From 4c14e3181ca416f19c83c4b3489377bcdd55460f Mon Sep 17 00:00:00 2001
From: John Mandereau <john.mandereau@gmail.com>
Date: Sat, 22 Nov 2008 20:04:38 +0100
Subject: [PATCH] Docs: hack HTML gettextization again

Section titles which contain @-commands are still not translated in
<title>, because there is no way to guess them easily in HTML output;
this will remain broken until we drop 'makeinfo --html' support and
translate node names only in the source.
---
 buildscripts/html-gettext.py | 70 +++++++++++++++++++-----------------
 1 file changed, 38 insertions(+), 32 deletions(-)

diff --git a/buildscripts/html-gettext.py b/buildscripts/html-gettext.py
index aaa8888a3f..6920e50d22 100644
--- a/buildscripts/html-gettext.py
+++ b/buildscripts/html-gettext.py
@@ -29,37 +29,42 @@ my_gettext = langdefs.translation[lang]
 html_codes = ((' -- ', ' &ndash; '),
               (' --- ', ' &mdash; '),
               ("'", '&rsquo;'))
-html2texi = {'command':
-                 (re.compile (r'<samp><span class="command">(.*?)</span></samp>'),
-                  r'@command{\1}'),
-             'code':
-                 (re.compile (r'<code>(.*?)</code>'),
-                  r'@code{\1}')
-             }
-texi2html = {'command':
-                 (re.compile (r'@command{(.*?)}'),
-                  r'<samp><span class="command">\1</span></samp>'),
-             'code':
-                 (re.compile (r'@code{(.*?)}'),
-                  r'<code>\1</code>')
-             }
+texi_html_conversion = {
+    'command': {
+        'html2texi':
+            (re.compile (r'(?:<samp><span class="command">|<code>)(.*?)(?:</span></samp>|</code>)'),
+             r'@command{\1}'),
+        'texi2html':
+            (re.compile (r'@command{(.*?)}'),
+             r'<code>\1</code>'),
+        },
+    'code': {
+        'html2texi':
+            (re.compile (r'<code>(.*?)</code>'),
+             r'@code{\1}'),
+        'texi2html':
+            (re.compile (r'@code{(.*?)}'),
+             r'<code>\1</code>'),
+        },
+    }
+
 whitespaces = re.compile (r'\s+')
 
 
 def _ (s):
     if not s:
         return ''
-    s = whitespaces.sub (' ', s)
+    str = whitespaces.sub (' ', s)
     for c in html_codes:
-        s = s.replace (c[1], c[0])
-    for u in html2texi.values():
-        s = u[0].sub (u[1], s)
-    s = my_gettext (s)
-    for u in texi2html.values():
-        s = u[0].sub (u[1], s)
+        str = str.replace (c[1], c[0])
+    for command in texi_html_conversion:
+        d = texi_html_conversion[command]
+        str = d['html2texi'][0].sub (d['html2texi'][1], str)
+        str = my_gettext (str)
+        str = d['texi2html'][0].sub (d['texi2html'][1], str)
     for c in html_codes:
-        s = s.replace (c[0], c[1])
-    return s
+        str = str.replace (c[0], c[1])
+    return str
 
 link_re =  re.compile (r'<link rel="(up|prev|next)" (.*?) title="([^"]*?)">')
 
@@ -72,27 +77,28 @@ makeinfo_title_re = re.compile (r'<title>([^<]*?) - ([^<]*?)</title>')
 def makeinfo_title_gettext (m):
     return '<title>' + _ (m.group (1)) + ' - ' + m.group (2) + '</title>'
 
-texi2html_title_re = re.compile (r'<title>(.+?): ([A-Z\d.]+ |)(.+?)</title>')
+texi2html_title_re = re.compile (r'<title>(.+): ([A-Z\d.]+ |)(.+?)</title>')
 
 def texi2html_title_gettext (m):
     return '<title>' + _ (m.group (1)) + double_punct_char_separator + ': ' \
         + m.group (2) + _ (m.group (3)) + '</title>'
 
-a_href_re = re.compile ('(?s)<a ([^>]*?href="[\\w.#-_]+"[^>]*>(?:<code>|))\
-(Appendix |)([A-Z0-9.]+ | (?:&lt;){1,2} |&nbsp;[^:<]+?:&nbsp;|&nbsp;|)\
-(.+?)(</code>| (?:&gt;){1,2} |&nbsp;|)</a>:?')
+a_href_re = re.compile ('(?s)<a (?P<attributes>[^>]*?href="[\\w.#-_]+"[^>]*>)(?P<code><code>)?\
+(?P<appendix>Appendix )?(?P<leading>[A-Z0-9.]+ | (?:&lt;){1,2} |&nbsp;[^:]+?:&nbsp;|&nbsp;|)\
+(?P<name>.+?)(?P<end_code>(?(code)</code>|))(?P<trailing>| (?:&gt;){1,2} |&nbsp;|)</a>:?')
 
 def a_href_gettext (m):
     s = ''
     if m.group(0)[-1] == ':':
         s = double_punct_char_separator + ':'
     t = ''
-    if m.group (2):
-        t = _ (m.group (2))
-    return '<a ' + m.group (1) + t + m.group (3) + _ (m.group (4)) + \
-        m.group (5) + '</a>' + s
+    if m.group ('appendix'):
+        t = _ (m.group ('appendix'))
+    return '<a ' + m.group ('attributes') + (m.group ('code') or '') + \
+        t + m.group ('leading') + _ (m.group ('name')) + \
+        m.group ('end_code') + m.group ('trailing') + '</a>' + s
 
-h_re = re.compile (r'<h(\d)( class="\w+"|)>\s*(Appendix |)([A-Z\d.]+ |)?([^<]+)\s*</h\1>')
+h_re = re.compile (r'<h(\d)( class="\w+"|)>\s*(Appendix |)([A-Z\d.]+ |)(.+?)\s*</h\1>')
 
 def h_gettext (m):
     if m.group (3):
-- 
2.39.5