! means iterate this build script. Note that ! can also
indicate a directive; the rule is that it indicates a directive if it can; otherwise, it
is a prefix.
! echo $(.SOURCES) Does not iterate: !echo is a directive
!| echo $(.SOURCES) Iterates: | indicates the end of prefixes
! add $? to $@ Iterates: !add isn't a directive
@! echo $(.SOURCES) Does not iterate: !echo is a directive
!@ echo $(.SOURCES) Iterates: !@ is not a directive
The noiterate feature turns off any interpretation of ! as a shell-line prefix.

twopass Opus Make v5.2x used two passes to macro expand build scripts. The first pass
expanded all macros. The second pass tokenized the line and replaced any
found targets with the pathname to the target. omake does not do the second
pass unless you choose this feature. omake has a macro modifier, X, that allows
selective expansion of names into pathnames.
D.4 Borland Make Compatibility

The following sections describe omake’s compatibility with Borland Make.
➤ Automatic dependencies
omake knows how to read the dependency information stored in object files created from the
Borland bcc compiler.
➤ Special targets
.SWAP : .SWAP should only be found in makefiles produced by the Borland
prj2mak program.
➤ Inline response files
omake accepts the inline response file syntax:

target :
command [ prolog ]&&X
build script
.
.
.
X [ epilog ]
where prolog is usually @ and X is a single character (Borland uses ! in its examples). The
build scripts are placed in a response file and the command is invoked with the following
command, where tempfile is a unique filename:
command prolog tempfile epilog
➤ Borland Make emulation
There is no specific Borland Make emulation mode.
D.5 UNIX Make Compatibility

omake supports makefiles produced for SunOS (Solaris) Make. The following features are
supported:
➤ makefile syntax
The makefile expression
include filename
where include is in the leftmost column of the makefile includes filename at this point in the
makefile. The makefile expression
$(name:str1=str2)
is a macro modification with string substitution, with strings str1 and str2 both capable of
using the % character as a wildcard to match zero or more characters in the expansion of
name.
➤ Inference rules
omake supports the suffix rules and metarules of UNIX Make. Suffix rules are converted into
omake inference rules. The .SUFFIXES directive is also supported.

➤ UNIX make emulation
There is no specific UNIX make emulation mode.

E Regular Expressions
E
Regular expression matching is a scheme for string searching. A regular expression is a string of
characters, some normal, some special, that allows the specification of a pattern to be matched.
E.1 Configuration of Regular Expressions

The following directives control the special characters that appear in regular expressions:
Directive Use Default Value
.REGEX_CHAR Sets the escape character used to indicate \ (backslash)
special sequences
.REGEX_WILD Sets the wildcard character that matches any . (period)
single character
Regular Expression Components
Table 29 lists the forms, or components, of a regular expression.
E - Regular Expressions 167

Table 29 Regular Expression Components
Label Form Form Description
[1] character A normal character matches itself. Special characters: wild \ [ ] * +

^$
[2] wild The wildcard character, wild, matches any character.
[3] \ The escape character makes special characters literal: \wild is

literal wild, \\ is literal \, \[ is literal [, and so on. The only
exceptions are that $ and $ are special. See [7] below.
[4] [ set ] Matches one character in set. If the first character in set is ^, this
form matches characters not in set. A range start–end means
characters from start to end. The characters ] and – aren't special if
they appear as the first characters in set. \t matches a tab character.
For example:
Set Matches
[a-zA-Z@] Lowercase and uppercase alphabetic, or @
[^]–] Neither ] nor –
[^A–Z] Not uppercase alphabetic
[<SPACE> \t] A space or tab (that is, white space)
[5] form* Any regular expression form labeled [1] to [4] followed by the
closure character * matches zero or more of the form.
[6] form+ + is like *, except it matches one or more of the form.
[7] $ form $ A regular expression in the form [1] to [10], enclosed as $ form $
matches what form matches. The substring matched by form can be
referenced with a tag (see below).
[8] \1 ... \9 Matches a previously tagged regular expression (see [7]).
[9] form1form2 A composite regular expression form1form2, where form1 and form2
are in the form [1] to [9], matches the longest match of form1
followed by a match of form2.
[10] ^ and $ A regex starting with ^ and/or ending with $ restricts the regex to
the beginning of the line and/or the end of line. Elsewhere in the
regex, ^ and $ are ordinary characters.

Referencing the Matched Expression
Once a regular expression has matched, you can reference the matched part:
➤ & refers to the entire matched string (but \& is a literal ampersand).
➤ If  is used to delimit a part of the regular expression, the tag \1 refers to the first
delimited part of the matched substring. Successive pairs of  are tagged \2, \3, ..., \9.
E.2 Macro Modifiers in OMAKE

The M, N, and S modifiers use regular expressions to match macro elements. The regular
expression is matched against each macro element individually. For the following examples,
assume the following macro definitions:
SRCS = main.c sub.cpp io.cpp
CFLAGS = –AX –Ifoo –Ibar /Ibaz –DX=–IT xI.c yi.c
Regular Expressions for the M Modifier
To select files whose names include .c:
$(SRCS,M.c) is main.c sub.cpp
To select files that end in .c, anchor the search to the end with the regular expression character $.
To get $ to the regular expression, use $$ in the makefile:
$(SRCS,M.c$$) is main.c
The ^ regular expression character anchors the search to the front of the macro element:

The [set] regular expression characters indicate a set of characters, where set can be single
characters ([aA@] matches a, A, or @), a range of characters ([a-z] matches a through z), or a
mixture. For example:
E - Regular Expressions 169

Regular Expressions for the S Modifier
One powerful feature of regular expressions is that when they are used in substitutions, they can
access the matched parts of the string. The S/rfrom/rto/ (substitution) modifier uses regular
expression rfrom to substitute the matched part of an element with the rto regular expression. For
example, when DIR = NT_L, the expression $(DIR,S/$wild*$_wild*/\1/) is NT
The  pair surround part of the regular expression that can be referenced later. Inside the pair
is wild*, which matches any character repeated zero or more times. Taken together, they mean
instruct omake to match any character, zero or more times, and tag it. The rest of the regular
expression is _, which matches _ and wild*, which matches any character repeated zero or more
times.
The substitution replaces the matched part of the element with the expression \1, which is the
stuff matched in the first pair of . The entire element was matched, so the substitution
produces NT. Table 30 shows other expressions.
Table 30 Examples of Regular Expressions
Expression Result
$(DIR,S/wild*_$wild*$/\1/) L
$(DIR,S/wild*_$wild*$/&/) NT_L
$(DIR,S/$wild*$_ /\1/) NTL
$(DIR,S/_ //) NTL
$(DIR,S/does not match/xyzzy/) NT_L

Glossary
ATTRIBUTES. Properties that can be associated with targets. They include .MAKE, .REREAD,
.SILENT and others.
BASE NAME. A file name without any directory components or extension.
BUILD. To bring a target up to date (make it current).
BUILD AVOIDANCE. The ability to fulfill a build request by using an existing derived object, instead
of creating a new derived object by executing a build script.
BUILD SCRIPT. A command executed by omake or passed to the shell program for execution. Build
scripts are used to update targets.
COMSPEC. This environment variable names the program that is the operating system
command processor or shell program. For Windows NT, this is usually cmd.exe.
CONFIGURATION LOOKUP. The process by which omake determines whether to produce a derived
object by performing a target rebuild (executing a build script) or by reusing an existing
instance of the derived object. This involves comparing the configuration records of existing
derived objects with the build configuration of the current view: its set of source versions, the
current build script that would be executed, and the current build options.
DEFAULT TARGET. The first normal target defined in the first makefile. A normal target is not a
directive, special target, attribute, or inference rule. It is the target that is made when omake
is run without any command-line targets.
DEPENDENCY. Anything a target depends on. Dependencies are themselves targets because they
can be made.
DEPENDENCY LINE. The line in a build script that indicates the dependence of a target on its
dependencies. For example,
test.exe : main.obj sub.obj
declares that test.exe depends on main.obj and sub.obj.
DIRECTIVE.The instructions that control how omake proceeds. There are run-time directives and
read-time directives, and some work both at read and run time. Directives are indicated by
Glossary 171
the appearance of either % or ! as the first non-white-space character on a makefile line,
followed by the directive name. If the % (or !) is in the leftmost column of the makefile, the
directive is interpreted at read time; otherwise, it is interpreted at run time.
DIRECTORY SEPARATOR. The character in a pathname that separates directory names and the file
name. For Windows NT, the separator is either \ or /. omake uses either character as the
directory separator.
DOT DIRECTIVES. A read-time directive that modifies the operation of omake from the point at
which it is encountered in the makefile.
EXIT STATUS. A number returned by an executed program and testable by omake. At the
command line, you can check the exit status of the last executed program by using the if
errorlevel command.
By convention, programs return a zero exit status when they finish without error and nonzero
when an error occurs.
EXPLICIT RULE. A dependency line and build scripts that are used to make a target, in this form:
target : dependencies
build script
.
.
.
EXPLICIT DEPENDENCY.A dependency declared in a dependency line. For example, def.h is an
explicit dependency of main.obj in the following:
main.obj : def.h
EXTENSION. The suffix part of the file name, usually used to denote the type of file. The extension
consists of the characters that follow the last period in the file name to the end of the name.
FILE NAME. The part of the pathname after the last directory separator.
HEADER FILE. A file, containing source code, that is included into the body of a source file.
INCLUDE FILE. See header file.
INFERENCE RULE. A rule that generalizes the build process so that you do not have to specify how
to build each target. For example, here is the built-in inference rule for making an .obj file
from a .c file:
%.obj : %.c
$(CC) $(CFLAGS) –c $.SOURCE
INFERRED DEPENDENCY. The dependency determined with an inference rule. For example, if
main.obj is being built and the inference rule %.obj : %.c is used, main.c is an inferred
dependency of main.obj.
INIT. An environment variable whose value is a semicolon-separated list of directory names. For
example:
INIT = c:\home;c:\msc

KEEP WORKING. An operating mode of omake. While updating a target, if an executed build script
returns a nonzero exit status, omake stops updating this target immediately. Any other
targets that depend on this target as a source are not updated. This mode maximizes the
amount of safe making and is ideal for running unattended builds (for example, rebuilding a
large project overnight).
MACRO. The association of a name and a value. The macro expansion of macro name returns the
value. Macros are used at read time as a means of organizing names of files, compiler options,
and so on. At run time, macros also allow you to refer to the current target being built.
MAKEFILE. A file from which omake reads its instructions. An initialization file is read first,
followed by one or more makefiles. The initialization file holds instructions for all make
programs and is used to customize the operation of omake. The makefile has instructions for
a specific project.
MODULE. A single file, such as a source or object file, that is combined with other files to build a
project, such as an executable.
NULL STRING. A string with no characters.
PATH. The environment variable used to indicate the order in which directories are searched for
executable files. For example, with a PATH of c:\bin;c:\utils, the program searches for
executables in the current directory, in directory c:\bin, and in directory c:\utils in that order.
PATHNAME. The location of a target or file on disk, including any directory components.
READ TIME. The phase of omake in which it reads the makefiles.
READ-TIME DIRECTIVE. A directive that appears in the makefile with the directive character % or !
in the leftmost column. Read-time directives are interpreted while the makefile is being read.
REGULAR EXPRESSION. A string of characters (some normal, some special) that allows the
specification of a pattern to be matched.
RECURSION, RECURSIVE MAKE. The act of calling a program (or function) from itself. The usual
context is in the expression “call omake recursively.” This means a running copy of omake
uses a build script to call a second copy of omake. Recursive makes are often used for projects
that are split into multiple directories.
RESPONSE FILE. A text file used to hold long command lines. For many programs, the
command-line option @file indicates that the program is to read its command line from the file
response file.
ROOT NAME. For a target, its pathname minus its file name extension. This pathname includes any
directory components.
RUN TIME. The phase of omake when it builds targets.
RUN-TIME DIRECTIVE. A directive that appears as a target's build script (the directive character % or
! is indented from the leftmost column). Run-time directives are interpreted when the target's
build scripts are executed.
SHELL LINE. See build script.
SOURCE. See dependency.
Glossary 173
SOURCE FILE. A file containing source code that can be compiled into object code.
SPECIAL TARGET. A target whose name is of the form .NAME and that has special meaning to
omake.
STANDARD ERROR/OUTPUT. The two output streams (or file descriptors) that write to the console.
Standard output is usually used to output general messages. Standard error is usually used
to output error messages.
TARGET. Something that can be made. A target is usually a file, such as a source, object, or
executable. The target is said to exist if the file is present on disk.
TARGET ATTRIBUTE. Properties assigned to targets.
TIME STAMP. The time stamp of a target is the time and date that a target was last changed. It is
usually the creation or modification time of the file, as stored by the operating system.
UPDATE (A TARGET) . To execute a target’s build scripts. This brings the target up to date.
UP TO DATE TARGET. A target that is current; that is, it has been compiled, linked, and so on, and is
newer than all its dependencies.
WHITE SPACE. One or more space or tab characters.
WILDCARD CHARACTER. Characters used to match ambiguous part of a file name. Windows NT
treats ? as a wildcard character that matches any single character and * as a wildcard character
that matches any number of characters in a file name or file name extension.

Index
– – shell-line prefix 70 %ifndef directive 75, 77

– shell-line prefix 70, 104 %include directive 76, 89, 95
! directive character 163, 172 %length() operator 80
! logical operator (not) 82 %make() operator 80
! shell-line prefix 71, 85, 163 %member() operator 80
!= comparison operator 79 %null() operator 80
!CMDSWITCHES directive 91, 158 %restart directive 76, 89
!message directive 91, 158 %set directive 76, 90
"" double quotes 25 %setenv directive 76, 90
# comment character 12 %status function 91, 148
$ macro character 21 %time() operator 80, 86
$$@ predefined macro 62 %undef directive 24, 76, 91
$[any transformation macros] 147 %while directive 75, 84–86
% directive character 172 %writable() operator 80
% rule character 30, 36 & shell-line prefix 71, 104
%abort directive 39, 76, 86 && logical operator (and) 82
%break directive 75, 85–86 '' double quotes 63, 79, 87, 162
%chdir directive 76, 87 ( ) logical operator (order evaluation) 82
%continue directive 75, 85 * shell-line prefix 156
%defined() operator 80 ** run-time macro 61, 71, 85
%dir() operator 80 + shell-line prefix 71, 99
%do directive 35, 76, 87 + target-group character 32, 51
%echo directive 76, 88, 158 += appended macro definition 24–25
%elif directive 75, 77 .AUTO keyword 99
%else directive 75, 77 .NOMULTI keyword 99
%elseif directive 75, 77 .NOREDIR keyword 99
%end directive 75, 83, 155 .RULE_CHAR directive 36
%endfor directive 83 : drive separator 16
%endif directive 75, 77 : shell-line prefix 71, 99, 156
%error directive 76, 86, 88 := expanded macro definition 24–25, 54
%exec directive 76, 88 ; multiple-command separator 73
%exists() operator 80 < comparison operator 79
%exit directive 91, 148 < run-time macro 61
%file() operator 80 <= comparison operator 79
%foreach directive 75, 82 = = comparison operator 79
%if directive 75, 77 = standard macro definition 23
%ifdef directive 75, 77 > comparison operator 79
Index 175
/vobs/doc/ccase/omake/cc_omakeIX.fm — September 10, 1999 12:08 pm

> shell-line prefix 72 C
>= comparison operator 79
? run-time macro 60, 71, 85 case sensitivity 24, 58, 79
?= conditional macro definition 23 .CASE_MACRO directive 24, 64, 92
@ response file prefix 113, 115 .CASE_MACRO state macro 64
@ run-time macro 61 .CASE_TARGET directive 49, 64, 79, 92
@ shell-line prefix 69, 87, 105 .CASE_TARGET state macro 64
@@ shell-line prefix 69 CC built-in macro 67
[ command ] command-execution operator 81 ccase-home-dir directory xvii
\& literal "&" (in regex) 169 .CHAIN attribute 103
\; literal ";" (in multiple-command shell lines) 73 chained targets 33
\= literal "—" (in string substitution macro modifier) 57 changing directory 72, 74, 87
\ENTER line continuation 12, 154 command line
BUILD environment variable 153
^ run-time macro 61 MAKEOPTS environment variable 153
| shell-line prefix 72, 163 MFLAGS environment variable 47
|| logical operator (or) 82 OMAKEOPTS environment variable 47, 153
~ ~ shell-line prefix 71 command line flags
~ shell-line prefix 71 –# debug options 64, 93
–#1 read-time debug 119, 121
–#2 undefined macro 120
A –#4 unknown line 120
–#8 leave temporaries 116
–a all targets 64, 103
.AFTER special target 48, 106
–b file 42, 46
.ALWAYS attribute 64, 103 –D keep directory 65, 96, 110
(Always) shell-line prefix 152 –d run-time debug 64, 93, 119
.ALWAYS state macro 64 –e environment override 65, 95
attache-home-dir directory xvii –E2 emulate OMAKE v5.2x 65
–EN emulate NMAKE 65, 97, 158
attributes 31, 102 –EO emulate OMAKE 65, 98
.AUTO keyword 99 –EP emulate PolyMake 65, 98, 153
auto-detection mode 19 –f file 12, 43
automatic response files 110 –h help 43
–i ignore errors 20, 65, 70, 104
–k keep working 20, 39, 65, 96
B –M make makefile 65, 97
–n no execute, overriding 67, 104
–p print debug info 64, 93, 119, 122
base name 53, 171
–q query 65
.BEFORE special target 48, 90, 106 –q query, overriding 104
Borland Make 37, 163 –r reject rules 66, 98, 142
BUILD environment variable 153 –s silent 66, 105
build scripts 10, 18 –x redirect error messages 45
executing 19 –z ignore MFLAGS 47, 65
inheriting 31–32 comments 12–13, 162
prefixes 20 compatibility 145
builds 10, 171 Borland Make 163
built-in macros 60, 140 Configuration Builder 145
NMAKE 97, 158
BUILTINS environment variable 154 OMAKE v5.2x 162
BUILTINS predefined macro 62 PM/CB 145
PolyMake 145
UNIX Make 164

COMSPEC environment variable 20, 171 E
conditional directives 23, 76–77
conditional expressions 78 –e file-test operator 81
configuration 2 ECHO inline response file keyword 115
continued lines 12–13, 77 emulation 145
Configuration Builder 152
conventions, typographical xvii
NMAKE 2, 158
CWD predefined macro 63 OMAKE v5.2x 162
PM/CB 2, 152–153
PolyMake 152
D .ENV_OVERRIDE directive 65, 95
.ENV_OVERRIDE state macro 65
–d file-test operator 81
environment variables
.DEBUG directive 64, 93 accessing 28
.DEBUG state macro 64, 66 BUILD 153
.DEBUG_GRAPHICS directive 93 BUILTINS 154
.DEBUG_PRINT directive 64, 93 COMSPEC 20
giving higher precedence 95
.DEBUG_PRINT state macro 64
INIT 89
.DEBUG_RUN directive 64, 93 MAKEOPTS 153
.DEBUG_RUN state macro 64 OMAKECFG 46
.DEFAULT attribute 103 OMAKEOPTS 47, 153
.DEFAULT special target 106 setting 90
default targets 47, 50 .ENVMACROS directive 64, 95
deinitialization 20, 48, 106, 127 .ENVMACROS state macro 64
dependencies 10, 14 exit status 10, 20, 135, 172
explicit 14 MAKESTATUS macro 63, 135
inferred 48 non-zero 20
dependency lines 14 explicit dependencies 14
double colon 51 explicit rules 14, 18
wildcards 16 explicit sources 48
.DEPENDENCY_IGNORED_FOR_REUSE directive 94 extensions 172
directives 38 (ExtraLine) shell-line prefix 152
! character 163, 172
% character 172
.NOCMP_SCRIPT 98 F
conditional 75–76
dot 91 –f file-test operator 81
iteration 75, 85
FC built-in macro 67
other 76
read-time 75, 173 file names 53, 172
run-time 75, 173 FIRSTTARGET predefined macro 63
target 91
directory separator 172
.DO_FOR_SIBLING directive 94 H
documentation
online help description xviii header files 172
double-colon dependency lines 51 help 43
dummy target 50
Index 177

I L
I/O redirection 45 LIBEXE built-in macro 67
.IGNORE attribute 65, 70, 104 line continuation 12, 154
ignore mode 70 and comments 13
and conditional directives 77
(Ignore) shell-line prefix 152
LINK built-in macro 67
.IGNORE state macro 65
local input scripts 115, 117
.IGNORE_MFLAGS state macro 65
.INCLUDE directive 95, 149
include file 89, 95, 172 M
.INCREMENTAL_REPOSITORY_SIBLING directive 95
.INFER attribute 32, 104, 163 macro modifiers 26, 52, 141
inference rules 14, 29, 48 *D wild-card expand directory 57
attributes 36 *F wild-card expand file 57
built-in 36, 142 < prepend string 54
debugging 123 = string substitution 27, 56
defining 30 > append string 54
disabling 36, 98 @ include 54
disabling search 36 A absolute pathname 53
multiple-step 33–34 B base name 53, 107
redefining 35 D directory 27, 53
rejecting 36, 98 E extension 27, 53
search order 34 F file name 27, 53
suffix-only form 31 LC lowercase 54
target groups 32, 56 M member 55, 58, 169
using 32 N non-member 55, 58, 169
num numbered element 33, 56
inferred dependency 30 R root name 25, 53, 62
INIT environment variable 89, 172 S regex substitution 56, 59, 170
initialization 48, 106 UC uppercase 54
initialization file 11, 42, 46 W tokenize 27, 57, 116
X expand to pathname 54, 57, 163
inline response files 38, 110, 115, 162–163
Z drive 53
INPUTFILE predefined macro 63
.MACRO_CHAR directive 96
(Iterate) shell-line prefix 152
macros 21, 52
iteration 71, 85, 156, 163 built-in 60, 140
character 21, 96
definition
K += appended 24–25
:= expanded 24–25, 54
KEEP inline response file keyword 115 = standard 23, 25
.KEEPDIR directive 65, 96 ?= conditional 23
accessing environment variables 28
.KEEPDIR state macro 65
at read time 22
keep-directory mode 96 in the makefile 22
.KEEPWORKING directive 39, 65, 96, 128 indenting 24
keep-working mode 39, 96, 173 location 24
.KEEPWORKING state macro 65 on the command line 25
precedence 22
expansion 21, 28–29
precedence 22
predefined 60, 137
reference 21
run-time 26, 60

state 63, 138 .NOINFER attribute 36, 104, 163
transformation 147 NOKEEP inline response file keyword 115
undefining 24 .NOKEEPDIR directive 96
make 10 .NOKEEPWORKING directive 96
.MAKE attribute 71, 104 .NOMULTI keyword 99
MAKE built-in macro 67, 104 .NOREDIR keyword 99
.MAKE_MAKEFILE directive 65, 97 .NORULE attribute 36, 105
.MAKE_MAKEFILE state macro 65 .NOSHELL directive 71, 99
MAKE_TMP built-in macro 67, 116, 128 (NoShell) shell-line prefix 152
MAKEARGS predefined macro 63 .NOUNIXPATHS directive 101
MAKEDIR predefined macro 63, 87 .NOWINK_IN directive 101
MAKEFILE macro 43
makefiles 9, 12
debugging 119 O
line continuation 12
line length 12 .OMAKE directive 65, 98
making before reading 97 .OMAKE state macro 65
MAKEMACROS predefined macro 63 OMAKE v5.2x
MAKEOPTS environment variable 153 emulation 162
MAKESTATUS predefined macro 63, 135 incompatibilities 162
MAKETARGETS predefined macro 63 OMAKECFG environment variable 46
MAKEVERSION predefined macro 63 online help, accessing xviii
making the makefile 97 OPUS predefined macro 63
meta rules 30 .OPUS_52X state macro 65
MFLAGS built-in macro 68 OS built-in macro 68
.MS_NMAKE directive 65, 97, 159
.MS_NMAKE state macro 65
P
multiple-command shell lines
; command separator 73
\; literal ";" 73 .PATH macro 54, 107
path separator 50
.PATH.obj macro 54, 108
N pathnames 9, 49, 173
PM/CB 2, 37
.NEWSOURCES run-time macro 60, 71, 85 command-line flags 157
NMAKE 37 compatibility 72, 83, 91, 98, 106, 110, 145
compatibility 62, 91, 158 directives 148
directives 158 emulation 2, 68, 153
emulation 2, 29, 68, 158–159 iteration groups 156
.NOCASE_MACRO directive 24, 92 macros 145
operation-line modifiers 157
.NOCASE_TARGET directive 49, 79, 92
reserved targets 151
.NOCHAIN attribute 34, 103 suffix dependencies 157
.NOCMP_NON_MF_DEPS attribute 104 transformation macros 147
.NOCMP_SCRIPT directive 98 unimplemented features 156
.NOCONFIG_REC attribute 105 .POLY_MAKE directive 65, 98, 153
.NODEBUG directive 66, 93 .POLY_MAKE state macro 65
.NODEFAULT attribute 103 PolyMake 145
.NOENV_OVERRIDE directive 95
.NOENVMACROS directive 28, 95
Index 179

.PRECIOUS attribute 34, 105, 132 S
predefined macros 60, 137
PVCS 145, 149 search directories 54, 106
.PATH macro 54, 107
debugging 110, 123
Q location of 107
run-time macros 108
target names 49
.QUERY state macro 65
VPATH macro 108
setting environment variables 90
R .SHELL directive 66, 71, 99, 150
shell lines
–r file-test operator 81 multiple-command 72
RC built-in macro 68 prefixes
– – really ignore 70
read-time 9, 47, 173 – ignore 70, 104
read-time directives 75, 173 ! iterate 71, 85, 163
recursive macro 29, 131 & override –n 71, 104
recursive make 67, 104, 173 (Always) 152
(ExtraLine) 152
redirection of I/O 45 (Ignore) 152
.REGEX_BACK state macro 65 (Iterate) 152
.REGEX_CHAR directive 59, 65, 98 (NoShell) 152
.REGEX_CHAR state macro 65 (Shell) 152
(Silent) 152
.REGEX_DOT state macro 65
(TrackErrors) 152
.REGEX_WILD directive 59, 65, 98 + use shell 71, 99
.REGEX_WILD state macro 65 : suppress shell 71, 99
regular expressions 52, 55–56, 58, 167 > extra line 72
configuration 58, 167 @ silent 69, 87, 105
for the "M" modifier 58, 169 @@ really silent 69
for the "S" modifier 59, 170 | end of prefixes 72, 163
referencing matched expression 169 ~ ~ really ignore, keep status 71
.REJECT_RULES directive 36, 66, 98, 142 ~ ignore, keep status 71
PM/CB compatible 152
.REJECT_RULES state macro 66
(Shell) shell-line prefix 152
.REREAD attribute 66, 109
.SHELL state macro 66
.REREAD state macro 66
SHELLCOMMANDS built-in macro 19, 68
.RESPONSE directive 98, 111
SHELLSUFFIX built-in macro 19, 68
response files 37, 110
automatic 110, 114–115, 123 .SIBLINGS_AFFECT_REUSE directive 100
inline 36, 38, 110, 115, 162–163 .SILENT attribute 66, 69, 105
root name 53, 173 silent mode 69
.RULE attribute 105 (Silent) shell-line prefix 152
rule character 36, 99, 105 .SILENT state macro 66
.RULE_CHAR directive 66, 99 .SOURCE directive 110, 150
.RULE_CHAR state macro 66 .SOURCE run-time macro 26, 61
rules 13 sources
explicit 14 explicit 48
suffix 37 .SOURCES run-time macro 26, 61, 71, 85
run-time 9, 47, 173 special targets 48, 105
directives 75, 173 state macros 63, 138
status predefined macro 63, 85, 88

suffix rules 37, 124
.SUFFIXES directive 37, 66, 100, 150
.SUFFIXES state macro 66
T
target groups 51
inference rules 32, 56
.TARGET run-time macro 26, 61
.TARGETROOT run-time macro 62
targets 9
default 47, 50
dummy 50
existence 9
targets attributes 101
technical support xviii
timestamps 9, 174
(TrackErrors) shell-line prefix 152
transformation macros 147
typographical conventions xvii
U
.UNIXPATHS directive 35, 66, 101
.UNIXPATHS state macro 66
updating targets 48, 174
up-to-date targets 10, 171, 174
V
VPATH macro 108
W
–w file-test operator 81
whitespace 174
wildcards 16, 57, 174
.WINK_IN directive 101
Z
–z file-test operator 81
Index 181


CC Omake

Uploaded by

Copyright:

Available Formats

CC Omake

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CC Omake

Uploaded by

Copyright:

Available Formats

CLEARCASE OMAKE MANUAL

Release 4.0 and later

/vobs/doc/ccase/omake/cc_omakeTTL.fm — October 8, 1999 5:44 pm

/vobs/doc/ccase/omake/cc_omakeTOC.fm — September 10, 1999 12:08 pm

iv ClearCase OMAKE Manual

/vobs/doc/ccase/omake/cc_omakeTOC.fm — September 10, 1999 12:08 pm

/vobs/doc/ccase/omake/cc_omakeTOC.fm — September 10, 1999 12:08 pm

vi ClearCase OMAKE Manual

/vobs/doc/ccase/omake/cc_omakeTOC.fm — September 10, 1999 12:08 pm

/vobs/doc/ccase/omake/cc_omakeTOC.fm — September 10, 1999 12:08 pm

viii ClearCase OMAKE Manual

/vobs/doc/ccase/omake/cc_omakeTOC.fm — September 10, 1999 12:08 pm

4. Debugging Makefiles ...................................................................................................... 119

/vobs/doc/ccase/omake/cc_omakeTOC.fm — September 10, 1999 12:08 pm

A. Errors and Warnings.......................................................................................................127

B. Exit Status Values ...........................................................................................................135

C. Built-In Macros and Rules .............................................................................................137

D. Compatibility and Emulation ........................................................................................145

x ClearCase OMAKE Manual

/vobs/doc/ccase/omake/cc_omakeTOC.fm — September 10, 1999 12:08 pm

/vobs/doc/ccase/omake/cc_omakeTOC.fm — September 10, 1999 12:08 pm

xii ClearCase OMAKE Manual

/vobs/doc/ccase/omake/cc_omakeTOC.fm — September 10, 1999 12:08 pm

Table 1 Macro Modifiers................................................................................................... 27

/vobs/doc/ccase/omake/cc_omakeLOT.fm — September 10, 1999 11:43 am

/vobs/doc/ccase/omake/cc_omakeLOT.fm — September 10, 1999 11:43 am

About This Manual

➤ Read Chapter 1, Introduction, for configuration and usage information

➤ Read Chapter 2, Overview of Using omake

xvi ClearCase OMAKE Manual

This manual uses the following typographical conventions:

➤ [] Brackets enclose optional items in format and syntax descriptions.

➤ | A vertical bar separates items in a list of choices.

The ClearCase graphical interface includes a standard Windows help system.

> cleartool man (display the cleartool overview page)

> cleartool lsprivate –help

xviii ClearCase OMAKE Manual

North America 800-433-5444 408-863-4194 support@rational.com

Europe, Middle +31-(0)20-4546-200 +31-(0)20-4546-201 support@europe.rational.com

Asia Pacific 61-2-9419-0111 61-2-9419-0123 support@apac.rational.com

1.1 File Manifest

1.2 Configuring omake

1.3 Notes on Using omake

➤ omake reference page (in the ClearCase Reference Manual)

omake’s Configuration Management Features

2 ClearCase OMAKE Manual

Derived Object Sharing

Configuration Record Creation

➤ The following options have been added to omake:

omake supports local and remote non-MVFS dependencies:

Differences Between omake and Standard Make Tools

4 ClearCase OMAKE Manual

c:\> net use f: \\view\myview

2. Change to the appropriate directory and run omake:

f:\myvob\src> omake options

Supporting Both omake and clearmake

Parallel and Distributed Build Operations

Build Scripts and the rm Command

6 ClearCase OMAKE Manual