From 068aaea8b5ac80809a1b860f63c625b3ea389ded Mon Sep 17 00:00:00 2001
From: Philip Hazel <ph10@hermes.cam.ac.uk>
Date: Thu, 10 Nov 2005 12:30:13 +0000
Subject: First pass update for 4.60 documentation.

---
 doc/doc-docbook/TidyHTML-spec | 48 +++++++++++++++++++++++++++++++++++++++----
 1 file changed, 44 insertions(+), 4 deletions(-)

(limited to 'doc/doc-docbook/TidyHTML-spec')

diff --git a/doc/doc-docbook/TidyHTML-spec b/doc/doc-docbook/TidyHTML-spec
index 8459bcf17..05a9d826b 100755
--- a/doc/doc-docbook/TidyHTML-spec
+++ b/doc/doc-docbook/TidyHTML-spec
@@ -1,6 +1,6 @@
 #! /usr/bin/perl
 
-# $Cambridge: exim/doc/doc-docbook/TidyHTML-spec,v 1.1 2005/06/16 10:32:31 ph10 Exp $
+# $Cambridge: exim/doc/doc-docbook/TidyHTML-spec,v 1.2 2005/11/10 12:30:13 ph10 Exp $
 
 # Script to tidy up the spec HTML files that are generated by xmlto. The
 # following changes are made:
@@ -9,6 +9,11 @@
 # 2. Create reverse links from chapter and section titles back to the TOC.
 # 3. Tidy the ix01.html file - the actual index - by splitting long lines.
 # 4. Insert links from the letter divisions to the top of the Index.
+# 5. Turn <div class="literallayout"><p> into <div class="literallayout"> and
+#    a matching </p></div> into </div> to get rid of unwanted vertical white
+#    space.
+# 6. Before each occurrence of </td> insert &nbsp; so that the table's cell
+#    is a little bit wider than the text itself.
 
 chdir "spec.html";
 
@@ -52,7 +57,10 @@ open (OUT, ">index.html") || die "Failed to open index.html for writing: $!\n";
 print OUT @toc;
 close(OUT);
 
-# Now scan each of the other page files and insert the reverse links.
+# Now scan each of the other page files and insert the reverse links. While
+# we are at it, we tidy up <div class="literallayout"> by removing unwanted
+# paragraph marks, which generate unwanted vertical space. We also insert
+# &nbsp; before </td> to push table cells apart from each other.
 
 foreach $file (@chlist)
   {
@@ -60,9 +68,26 @@ foreach $file (@chlist)
   @text = <IN>;
   close(IN);
 
+  # Insert a newline after certain elements, and split the lines so that each
+  # one is a separate element in the vector. This makes it easier to recognize
+  # these elements.
+
   foreach $line (@text)
     {
-    if ($line =~ /^(.*?)<a( xmlns="[^"]+")? id="([^"]+)"><\/a>(.+?)<\/h(.*)$/)
+    $line =~ s/<p>\s*(?!\n)/<p>\n/g;
+    $line =~ s/<\/p>\s*(?!\n)/<\/p>\n/g;
+    $line =~ s/<\/div>\s*(?!\n)/<\/div>\n/g;
+    $line =~ s/<div([^>]*)>(?!\n)/<div$1>\n/g;
+    }
+
+  for ($i = 0; $i < scalar(@text); $i++)
+    { splice @text, $i, 1, (split /(?<=\n)/, $text[$i]); }
+
+  $thisdiv = 0;
+
+  for ($i = 0; $i < scalar(@text); $i++)
+    {
+    if ($text[$i] =~ /^(.*?)<a( xmlns="[^"]+")? id="([^"]+)"><\/a>(.+?)<\/h(.*)$/)
       {
       my($pre, $opt, $id, $title, $post) = ($1, $2, $3, $4, $5);
 
@@ -73,7 +98,22 @@ foreach $file (@chlist)
       $ref = $backref{"$file"} if !defined $ref;
 
       # Adjust the line
-      $line = "$pre<a$opt href=\"index.html#$ref\" id=\"$id\">$title</a></h$post";
+      $text[$i]= "$pre<a$opt href=\"index.html#$ref\" id=\"$id\">$title</a></h$post";
+      }
+
+    elsif ($text[$i] eq "<div class=\"literallayout\">\n" && $text[$i+1] eq "<p>\n")
+      {
+      $text[++$i] = "";
+      $thisdiv = 1;
+      }
+    elsif ($thisdiv && $text[$i] eq "</p>\n" && $text[$i+1] eq "</div>\n")
+      {
+      $text[$i] = "";
+      $thisdiv = 0;
+      }
+    elsif ($text[$i] =~ /^\s*<\/td>/)
+      {
+      $text[$i] = "&nbsp;$text[$i]";
       }
     }
 
-- 
cgit v1.2.3