meson: docs: Add xml{lint,proc} wrapper to collect dependencies

meson/ninja do not support specifying dependencies via globs (as those make it
significantly more expensive to check the current build state). Instead
targets should emit dependency information when running that then can be
cheaply re-checked during future builds.

To handle xmllint and xsltproc invocations in the docs, add and use a wrapper
that uses --load-trace to collect dependency information.

Reviewed-by: Peter Eisentraut <peter.eisentraut@enterprisedb.com>
Author: Nazir Bilal Yavuz <byavuz81@gmail.com>
Author: Andres Freund <andres@anarazel.de>
Discussion: https://postgr.es/m/c5736f70-bb6d-8d25-e35c-e3d886e4e905@enterprisedb.com

Author: anarazel

doc/src/sgml/meson.build

@@ -2,7 +2,7 @@ docs = []
 alldocs = []
 doc_generated = []
 
-xmllint = find_program(get_option('XMLLINT'), native: true, required: false)
+xmllint_bin = find_program(get_option('XMLLINT'), native: true, required: false)
 
 
 version_sgml = configure_file(
@@ -60,21 +60,31 @@ doc_generated += custom_target('keywords-table.sgml',
 )
 
 # For everything else we need at least xmllint
-if not xmllint.found()
+if not xmllint_bin.found()
   subdir_done()
 endif
 
 pandoc = find_program('pandoc', native: true, required: false)
-xsltproc = find_program(get_option('XSLTPROC'), native: true, required: false)
+xsltproc_bin = find_program(get_option('XSLTPROC'), native: true, required: false)
 fop = find_program('fop', native: true, required: false)
 
+xmltools_wrapper = [
+  python, files('xmltools_dep_wrapper'),
+  '--targetname', '@OUTPUT@', '--depfile', '@DEPFILE@'
+]
+
+xmllint = xmltools_wrapper + [
+  '--tool', xmllint_bin, '--',
+]
+
 # Run validation only once, common to all subsequent targets.  While
 # we're at it, also resolve all entities (that is, copy all included
 # files into one big file).  This helps tools that don't understand
 # vpath builds (such as dbtoepub).
 postgres_full_xml = custom_target('postgres-full.xml',
   input: 'postgres.sgml',
   output: 'postgres-full.xml',
+  depfile: 'postgres-full.xml.d',
   command: [xmllint, '--noent', '--valid', '--path', '@OUTDIR@', '-o', '@OUTPUT@', '@INPUT@'],
   depends: doc_generated,
   build_by_default: false,
@@ -86,18 +96,20 @@ alldocs += postgres_full_xml
 #
 # Full documentation as html, text
 #
-if xsltproc.found()
+if xsltproc_bin.found()
   xsltproc_flags = [
     '--stringparam', 'pg.version', pg_version,
     '--param', 'website.stylesheet', '1'
   ]
 
+  xsltproc = xmltools_wrapper + [
+    '--tool', xsltproc_bin, '--',
+  ]
 
-  # FIXME: Should use a wrapper around xsltproc --load-trace to compute a
-  # depfile
   html = custom_target('html',
     input: ['stylesheet.xsl', postgres_full_xml],
     output: 'html',
+    depfile: 'html.d',
     depends: doc_generated,
     command: [xsltproc, '-o', '@OUTDIR@/', xsltproc_flags, '@INPUT@'],
     build_by_default: false,
@@ -110,6 +122,7 @@ if xsltproc.found()
   html_help = custom_target('html_help',
     input: ['stylesheet-hh.xsl', postgres_full_xml],
     output: 'htmlhelp',
+    depfile: 'htmlhelp.d',
     depends: doc_generated,
     command: [xsltproc, '--path', '@OUTDIR@', '-o', '@OUTDIR@/', xsltproc_flags, '@INPUT@'],
     build_by_default: false,
@@ -121,6 +134,7 @@ if xsltproc.found()
   postgres_html = custom_target('postgres.html',
     input: ['stylesheet-html-nochunk.xsl', postgres_full_xml],
     output: 'postgres.html',
+    depfile: 'postgres.html.d',
     depends: doc_generated,
     command: [xsltproc, '--path', '@OUTDIR@', '-o', '@OUTPUT@', xsltproc_flags, '@INPUT@'],
     build_by_default: false,
@@ -144,17 +158,19 @@ endif
 #
 # INSTALL in html, text
 #
-if xsltproc.found()
+if xsltproc_bin.found()
   install_xml = custom_target('INSTALL.xml',
     input: ['standalone-profile.xsl', 'standalone-install.xml'],
     output: 'INSTALL.xml',
+    depfile: 'INSTALL.xml.d',
     depends: doc_generated + [postgres_full_xml],
     command: [xsltproc, '--path', '@OUTDIR@', '-o', '@OUTPUT@', xsltproc_flags, '--xinclude', '@INPUT@'],
     build_by_default: false,
   )
   install_html = custom_target('INSTALL.html',
     input: ['stylesheet-text.xsl', install_xml],
     output: 'INSTALL.html',
+    depfile: 'INSTALL.html.d',
     command: [xsltproc, '--path', '@OUTDIR@', '-o', '@OUTPUT@', xsltproc_flags, '@INPUT@'],
     build_by_default: false,
   )
@@ -177,11 +193,12 @@ endif
 #
 # Man pages
 #
-if xsltproc.found()
+if xsltproc_bin.found()
   # FIXME: implement / consider sqlmansectnum logic
   man = custom_target('man',
     input: ['stylesheet-man.xsl', postgres_full_xml],
     output: ['man1', 'man3', 'man7'],
+    depfile: 'man.d',
     depends: doc_generated,
     command: [xsltproc, '--path', '@OUTDIR@', '-o', '@OUTDIR@/', xsltproc_flags, '@INPUT@'],
     build_by_default: false,
@@ -195,17 +212,19 @@ endif
 #
 # Full documentation as PDF
 #
-if fop.found() and xsltproc.found()
+if fop.found() and xsltproc_bin.found()
   xsltproc_fo_flags = xsltproc_flags + ['--stringparam', 'img.src.path', meson.current_source_dir() + '/']
 
   foreach format, detail: {'A4': 'A4', 'US': 'USletter'}
     postgres_x_fo_f = 'postgres-@0@.fo'.format(format)
+    postgres_x_fo_dep = 'postgres-@0@.fo.d'.format(format)
     postgres_x_pdf_f = 'postgres-@0@.pdf'.format(format)
 
     postgres_x_fo = custom_target(postgres_x_fo_f,
       input: ['stylesheet-fo.xsl', postgres_full_xml],
-      output: [postgres_x_fo_f],
+      output: postgres_x_fo_f,
       depends: doc_generated,
+      depfile: postgres_x_fo_dep,
       command: [xsltproc, '--path', '@OUTDIR@/', xsltproc_fo_flags,
                 '--stringparam', 'paper.type', detail,
                 '-o', '@OUTPUT@', '@INPUT@'],
@@ -230,7 +249,7 @@ endif
 # This was previously implemented using dbtoepub - but that doesn't seem to
 # support running in build != source directory (i.e. VPATH builds already
 # weren't supported).
-if pandoc.found() and xsltproc.found()
+if pandoc.found() and xsltproc_bin.found()
   postgres_epub = custom_target('postgres.epub',
     input: postgres_full_xml,
     output: 'postgres.epub',

doc/src/sgml/xmltools_dep_wrapper

@@ -0,0 +1,54 @@
+#!/usr/bin/env python3
+
+# A small wrapper around xmllint and xsltproc that collects dependency
+# information (in gcc's format) using --load-trace.
+
+import argparse
+import re
+import subprocess
+import sys
+
+parser = argparse.ArgumentParser(
+    description='generate dependency file for docs')
+
+parser.add_argument('--targetname', type=str, required=False, nargs='+')
+parser.add_argument('--depfile', type=str, required=False)
+parser.add_argument('--tool', type=str, required=True)
+parser.add_argument('flags', nargs='*')
+
+args = parser.parse_args()
+
+if args.depfile:
+    command = [args.tool, '--load-trace'] + args.flags
+
+    # list of targets that depend on the loaded files we see via --load-trace
+    line_start = ' '.join(args.targetname) + ': '
+
+    # --load-trace flag displays all the documents loaded during the processing
+    # to stderr
+    res = subprocess.run(command, stderr=subprocess.PIPE,
+                         universal_newlines=True)
+
+    line_re = re.compile('^Loaded URL="([^"]+)"')
+    with open(args.depfile, 'w') as f:
+        for line in res.stderr.splitlines():
+            m = re.match(line_re, line)
+
+            # continue to show errors
+            if m is None:
+                print(line, file=sys.stderr)
+                continue
+            # Absolute paths are printed as file://, relative paths as-is. We
+            # don't care about http://, as a) those will be printed even if
+            # resolved locally b) we couldn't have a dependency anyway.
+            fname = m.group(1)
+            if fname.startswith('http://'):
+                continue
+            if fname.startswith('file://'):
+                fname = fname.split('file://')[1]
+            f.write(line_start + fname + '\n')
+else:
+    command = [args.tool] + args.flags
+    res = subprocess.run(command)
+
+exit(res.returncode)