From 5b5e8a29c13aa583f8a674cb65f92a204da24be8 Mon Sep 17 00:00:00 2001
From: Andrew Dunstan <andrew@dunslane.net>
Date: Mon, 6 Oct 2025 07:53:31 -0400
Subject: [PATCH] Revert "Improve docs syntax checking"

This reverts commit b292256272623d1a7532f3893a4565d1944742b4.

Further discussion is needed

Discussion: https://postgr.es/m/0198ec0f-0269-4cf4-b4a7-22c05b3047cb@eisentraut.org
---
 .cirrus.tasks.yml                 |  3 --
 doc/src/sgml/Makefile             | 18 ++++++--
 doc/src/sgml/meson.build          | 23 ---------
 doc/src/sgml/sgml_syntax_check.pl | 77 -------------------------------
 4 files changed, 15 insertions(+), 106 deletions(-)
 delete mode 100755 doc/src/sgml/sgml_syntax_check.pl

diff --git a/.cirrus.tasks.yml b/.cirrus.tasks.yml
index 1c937247a9a..eca9d62fc22 100644
--- a/.cirrus.tasks.yml
+++ b/.cirrus.tasks.yml
@@ -627,8 +627,6 @@ task:
     TEST_JOBS: 8
     IMAGE: ghcr.io/cirruslabs/macos-runner:sonoma
 
-    XML_CATALOG_FILES: /opt/local/share/xml/docbook/4.5/catalog.xml
-
     CIRRUS_WORKING_DIR: ${HOME}/pgsql/
     CCACHE_DIR: ${HOME}/ccache
     MACPORTS_CACHE: ${HOME}/macports-cache
@@ -643,7 +641,6 @@ task:
 
     MACOS_PACKAGE_LIST: >-
       ccache
-      docbook-xml-4.5
       icu
       kerberos5
       lz4
diff --git a/doc/src/sgml/Makefile b/doc/src/sgml/Makefile
index 4d45b805da2..b53b2694a6b 100644
--- a/doc/src/sgml/Makefile
+++ b/doc/src/sgml/Makefile
@@ -124,7 +124,7 @@ ifeq ($(STYLE),website)
 XSLTPROC_HTML_FLAGS += --param website.stylesheet 1
 endif
 
-html: check html-stamp
+html: html-stamp
 
 html-stamp: stylesheet.xsl postgres-full.xml $(ALL_IMAGES)
 	$(XSLTPROC) $(XMLINCLUDE) $(XSLTPROCFLAGS) $(XSLTPROC_HTML_FLAGS) $(wordlist 1,2,$^)
@@ -200,8 +200,8 @@ MAKEINFO = makeinfo
 ##
 
 # Quick syntax check without style processing
-check: postgres.sgml $(ALL_SGML)
-	$(PERL) $(srcdir)/sgml_syntax_check.pl --xmllint "$(XMLLINT)" --srcdir $(srcdir)
+check: postgres.sgml $(ALL_SGML) check-tabs check-nbsp
+	$(XMLLINT) $(XMLINCLUDE) --noout --valid $<
 
 
 ##
@@ -261,6 +261,18 @@ clean-man:
 
 endif # sqlmansectnum != 7
 
+# tabs are harmless, but it is best to avoid them in SGML files
+check-tabs:
+	@( ! grep '	' $(wildcard $(srcdir)/*.sgml $(srcdir)/func/*.sgml $(srcdir)/ref/*.sgml $(srcdir)/*.xsl) ) || \
+	(echo "Tabs appear in SGML/XML files" 1>&2;  exit 1)
+
+# Non-breaking spaces are harmless, but it is best to avoid them in SGML files.
+# Use perl command because non-GNU grep or sed could not have hex escape sequence.
+check-nbsp:
+	@ ( $(PERL) -ne '/\xC2\xA0/ and print("$$ARGV:$$_"),$$n++; END {exit($$n>0)}' \
+	  $(wildcard $(srcdir)/*.sgml $(srcdir)/func/*.sgml $(srcdir)/ref/*.sgml $(srcdir)/*.xsl $(srcdir)/images/*.xsl) ) || \
+	(echo "Non-breaking spaces appear in SGML/XML files" 1>&2;  exit 1)
+
 ##
 ## Clean
 ##
diff --git a/doc/src/sgml/meson.build b/doc/src/sgml/meson.build
index ce0dea587cd..6ae192eac68 100644
--- a/doc/src/sgml/meson.build
+++ b/doc/src/sgml/meson.build
@@ -306,26 +306,3 @@ endif
 if alldocs.length() != 0
   alias_target('alldocs', alldocs)
 endif
-
-sgml_syntax_check = files(
-  'sgml_syntax_check.pl'
-)
-
-test(
-  'sgml_syntax_check',
-  perl,
-  protocol: 'exitcode',
-  suite: 'doc',
-  args: [
-    sgml_syntax_check,
-    '--xmllint',
-      '@0@ --nonet'.format(xmllint_bin.full_path()),
-    '--srcdir',
-      meson.current_source_dir(),
-    '--builddir',
-      meson.current_build_dir(),
-  ],
-  depends: doc_generated
-)
-
-testprep_targets += doc_generated
diff --git a/doc/src/sgml/sgml_syntax_check.pl b/doc/src/sgml/sgml_syntax_check.pl
deleted file mode 100755
index 2264700a453..00000000000
--- a/doc/src/sgml/sgml_syntax_check.pl
+++ /dev/null
@@ -1,77 +0,0 @@
-# /usr/bin/perl
-
-# Copyright (c) 2025, PostgreSQL Global Development Group
-
-# doc/src/sgml/sgml_syntax_check.pl
-
-use strict;
-use warnings FATAL => 'all';
-use Getopt::Long;
-
-use File::Find;
-
-my $xmllint;
-my $srcdir;
-my $builddir;
-
-GetOptions(
-	'xmllint:s' => \$xmllint,
-	'srcdir:s' => \$srcdir,
-	'builddir:s' => \$builddir) or die "$0: wrong arguments";
-
-die "$0: --srcdir must be specified\n" unless defined $srcdir;
-
-my $xmlinclude = "--path . --path $srcdir";
-$xmlinclude .= " --path $builddir" if defined $builddir;
-
-# find files to process - all the sgml and xsl files (including in subdirectories)
-my @files_to_process;
-my @dirs_to_search = ($srcdir);
-push @dirs_to_search, $builddir if defined $builddir;
-find(
-	sub {
-		return unless -f $_;
-		return if $_ !~ /\.(sgml|xsl)$/;
-		push @files_to_process, $File::Find::name;
-	},
-	@dirs_to_search,);
-
-# tabs and non-breaking spaces are harmless, but it is best to avoid them in SGML files
-sub check_tabs_and_nbsp
-{
-	my $errors = 0;
-	for my $f (@files_to_process)
-	{
-		open my $fh, "<:encoding(UTF-8)", $f or die "Can't open $f: $!";
-		my $line_no = 0;
-		while (<$fh>)
-		{
-			$line_no++;
-			if (/\t/)
-			{
-				print STDERR "Tab found in $f:$line_no\n";
-				$errors++;
-			}
-			if (/\xC2\xA0/)
-			{
-				print STDERR "$f:$line_no: contains non-breaking space\n";
-				$errors++;
-			}
-		}
-		close($fh);
-	}
-
-	if ($errors)
-	{
-		die "Tabs and/or non-breaking spaces appear in SGML/XML files\n";
-	}
-}
-
-sub run_xmllint
-{
-	my $cmd = "$xmllint $xmlinclude --noout --valid postgres.sgml";
-	system($cmd) == 0 or die "xmllint validation failed\n";
-}
-
-run_xmllint();
-check_tabs_and_nbsp();
-- 
2.47.3