Run check_keywords.pl anytime gram.c is rebuilt.

This script is a bit slow, but still it only takes a fraction of the time
the bison run does, so the overhead doesn't seem intolerable.  And we
definitely need some mechanical aid here, because people keep missing
the need to add new keywords to the appropriate keyword-list production.

While at it, I moved check_keywords.pl from src/tools into
src/backend/parser where it's actually used, and did some very minor
cleanup on the script.

Author: tglsfdc

src/backend/parser/Makefile

@@ -41,6 +41,7 @@ endif
 gram.h: gram.c ;
 
 gram.c: gram.y
+	$(PERL) $(srcdir)/check_keywords.pl $< $(top_srcdir)/src/include/parser/kwlist.h
 ifdef BISON
 	$(BISON) -d $(BISONFLAGS) -o $@ $<
 else
@@ -65,7 +66,3 @@ gram.o keywords.o parser.o: gram.h
 # are not cleaned here.
 clean distclean maintainer-clean:
 	rm -f lex.backup
-
-
-maintainer-check:
-	$(PERL) $(top_srcdir)/src/tools/check_keywords.pl $(top_srcdir)

src/tools/check_keywords.pl renamed to src/backend/parser/check_keywords.pl

@@ -1,31 +1,25 @@
-#!/usr/bin/perl -w
+#!/usr/bin/perl
 
+# Check that the keyword lists in gram.y and kwlist.h are sane.
+# Usage: check_keywords.pl gram.y kwlist.h
+
+# src/backend/parser/check_keywords.pl
+# Copyright (c) 2009-2012, PostgreSQL Global Development Group
+
+use warnings;
 use strict;
 
-# Check that the keyword lists in gram.y and kwlist.h are sane. Run from
-# the top directory, or pass a path to a top directory as argument.
-#
-# src/tools/check_keywords.pl
+my $gram_filename = $ARGV[0];
+my $kwlist_filename = $ARGV[1];
 
 my $errors = 0;
-my $path;
 
 sub error(@)
 {
 	print STDERR @_;
 	$errors = 1;
 }
 
-if (@ARGV)
-{
-	$path = $ARGV[0];
-	shift @ARGV;
-}
-else
-{
-	$path = ".";
-}
-
 $, = ' ';     # set output field separator
 $\ = "\n";    # set output record separator
 
@@ -35,7 +29,6 @@ (@)
 $keyword_categories{'type_func_name_keyword'} = 'TYPE_FUNC_NAME_KEYWORD';
 $keyword_categories{'reserved_keyword'}       = 'RESERVED_KEYWORD';
 
-my $gram_filename = "$path/src/backend/parser/gram.y";
 open(GRAM, $gram_filename) || die("Could not open : $gram_filename");
 
 my ($S, $s, $k, $n, $kcat);
@@ -59,7 +52,6 @@ (@)
 
 	if (!($kcat))
 	{
-
 		# Is this the beginning of a keyword list?
 		foreach $k (keys %keyword_categories)
 		{
@@ -89,7 +81,6 @@ (@)
 		}
 		elsif ($arr[$fieldIndexer] eq '/*')
 		{
-
 			# start of a multiline comment
 			$comment = 1;
 			next;
@@ -101,7 +92,6 @@ (@)
 
 		if ($arr[$fieldIndexer] eq ';')
 		{
-
 			# end of keyword list
 			$kcat = '';
 			next;
@@ -118,46 +108,43 @@ (@)
 }
 close GRAM;
 
-# Check that all keywords are in alphabetical order
+# Check that each keyword list is in alphabetical order (just for neatnik-ism)
 my ($prevkword, $kword, $bare_kword);
 foreach $kcat (keys %keyword_categories)
 {
 	$prevkword = '';
 
 	foreach $kword (@{ $keywords{$kcat} })
 	{
-
 		# Some keyword have a _P suffix. Remove it for the comparison.
 		$bare_kword = $kword;
 		$bare_kword =~ s/_P$//;
 		if ($bare_kword le $prevkword)
 		{
 			error
 			  "'$bare_kword' after '$prevkword' in $kcat list is misplaced";
-			$errors = 1;
 		}
 		$prevkword = $bare_kword;
 	}
 }
 
 # Transform the keyword lists into hashes.
-# kwhashes is a hash of hashes, keyed by keyword category id, e.g.
-# UNRESERVED_KEYWORD. Each inner hash is a keyed by keyword id, e.g. ABORT_P
-# with a dummy value.
+# kwhashes is a hash of hashes, keyed by keyword category id,
+# e.g. UNRESERVED_KEYWORD.
+# Each inner hash is keyed by keyword id, e.g. ABORT_P, with a dummy value.
 my %kwhashes;
 while (my ($kcat, $kcat_id) = each(%keyword_categories))
 {
 	@arr = @{ $keywords{$kcat} };
 
 	my $hash;
-	foreach my $item (@arr) { $hash->{$item} = 1 }
+	foreach my $item (@arr) { $hash->{$item} = 1; }
 
 	$kwhashes{$kcat_id} = $hash;
 }
 
 # Now read in kwlist.h
 
-my $kwlist_filename = "$path/src/include/parser/kwlist.h";
 open(KWLIST, $kwlist_filename) || die("Could not open : $kwlist_filename");
 
 my $prevkwstring = '';
@@ -173,7 +160,7 @@ (@)
 		my ($kwname)   = $2;
 		my ($kwcat_id) = $3;
 
-		# Check that the list is in alphabetical order
+		# Check that the list is in alphabetical order (critical!)
 		if ($kwstring le $prevkwstring)
 		{
 			error
@@ -182,14 +169,14 @@ (@)
 		$prevkwstring = $kwstring;
 
 		# Check that the keyword string is valid: all lower-case ASCII chars
-		if ($kwstring !~ /^[a-z_]*$/)
+		if ($kwstring !~ /^[a-z_]+$/)
 		{
 			error
 "'$kwstring' is not a valid keyword string, must be all lower-case ASCII chars";
 		}
 
 		# Check that the keyword name is valid: all upper-case ASCII chars
-		if ($kwname !~ /^[A-Z_]*$/)
+		if ($kwname !~ /^[A-Z_]+$/)
 		{
 			error
 "'$kwname' is not a valid keyword name, must be all upper-case ASCII chars";
@@ -209,8 +196,7 @@ (@)
 
 		if (!(%kwhash))
 		{
-
-			#error "Unknown kwcat_id: $kwcat_id";
+			error "Unknown keyword category: $kwcat_id";
 		}
 		else
 		{
@@ -220,9 +206,9 @@ (@)
 			}
 			else
 			{
-
-				# Remove it from the hash, so that we can complain at the end
-				# if there's keywords left that were not found in kwlist.h
+				# Remove it from the hash, so that we can
+				# complain at the end if there's keywords left
+				# that were not found in kwlist.h
 				delete $kwhashes{$kwcat_id}->{$kwname};
 			}
 		}