Improve parser so that we can show an error cursor position for errors

during parse analysis, not only errors detected in the flex/bison stages.
This is per my earlier proposal.  This commit includes all the basic
infrastructure, but locations are only tracked and reported for errors
involving column references, function calls, and operators.  More could
be done later but this seems like a good set to start with.  I've also
moved the ReportSyntaxErrorPosition logic out of psql and into libpq,
which should make it available to more people --- even within psql this
is an improvement because warnings weren't handled by ReportSyntaxErrorPosition.

1 files changed, 10 insertions, 1 deletions

diff --git a/src/include/parser/gramparse.h b/src/include/parser/gramparse.h
index 054604521b..13af69116c 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 1.33 2006/03/07 01:00:18 tgl Exp $
+ * $PostgreSQL: pgsql/src/include/parser/gramparse.h,v 1.34 2006/03/14 22:48:22 tgl Exp $
  *
  *-------------------------------------------------------------------------
  */
@@ -17,6 +17,15 @@
 
 #include "nodes/parsenodes.h"
 
+/*
+ * We track token locations in terms of byte offsets from the start of the
+ * source string, not the column number/line number representation that
+ * bison uses by default.  Also, to minimize overhead we track only one
+ * location (usually the first token location) for each construct, not
+ * the beginning and ending locations as bison does by default.  It's
+ * therefore sufficient to make YYLTYPE an int.
+ */
+#define YYLTYPE  int
 
 /* from scan.l */
 extern void scanner_init(const char *str);