Code review for standard_conforming_strings patch. Fix it so it does not

throw warnings for 100%-SQL-standard constructs, clean up some minor
infelicities, try to un-break ecpg to the best of my ability.  (It's not clear
how ecpg is going to find out the setting of standard_conforming_strings,
though.)  I think pg_dump still needs work, too.

10 files changed, 80 insertions, 89 deletions

diff --git a/doc/src/sgml/config.sgml b/doc/src/sgml/config.sgml
index 933ca3811dc..a2e6d46ca6a 100644
--- a/doc/src/sgml/config.sgml
+++ b/doc/src/sgml/config.sgml
@@ -1,4 +1,4 @@
-<!-- $PostgreSQL: pgsql/doc/src/sgml/config.sgml,v 1.57 2006/05/02 18:07:51 tgl Exp $ -->
+<!-- $PostgreSQL: pgsql/doc/src/sgml/config.sgml,v 1.58 2006/05/11 19:15:35 tgl Exp $ -->
 
 <chapter Id="runtime-config">
   <title>Server Configuration</title>
@@ -3734,32 +3734,15 @@ dynamic_library_path = 'C:\tools\postgresql;H:\my_project\lib;$libdir'
        <para>
         When on, a warning is issued if a backslash (<literal>\</>)
         appears in an ordinary string literal (<literal>'...'</>
-        syntax). The default is <literal>on</>.
+        syntax) and <varname>standard_conforming_strings</varname> is off.
+        The default is <literal>on</>.
        </para>
        <para>
-        Escape string syntax (<literal>E'...'</>) should be used for
-        backslash escape sequences, because ordinary strings have
-        the standard-conforming behavior of treating backslashes
-        literally when the <literal>standard-conforming-strings</>
-        option is set <literal>on</>.
-       </para>
-      </listitem>
-     </varlistentry>
-
-     <varlistentry id="guc-standard-conforming-strings" xreflabel="standard_conforming_strings">
-      <term><varname>standard_conforming_strings</varname> (<type>boolean</type>)</term>
-      <indexterm><primary>strings</><secondary>standard conforming</></>
-      <indexterm>
-       <primary><varname>standard_conforming_strings</> configuration parameter</primary>
-      </indexterm>
-      <listitem>
-       <para>
-        Controls whether ordinary string literals
-        (<literal>'...'</>) treat backslashes literally, as specified in
-        the SQL standard.  Applications may check this
-        parameter to determine how string literals will be processed.
-        The presence of this parameter can also be taken as an indication
-        that the escape string syntax (<literal>E'...'</>) is supported.
+        Applications that wish to use backslash as escape should be
+        modified to use escape string syntax (<literal>E'...'</>),
+        because the default behavior of ordinary strings will change
+        in a future release for SQL compatibility.  This variable can
+        be enabled to help detect applications that will break.
        </para>
       </listitem>
      </varlistentry>
@@ -3799,6 +3782,32 @@ dynamic_library_path = 'C:\tools\postgresql;H:\my_project\lib;$libdir'
       </listitem>
      </varlistentry>
 
+     <varlistentry id="guc-standard-conforming-strings" xreflabel="standard_conforming_strings">
+      <term><varname>standard_conforming_strings</varname> (<type>boolean</type>)</term>
+      <indexterm><primary>strings</><secondary>standard conforming</></>
+      <indexterm>
+       <primary><varname>standard_conforming_strings</> configuration parameter</primary>
+      </indexterm>
+      <listitem>
+       <para>
+        This controls whether ordinary string literals
+        (<literal>'...'</>) treat backslashes literally, as specified in
+        the SQL standard.
+        The default is currently <literal>off</>, causing
+        <productname>PostgreSQL</productname> to have its historical
+        behavior of treating backslashes as escape characters.
+        The default will change to <literal>on</> in a future release
+        to improve compatibility with the standard.
+        Applications may check this
+        parameter to determine how string literals will be processed.
+        The presence of this parameter can also be taken as an indication
+        that the escape string syntax (<literal>E'...'</>) is supported.
+        Escape string syntax should be used if an application desires
+        backslashes to be treated as escape characters.
+       </para>
+      </listitem>
+     </varlistentry>
+
      </variablelist>
     </sect2>
 
diff --git a/src/backend/parser/scan.l b/src/backend/parser/scan.l
index 0ac8d346a7b..78a888e02af 100644
--- a/src/backend/parser/scan.l
+++ b/src/backend/parser/scan.l
@@ -24,7 +24,7 @@
  * Portions Copyright (c) 1994, Regents of the University of California
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/parser/scan.l,v 1.133 2006/03/14 22:48:21 tgl Exp $
+ *	  $PostgreSQL: pgsql/src/backend/parser/scan.l,v 1.134 2006/05/11 19:15:35 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -51,12 +51,12 @@ static char    *dolqstart;      /* current $foo$ quote start string */
 /*
  * GUC variables.  This is a DIRECT violation of the warning given at the
  * head of gram.y, ie flex/bison code must not depend on any GUC variables;
- * as such, changing its value can induce very unintuitive behavior.
+ * as such, changing their values can induce very unintuitive behavior.
  * But we shall have to live with it as a short-term thing until the switch
  * to SQL-standard string syntax is complete.
  */
-bool			escape_string_warning;
-bool			standard_conforming_strings;
+bool			escape_string_warning = true;
+bool			standard_conforming_strings = false;
 
 static bool		warn_on_first_escape;
 
@@ -211,8 +211,7 @@ xehexesc		[\\]x[0-9A-Fa-f]{1,2}
  */
 xqstart			{quote}
 xqdouble		{quote}{quote}
-xqinside		[^\\']+
-xqbackslash	[\\]
+xqinside		[^']+
 
 /* $foo$ style quotes ("dollar quoting")
  * The quoted string starts with $foo$ where "foo" is an optional string
@@ -443,7 +442,7 @@ other			.
 					yylval.str = litbufdup();
 					return SCONST;
 				}
-<xq,xe>{xqdouble}  {
+<xq,xe>{xqdouble} {
 					addlitchar('\'');
 				}
 <xq>{xqinside}  {
@@ -452,10 +451,6 @@ other			.
 <xe>{xeinside}  {
 					addlit(yytext, yyleng);
 				}
-<xq>{xqbackslash} {
-					check_string_escape_warning(yytext[1]);
-					addlitchar('\\');
-                }
 <xe>{xeescape}  {
 					check_string_escape_warning(yytext[1]);
 					addlitchar(unescape_single_char(yytext[1]));
diff --git a/src/backend/utils/misc/guc.c b/src/backend/utils/misc/guc.c
index c2c1318b4cb..87df3fdbacd 100644
--- a/src/backend/utils/misc/guc.c
+++ b/src/backend/utils/misc/guc.c
@@ -10,7 +10,7 @@
  * Written by Peter Eisentraut <peter_e@gmx.net>.
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/backend/utils/misc/guc.c,v 1.318 2006/05/02 11:28:55 teodor Exp $
+ *	  $PostgreSQL: pgsql/src/backend/utils/misc/guc.c,v 1.319 2006/05/11 19:15:35 tgl Exp $
  *
  *--------------------------------------------------------------------
  */
@@ -46,6 +46,7 @@
 #include "optimizer/geqo.h"
 #include "optimizer/paths.h"
 #include "optimizer/planmain.h"
+#include "parser/gramparse.h"
 #include "parser/parse_expr.h"
 #include "parser/parse_relation.h"
 #include "parser/scansup.h"
diff --git a/src/backend/utils/misc/postgresql.conf.sample b/src/backend/utils/misc/postgresql.conf.sample
index 84c9f795848..dac59378a07 100644
--- a/src/backend/utils/misc/postgresql.conf.sample
+++ b/src/backend/utils/misc/postgresql.conf.sample
@@ -279,7 +279,7 @@
 					#   warning
 					#   error
 					#   panic(off)
-				 
+
 #log_min_duration_statement = -1	# -1 is disabled, 0 logs all statements
 					# and their durations, in milliseconds.
 
@@ -415,8 +415,8 @@
 #add_missing_from = off
 #array_nulls = on
 #default_with_oids = off
-escape_string_warning = on		# warn about backslashes in string literals
-#standard_conforming_strings = off	# SQL standard string literal processing
+#escape_string_warning = on
+#standard_conforming_strings = off
 #regex_flavor = advanced		# advanced, extended, or basic
 #sql_inheritance = on
 
diff --git a/src/bin/psql/common.c b/src/bin/psql/common.c
index ec0acc156e7..ba8e403dd2a 100644
--- a/src/bin/psql/common.c
+++ b/src/bin/psql/common.c
@@ -3,7 +3,7 @@
  *
  * Copyright (c) 2000-2006, PostgreSQL Global Development Group
  *
- * $PostgreSQL: pgsql/src/bin/psql/common.c,v 1.116 2006/03/14 22:48:22 tgl Exp $
+ * $PostgreSQL: pgsql/src/bin/psql/common.c,v 1.117 2006/05/11 19:15:35 tgl Exp $
  */
 #include "postgres_fe.h"
 #include "common.h"
@@ -1101,8 +1101,8 @@ is_superuser(void)
 /*
  * Test if the current session uses standard string literals.
  *
- * Note: this will correctly detect the setting only with a protocol-3.0
- * or newer backend; otherwise it will always say "false".
+ * Note: With a pre-protocol-3.0 connection this will always say "false",
+ * which should be the right answer.
  */
 bool
 standard_strings(void)
diff --git a/src/bin/psql/psqlscan.l b/src/bin/psql/psqlscan.l
index 6c04aaf30ba..4a344baf6b2 100644
--- a/src/bin/psql/psqlscan.l
+++ b/src/bin/psql/psqlscan.l
@@ -33,7 +33,7 @@
  * Portions Copyright (c) 1994, Regents of the University of California
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/bin/psql/psqlscan.l,v 1.17 2006/03/06 19:49:20 momjian Exp $
+ *	  $PostgreSQL: pgsql/src/bin/psql/psqlscan.l,v 1.18 2006/05/11 19:15:35 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -467,7 +467,7 @@ other			.
 					BEGIN(INITIAL);
 					ECHO;
 				}
-<xq,xe>{xqdouble}  {
+<xq,xe>{xqdouble} {
 					ECHO;
 				}
 <xq>{xqinside}  {
diff --git a/src/include/parser/gramparse.h b/src/include/parser/gramparse.h
index 13af69116cd..8430ad8c8e8 100644
--- a/src/include/parser/gramparse.h
+++ b/src/include/parser/gramparse.h
@@ -7,7 +7,7 @@
  * Portions Copyright (c) 1996-2006, PostgreSQL Global Development Group
  * Portions Copyright (c) 1994, Regents of the University of California
  *
- * $PostgreSQL: pgsql/src/include/parser/gramparse.h,v 1.34 2006/03/14 22:48:22 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/parser/gramparse.h,v 1.35 2006/05/11 19:15:35 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -27,6 +27,11 @@
  */
 #define YYLTYPE  int
 
+/* GUC variables in scan.l (every one of these is a bad idea :-() */
+extern bool escape_string_warning;
+extern bool standard_conforming_strings;
+
+
 /* from scan.l */
 extern void scanner_init(const char *str);
 extern void scanner_finish(void);
diff --git a/src/include/utils/guc.h b/src/include/utils/guc.h
index d60f5061b26..182a762cc02 100644
--- a/src/include/utils/guc.h
+++ b/src/include/utils/guc.h
@@ -7,7 +7,7 @@
  * Copyright (c) 2000-2006, PostgreSQL Global Development Group
  * Written by Peter Eisentraut <peter_e@gmx.net>.
  *
- * $PostgreSQL: pgsql/src/include/utils/guc.h,v 1.67 2006/03/07 03:01:22 momjian Exp $
+ * $PostgreSQL: pgsql/src/include/utils/guc.h,v 1.68 2006/05/11 19:15:35 tgl Exp $
  *--------------------------------------------------------------------
  */
 #ifndef GUC_H
@@ -120,8 +120,6 @@ extern bool SQL_inheritance;
 extern bool Australian_timezones;
 
 extern bool default_with_oids;
-extern bool escape_string_warning;
-extern bool standard_conforming_strings;
 
 extern int	log_min_error_statement;
 extern int	log_min_messages;
diff --git a/src/interfaces/ecpg/preproc/pgc.l b/src/interfaces/ecpg/preproc/pgc.l
index 67f2517d21f..a835a1da696 100644
--- a/src/interfaces/ecpg/preproc/pgc.l
+++ b/src/interfaces/ecpg/preproc/pgc.l
@@ -12,7 +12,7 @@
  *
  *
  * IDENTIFICATION
- *	  $PostgreSQL: pgsql/src/interfaces/ecpg/preproc/pgc.l,v 1.145 2006/03/06 19:49:20 momjian Exp $
+ *	  $PostgreSQL: pgsql/src/interfaces/ecpg/preproc/pgc.l,v 1.146 2006/05/11 19:15:36 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -29,8 +29,8 @@ extern YYSTYPE yylval;
 
 static int		xcdepth = 0;	/* depth of nesting in slash-star comments */
 static char	   *dolqstart;      /* current $foo$ quote start string */
-bool			escape_string_warning;
-bool			standard_conforming_strings;
+static bool		escape_string_warning;
+static bool		standard_conforming_strings;
 static bool 	warn_on_first_escape;
 
 /*
@@ -128,10 +128,10 @@ xnstart			[nN]{quote}
 
 /* Quoted string that allows backslash escapes */
 xestart 		[eE]{quote}
-xeinside                [^\\']+
-xeescape	                [\\][^0-7]
-xeoctesc	                [\\][0-7]{1,3}
-xehexesc	                [\\]x[0-9A-Fa-f]{1,2}
+xeinside		[^\\']+
+xeescape		[\\][^0-7]
+xeoctesc		[\\][0-7]{1,3}
+xehexesc		[\\]x[0-9A-Fa-f]{1,2}
 
 /* C version of hex number */
 xch				0[xX][0-9A-Fa-f]*
@@ -141,8 +141,7 @@ xch				0[xX][0-9A-Fa-f]*
  */
 xqstart			{quote}
 xqdouble		{quote}{quote}
-xqinside		[^\\']+
-xqbackslash	[\\]
+xqinside		[^']+
 
 /* $foo$ style quotes ("dollar quoting")
  * The quoted string starts with $foo$ where "foo" is an optional string
@@ -402,11 +401,23 @@ cppline			{space}*#(.*\\{space})*.*{newline}
 				/* National character.
 				 * Transfer it as-is to the backend.
 				 */
+				warn_on_first_escape = true;
 		        token_start = yytext;
-				BEGIN(xq);
+				state_before = YYSTATE;
+				if (standard_conforming_strings)
+					BEGIN(xq);
+				else
+					BEGIN(xe);
+				startlit();
+			}
+<C>{xqstart}	{
+				warn_on_first_escape = false;
+				token_start = yytext;
+				state_before = YYSTATE;
+				BEGIN(xe);
 				startlit();
 			}
-<C,SQL>{xqstart}	{
+<SQL>{xqstart}	{
 				warn_on_first_escape = true;
 				token_start = yytext;
 				state_before = YYSTATE;
@@ -416,7 +427,7 @@ cppline			{space}*#(.*\\{space})*.*{newline}
 					BEGIN(xe);
 				startlit();
 			}
-<C,SQL>{xestart}	{
+<SQL>{xestart}	{
 				warn_on_first_escape = false;
 				token_start = yytext;
 				state_before = YYSTATE;
@@ -433,10 +444,6 @@ cppline			{space}*#(.*\\{space})*.*{newline}
 <xq,xe>{xqdouble}		{ addlitchar('\''); }
 <xq>{xqinside}		{ addlit(yytext, yyleng); }
 <xe>{xeinside}		{ addlit(yytext, yyleng); }
-<xq>{xqbackslash} {
-					check_escape_warning();
-					addlitchar('\\');
-                }
 <xe>{xeescape}  	{ 
 				check_escape_warning();
 				addlit(yytext, yyleng);
diff --git a/src/test/regress/expected/strings.out b/src/test/regress/expected/strings.out
index 0c9b88ee8f0..20081f699b6 100644
--- a/src/test/regress/expected/strings.out
+++ b/src/test/regress/expected/strings.out
@@ -927,30 +927,6 @@ show standard_conforming_strings;
 (1 row)
 
 select 'a\bcd' as f1, 'a\b''cd' as f2, 'a\b''''cd' as f3, 'abcd\'   as f4, 'ab\''cd' as f5, '\\' as f6;
-WARNING:  nonstandard use of escape in a string literal
-LINE 1: select 'a\bcd' as f1, 'a\b''cd' as f2, 'a\b''''cd' as f3, 'a...
-               ^
-HINT:  Use the escape string syntax for escapes, e.g., E'\r\n'.
-WARNING:  nonstandard use of escape in a string literal
-LINE 1: select 'a\bcd' as f1, 'a\b''cd' as f2, 'a\b''''cd' as f3, 'a...
-                              ^
-HINT:  Use the escape string syntax for escapes, e.g., E'\r\n'.
-WARNING:  nonstandard use of escape in a string literal
-LINE 1: select 'a\bcd' as f1, 'a\b''cd' as f2, 'a\b''''cd' as f3, 'a...
-                                               ^
-HINT:  Use the escape string syntax for escapes, e.g., E'\r\n'.
-WARNING:  nonstandard use of escape in a string literal
-LINE 1: ...a\bcd' as f1, 'a\b''cd' as f2, 'a\b''''cd' as f3, 'abcd\'   ...
-                                                             ^
-HINT:  Use the escape string syntax for escapes, e.g., E'\r\n'.
-WARNING:  nonstandard use of escape in a string literal
-LINE 1: ...b''cd' as f2, 'a\b''''cd' as f3, 'abcd\'   as f4, 'ab\''cd' ...
-                                                             ^
-HINT:  Use the escape string syntax for escapes, e.g., E'\r\n'.
-WARNING:  nonstandard use of escape in a string literal
-LINE 1: ...b''''cd' as f3, 'abcd\'   as f4, 'ab\''cd' as f5, '\\' as f6...
-                                                             ^
-HINT:  Use the escape string syntax for escapes, e.g., E'\r\n'.
   f1   |   f2   |   f3    |  f4   |   f5   | f6 
 -------+--------+---------+-------+--------+----
  a\bcd | a\b'cd | a\b''cd | abcd\ | ab\'cd | \\