# NODE\tFILENAME\tANCHOR
# LANG is the document language in case it's not 'en'
# Note: The filename does not have any extension appended!
-# This file can then be used by our texi2html init script to determine
+# This file should then be used by our texi2html init script to determine
# the correct file name and anchor for external refs
+# For translated documentation: cross-references to nodes that exist
+# only in documentation in English are allowed, that's why the already
+# generated map file of docs in English is loaded with
+# --master-map-file option, then the node names that are defined in
+# the map for the manual in English but not in the translated manual
+# are added to the map for the translated manual.
+
+
import sys
import re
import os
import getopt
-options_list, files = getopt.getopt (sys.argv[1:],'o:s:hI:m:',
+options_list, files = getopt.getopt (sys.argv[1:],'o:s:hI:m:k:q',
['output=', 'split=',
'help', 'include=',
- 'master-map-file='])
+ 'master-map-file=',
+ 'known-missing-files=',
+ 'quiet'])
help_text = r"""Usage: %(program_name)s [OPTIONS]... TEXIFILE...
Extract files names for texinfo (sub)sections from the texinfo files.
-o, --output=DIRECTORY write .xref-map files to DIRECTORY
-s, --split=MODE split manual according to MODE. Possible values
are section and custom (default)
+ -k, --known-missing-files a filename which has a list of files known
+ to be missing for this make
+ -q, --quiet suppress most messages
"""
def help (text):
split = "custom"
include_path = ['.',]
master_map_file = ''
+known_missing_files = []
+known_missing_files_file = ''
+docs_without_directories = ['changes', 'music-glossary']
+suppress_output = False
initial_map = {}
for opt in options_list:
o = opt[0]
if os.path.isdir (a):
include_path.append (a)
else:
- print 'NOT A DIR from: ', os.getcwd (), a
+ path_list = a.split('/')
+ file_name = path_list[len(path_list)-1]
+ if not (file_name in docs_without_directories):
+ print a, 'is not a directory.'
+ print 'Please consider adding it to the list of '
+ print 'known missing files in extract_texi_filename.py.'
elif o == '-o' or o == '--output':
outdir = a
elif o == '-s' or o == '--split':
elif o == '-m' or o == '--master-map-file':
if os.path.isfile (a):
master_map_file = a
+ elif o == '--known-missing-files':
+ if os.path.isfile (a):
+ known_missing_files_file = a
+ else:
+ print 'Missing files list file not found: ', a
+ elif o == '-q' or o == '--quiet':
+ suppress_output = True
else:
raise Exception ('unknown option: ' + o)
+if known_missing_files_file:
+ missing_files = open (known_missing_files_file, 'r')
+ known_missing_files = missing_files.read().splitlines()
+ missing_files.close()
if not os.path.isdir (outdir):
if os.path.exists (outdir):
os.unlink (outdir)
os.makedirs (outdir)
-include_re = re.compile (r'@include ((?!../lily-).*?\.i?texi)$', re.M)
+# Only look at @include if it is not preceeded by a @c:
+include_re = re.compile (r'^(?!.*@c .*@include)@include ((?!../lily-).*?\.i?te(xi|ly))$', re.M)
whitespaces = re.compile (r'\s+')
section_translation_re = re.compile ('^@(node|(?:unnumbered|appendix)\
(?:(?:sub){0,2}sec)?|top|chapter|(?:sub){0,2}section|\
-(?:major|chap|(?:sub){0,2})heading|lydoctitle|translationof) \
+(?:major|chap|(?:sub){0,2})heading|lydoctitle|translationof|nodeprefix) \
(.+)$', re.MULTILINE)
external_node_re = re.compile (r'\s+@c\s+external.*')
filepath = os.path.join (directory, include_name)
if os.path.exists (filepath):
return extract_sections (filepath)[1]
- print 'No such file: ' + include_name
- print 'Search path: ' + ':'.join (include_path)
+ if not (include_name in known_missing_files):
+ # Not found
+ print 'Warning: No such file: ' + include_name + \
+ ' (search path: ' + ':'.join (include_path)+')'
return ''
lang_re = re.compile (r'^@documentlanguage (.+)', re.M)
result = 't_g' + result
return result
-texinfo_re = re.compile (r'@.*{(.*)}')
+texinfo_re = re.compile (r'@.*?{(.*?)}')
def remove_texinfo (title):
- return texinfo_re.sub (r'\1', title)
+ title = title.replace ('--', '-')
+ return texinfo_re.sub (r'\1', title).strip ()
def create_texinfo_anchor (title):
return texinfo_file_name (remove_texinfo (title))
sections = section_translation_re.findall (page)
basename = os.path.splitext (os.path.basename (filename))[0]
p = os.path.join (outdir, basename) + lang_suffix + '.xref-map'
+ if not suppress_output:
+ print 'writing:', p
f = open (p, 'w')
+ node_prefix_title = ''
this_title = ''
this_filename = 'index'
this_anchor = ''
this_filename = anchor
elif original_node in initial_map:
this_filename = initial_map[original_node][2]
+ elif sec[0] == "nodeprefix":
+ node_prefix_title = remove_texinfo (sec[1])
+ node_prefix_anchor = create_texinfo_anchor (sec[1])
else:
# Some pages might not use a node for every section, so
# treat this case here, too: If we already had a section
# and encounter another one before the next @node, we
# write out the old one and start with the new values
- if had_section and this_title:
+ if had_section and split != 'node' and this_title:
f.write (this_title + "\t" + this_filename + "\t" + this_anchor + "\n")
this_title = remove_texinfo (sec[1])
this_anchor = create_texinfo_anchor (sec[1])
had_section = True
+ if sec[0] == "lydoctitle" and node_prefix_title:
+ this_title = "%s: %s" % (node_prefix_title, this_title)
+ this_anchor = "%s-%s" % (node_prefix_anchor, this_anchor)
+
if split == 'custom':
# unnumbered nodes use the previously used file name,
# only numbered nodes get their own filename! However,
initial_map[m.group (1)] = (m.group (1), m.group (2), m.group (3))
for filename in files:
- print "extract_texi_filenames.py: Processing %s" % filename
+ if not suppress_output:
+ print "extract_texi_filenames.py: Processing %s" % filename
(lang_suffix, sections) = extract_sections (filename)
process_sections (filename, lang_suffix, sections)