Attached is a patch to add bytea support to JDBC.

This patch does the following:

- Adds binary datatype support (bytea)
- Changes getXXXStream()/setXXXStream() methods to be spec compliant
- Adds ability to revert to old behavior

Details:

Adds support for the binary type bytea.  The ResultSet.getBytes() and
PreparedStatement.setBytes() methods now work against columns of bytea
type.  This is a change in behavior from the previous code which assumed
the column type was OID and thus a LargeObject.  The new behavior is
more complient with the JDBC spec as BLOB/CLOB are to be used for
LargeObjects and the getBytes()/setBytes() methods are for the databases
binary datatype (which is bytea in postgres).

Changes the behavior of the getBinaryStream(), getAsciiStream(),
getCharacterStream(), getUnicodeStream() and their setXXXStream()
counterparts.  These methos now work against either the bytea type
(BinaryStream) or the text types (AsciiStream, CharacterStream,
UnicodeStream).  The previous behavior was that these all assumed the
underlying column was of type OID and thus a LargeObject.  The
spec/javadoc for these methods indicate that they are for LONGVARCHAR
and LONGVARBINARY datatypes, which are distinct from the BLOB/CLOB
datatypes.  Given that the bytea and text types support upto 1G, they
are the LONGVARBINARY and LONGVARCHAR datatypes in postgres.

Added support for turning off the above new functionality.  Given that
the changes above are not backwardly compatible (however they are more
spec complient), I added the ability to revert back to the old behavior.
  The Connection now takes an optional parameter named 'compatible'.  If
the value of '7.1' is passed, the driver reverts to the 7.1 behavior.
If the parameter is not passed or the value '7.2' is passed the behavior
is the new behavior.  The mechanism put in place can be used in the
future when/if similar needs arise to change behavior.  This is
patterned after how Oracle does this (i.e. Oracle has a 'compatible'
parameter that behaves in a similar manner).

Misc fixes.  Cleaned up a few things I encountered along the way.


Note that in testing the patch I needed to ignore whitespace differences
in order to get it to apply cleanly (i.e. patch -l -i byteapatch.diff).
Also this patch introduces a new file
(src/interfaces/jdbc/org/postgresql/util/PGbytea.java).

Barry Lind

Author: bmomjian

src/interfaces/jdbc/org/postgresql/Connection.java

@@ -11,7 +11,7 @@
 import org.postgresql.core.*;
 
 /**
- * $Id: Connection.java,v 1.28 2001/09/07 22:17:02 momjian Exp $
+ * $Id: Connection.java,v 1.29 2001/09/10 15:07:05 momjian Exp $
  *
  * This abstract class is used by org.postgresql.Driver to open either the JDBC1 or
  * JDBC2 versions of the Connection class.
@@ -22,15 +22,13 @@ public abstract class Connection
   // This is the network stream associated with this connection
   public PG_Stream pg_stream;
 
-  // This is set by org.postgresql.Statement.setMaxRows()
-  //public int maxrows = 0;		// maximum no. of rows; 0 = unlimited
-
   private String PG_HOST;
   private int PG_PORT;
   private String PG_USER;
   private String PG_PASSWORD;
   private String PG_DATABASE;
   private boolean PG_STATUS;
+  private String compatible;
 
   /**
    *  The encoding to use for this connection.
@@ -123,6 +121,11 @@ protected void openConnection(String host, int port, Properties info, String dat
     PG_PORT = port;
     PG_HOST = host;
     PG_STATUS = CONNECTION_BAD;
+    if(info.getProperty("compatible")==null) {
+      compatible = d.getMajorVersion() + "." + d.getMinorVersion();
+    } else {
+      compatible = info.getProperty("compatible");
+    }
 
     // Now make the initial connection
     try
@@ -968,6 +971,23 @@ public boolean haveMinimumServerVersion(String ver) throws SQLException
       return (getDBVersionNumber().compareTo(ver) >= 0);
   }
 
+  /**
+   * This method returns true if the compatible level set in the connection
+   * (which can be passed into the connection or specified in the URL)
+   * is at least the value passed to this method.  This is used to toggle
+   * between different functionality as it changes across different releases
+   * of the jdbc driver code.  The values here are versions of the jdbc client
+   * and not server versions.  For example in 7.1 get/setBytes worked on
+   * LargeObject values, in 7.2 these methods were changed to work on bytea
+   * values.  This change in functionality could be disabled by setting the
+   * "compatible" level to be 7.1, in which case the driver will revert to
+   * the 7.1 functionality.
+   */
+  public boolean haveMinimumCompatibleVersion(String ver) throws SQLException
+  {
+      return (compatible.compareTo(ver) >= 0);
+  }
+
 
   /**
    * This returns the java.sql.Types type for a PG type oid

src/interfaces/jdbc/org/postgresql/Driver.java.in

@@ -85,12 +85,26 @@ public class Driver implements java.sql.Driver
    * database.
    *
    * <p>The java.util.Properties argument can be used to pass arbitrary
-   * string tag/value pairs as connection arguments.  Normally, at least
+   * string tag/value pairs as connection arguments.  
+   *
+   * user - (optional) The user to connect as
+   * password - (optional) The password for the user
+   * charSet - (optional) The character set to be used for converting 
+   *   to/from the database to unicode.  If multibyte is enabled on the 
+   *   server then the character set of the database is used as the default,
+   *   otherwise the jvm character encoding is used as the default.
+   * compatible - This is used to toggle
+   *   between different functionality as it changes across different releases
+   *   of the jdbc driver code.  The values here are versions of the jdbc 
+   *   client and not server versions.  For example in 7.1 get/setBytes 
+   *   worked on LargeObject values, in 7.2 these methods were changed 
+   *   to work on bytea values.  This change in functionality could 
+   *   be disabled by setting the compatible level to be "7.1", in 
+   *   which case the driver will revert to the 7.1 functionality.
+   *
+   * <p>Normally, at least
    * "user" and "password" properties should be included in the
-   * properties.  In addition, the "charSet" property can be used to
-   * set a character set encoding (e.g. "utf-8") other than the platform
-   * default (typically Latin1).  This is necessary in particular if storing
-   * multibyte characters in the database.  For a list of supported
+   * properties.  For a list of supported
    * character encoding , see
    * http://java.sun.com/products/jdk/1.2/docs/guide/internat/encoding.doc.html
    * Note that you will probably want to have set up the Postgres database

src/interfaces/jdbc/org/postgresql/ResultSet.java

@@ -192,7 +192,8 @@ public String getFixedString(int col) throws SQLException {
       String s = getString(col);
 
       // Handle SQL Null
-      if(s==null)
+      wasNullFlag = (this_row[col - 1] == null);
+      if(wasNullFlag)
         return null;
 
       // Handle Money

src/interfaces/jdbc/org/postgresql/jdbc1/Connection.java

@@ -17,7 +17,7 @@
 import org.postgresql.util.*;
 
 /**
- * $Id: Connection.java,v 1.9 2001/09/06 03:13:34 momjian Exp $
+ * $Id: Connection.java,v 1.10 2001/09/10 15:07:05 momjian Exp $
  *
  * A Connection represents a session with a specific database.  Within the
  * context of a Connection, SQL statements are executed and results are
@@ -174,6 +174,7 @@ public int getSQLType(String pgTypeName)
     "float8",
     "bpchar","char","char2","char4","char8","char16",
     "varchar","text","name","filename",
+    "bytea",
     "bool",
     "date",
     "time",
@@ -197,6 +198,7 @@ public int getSQLType(String pgTypeName)
     Types.DOUBLE,
     Types.CHAR,Types.CHAR,Types.CHAR,Types.CHAR,Types.CHAR,Types.CHAR,
     Types.VARCHAR,Types.VARCHAR,Types.VARCHAR,Types.VARCHAR,
+    Types.BINARY,
     Types.BIT,
     Types.DATE,
     Types.TIME,

src/interfaces/jdbc/org/postgresql/jdbc1/PreparedStatement.java

@@ -82,7 +82,7 @@ public PreparedStatement(Connection connection, String sql) throws SQLException
 	 * A Prepared SQL query is executed and its ResultSet is returned
 	 *
 	 * @return a ResultSet that contains the data produced by the
-	 *	query - never null
+	 *     	 *	query - never null
 	 * @exception SQLException if a database access error occurs
 	 */
 	public java.sql.ResultSet executeQuery() throws SQLException
@@ -107,7 +107,7 @@ public java.sql.ResultSet executeQuery() throws SQLException
 	 * be executed.
 	 *
 	 * @return either the row count for INSERT, UPDATE or DELETE; or
-	 * 	0 for SQL statements that return nothing.
+	 *     	 * 	0 for SQL statements that return nothing.
 	 * @exception SQLException if a database access error occurs
 	 */
 	public int executeUpdate() throws SQLException
@@ -294,12 +294,22 @@ public void setString(int parameterIndex, String x) throws SQLException
    */
   public void setBytes(int parameterIndex, byte x[]) throws SQLException
   {
+    if (connection.haveMinimumCompatibleVersion("7.2")) {
+      //Version 7.2 supports the bytea datatype for byte arrays
+      if(null == x){
+        setNull(parameterIndex,Types.OTHER);
+      } else {
+        setString(parameterIndex, PGbytea.toPGString(x));
+      }
+    } else {
+      //Version 7.1 and earlier support done as LargeObjects
     LargeObjectManager lom = connection.getLargeObjectAPI();
     int oid = lom.create();
     LargeObject lob = lom.open(oid);
     lob.write(x);
     lob.close();
     setInt(parameterIndex,oid);
+  }
   }
 
 	/**
@@ -386,8 +396,29 @@ public void setTimestamp(int parameterIndex, Timestamp x) throws SQLException
 	 */
 	public void setAsciiStream(int parameterIndex, InputStream x, int length) throws SQLException
 	{
+          if (connection.haveMinimumCompatibleVersion("7.2")) {
+            //Version 7.2 supports AsciiStream for all PG text types (char, varchar, text)
+            //As the spec/javadoc for this method indicate this is to be used for
+            //large String values (i.e. LONGVARCHAR)  PG doesn't have a separate
+            //long varchar datatype, but with toast all text datatypes are capable of
+            //handling very large values.  Thus the implementation ends up calling
+            //setString() since there is no current way to stream the value to the server
+            try {
+              InputStreamReader l_inStream = new InputStreamReader(x, "ASCII");
+              char[] l_chars = new char[length];
+              int l_charsRead = l_inStream.read(l_chars,0,length);
+              setString(parameterIndex, new String(l_chars,0,l_charsRead));
+            } catch (UnsupportedEncodingException l_uee) {
+              throw new PSQLException("postgresql.unusual",l_uee);
+            } catch (IOException l_ioe) {
+              throw new PSQLException("postgresql.unusual",l_ioe);
+            }
+          } else {
+            //Version 7.1 supported only LargeObjects by treating everything
+            //as binary data
 		setBinaryStream(parameterIndex, x, length);
 	}
+        }
 
 	/**
 	 * When a very large Unicode value is input to a LONGVARCHAR parameter,
@@ -406,8 +437,29 @@ public void setAsciiStream(int parameterIndex, InputStream x, int length) throws
 	 */
 	public void setUnicodeStream(int parameterIndex, InputStream x, int length) throws SQLException
 	{
+          if (connection.haveMinimumCompatibleVersion("7.2")) {
+            //Version 7.2 supports AsciiStream for all PG text types (char, varchar, text)
+            //As the spec/javadoc for this method indicate this is to be used for
+            //large String values (i.e. LONGVARCHAR)  PG doesn't have a separate
+            //long varchar datatype, but with toast all text datatypes are capable of
+            //handling very large values.  Thus the implementation ends up calling
+            //setString() since there is no current way to stream the value to the server
+            try {
+              InputStreamReader l_inStream = new InputStreamReader(x, "UTF-8");
+              char[] l_chars = new char[length];
+              int l_charsRead = l_inStream.read(l_chars,0,length);
+              setString(parameterIndex, new String(l_chars,0,l_charsRead));
+            } catch (UnsupportedEncodingException l_uee) {
+              throw new PSQLException("postgresql.unusual",l_uee);
+            } catch (IOException l_ioe) {
+              throw new PSQLException("postgresql.unusual",l_ioe);
+            }
+          } else {
+            //Version 7.1 supported only LargeObjects by treating everything
+            //as binary data
 		setBinaryStream(parameterIndex, x, length);
 	}
+        }
 
 	/**
 	 * When a very large binary value is input to a LONGVARBINARY parameter,
@@ -425,7 +477,54 @@ public void setUnicodeStream(int parameterIndex, InputStream x, int length) thro
 	 */
 	public void setBinaryStream(int parameterIndex, InputStream x, int length) throws SQLException
 	{
-	    throw org.postgresql.Driver.notImplemented();
+          if (connection.haveMinimumCompatibleVersion("7.2")) {
+            //Version 7.2 supports BinaryStream for for the PG bytea type
+            //As the spec/javadoc for this method indicate this is to be used for
+            //large binary values (i.e. LONGVARBINARY)  PG doesn't have a separate
+            //long binary datatype, but with toast the bytea datatype is capable of
+            //handling very large values.  Thus the implementation ends up calling
+            //setBytes() since there is no current way to stream the value to the server
+            byte[] l_bytes = new byte[length];
+            int l_bytesRead;
+            try {
+              l_bytesRead = x.read(l_bytes,0,length);
+            } catch (IOException l_ioe) {
+              throw new PSQLException("postgresql.unusual",l_ioe);
+            }
+            if (l_bytesRead == length) {
+              setBytes(parameterIndex, l_bytes);
+            } else {
+              //the stream contained less data than they said
+              byte[] l_bytes2 = new byte[l_bytesRead];
+              System.arraycopy(l_bytes,0,l_bytes2,0,l_bytesRead);
+              setBytes(parameterIndex, l_bytes2);
+            }
+          } else {
+            //Version 7.1 only supported streams for LargeObjects
+            //but the jdbc spec indicates that streams should be
+            //available for LONGVARBINARY instead
+            LargeObjectManager lom = connection.getLargeObjectAPI();
+            int oid = lom.create();
+            LargeObject lob = lom.open(oid);
+            OutputStream los = lob.getOutputStream();
+            try {
+              // could be buffered, but then the OutputStream returned by LargeObject
+              // is buffered internally anyhow, so there would be no performance
+              // boost gained, if anything it would be worse!
+              int c=x.read();
+              int p=0;
+              while(c>-1 && p<length) {
+                los.write(c);
+                c=x.read();
+                p++;
+              }
+              los.close();
+            } catch(IOException se) {
+              throw new PSQLException("postgresql.unusual",se);
+            }
+            // lob is closed by the stream so don't call lob.close()
+            setInt(parameterIndex,oid);
+          }
 	}
 
 	/**
@@ -460,8 +559,8 @@ public void clearParameters() throws SQLException
 	 * @param x the object containing the input parameter value
 	 * @param targetSqlType The SQL type to be send to the database
 	 * @param scale For java.sql.Types.DECIMAL or java.sql.Types.NUMERIC
-	 *	types this is the number of digits after the decimal.  For 
-	 *	all other types this value will be ignored.
+	 *     	 *	types this is the number of digits after the decimal.  For
+	 *     	 *	all other types this value will be ignored.
 	 * @exception SQLException if a database access error occurs
 	 */
 	public void setObject(int parameterIndex, Object x, int targetSqlType, int scale) throws SQLException
@@ -572,7 +671,7 @@ else if (x instanceof PGobject)
 	 * statements handled by executeQuery and executeUpdate
 	 *
 	 * @return true if the next result is a ResultSet; false if it is an
-	 *	update count or there are no more results
+	 *     	 *	update count or there are no more results
 	 * @exception SQLException if a database access error occurs
 	 */
 	public boolean execute() throws SQLException