tools/make-man-index.py

   1 #!/usr/bin/env python3
   2 # SPDX-License-Identifier: LGPL-2.1-or-later
   3
   4 import collections
   5 import re
   6 import sys
   7
   8 from xml_helper import tree, xml_parse, xml_print
   9
  10 MDASH = ' — ' if sys.version_info.major >= 3 else ' -- '
  11
  12 TEMPLATE = '''\
  13 <refentry id="systemd.index">
  14
  15   <refentryinfo>
  16     <title>systemd.index</title>
  17     <productname>systemd</productname>
  18   </refentryinfo>
  19
  20   <refmeta>
  21     <refentrytitle>systemd.index</refentrytitle>
  22     <manvolnum>7</manvolnum>
  23   </refmeta>
  24
  25   <refnamediv>
  26     <refname>systemd.index</refname>
  27     <refpurpose>List all manpages from the systemd project</refpurpose>
  28   </refnamediv>
  29 </refentry>
  30 '''
  31
  32 SUMMARY = '''\
  33   <refsect1>
  34     <title>See Also</title>
  35     <para>
  36       <citerefentry><refentrytitle>systemd.directives</refentrytitle><manvolnum>7</manvolnum></citerefentry>
  37     </para>
  38
  39     <para id='counts' />
  40   </refsect1>
  41 '''
  42
  43 COUNTS = '\
  44 This index contains {count} entries, referring to {pages} individual manual pages.'
  45
  46
  47 def check_id(page, t):
  48     page_id = t.getroot().get('id')
  49     if not re.search('/' + page_id + '[.]', page.translate(str.maketrans('@', '_'))):
  50         raise ValueError(f"id='{page_id}' is not the same as page name '{page}'")
  51
  52 def make_index(pages):
  53     index = collections.defaultdict(list)
  54     for p in pages:
  55         t = xml_parse(p)
  56         check_id(p, t)
  57         section = t.find('./refmeta/manvolnum').text
  58         refname = t.find('./refnamediv/refname').text
  59         purpose_text = ' '.join(t.find('./refnamediv/refpurpose').itertext())
  60         purpose = ' '.join(purpose_text.split())
  61         for f in t.findall('./refnamediv/refname'):
  62             infos = (f.text, section, purpose, refname)
  63             index[f.text[0].upper()].append(infos)
  64     return index
  65
  66 def add_letter(template, letter, pages):
  67     refsect1 = tree.SubElement(template, 'refsect1')
  68     title = tree.SubElement(refsect1, 'title')
  69     title.text = letter
  70     para = tree.SubElement(refsect1, 'para')
  71     for info in sorted(pages, key=lambda info: str.lower(info[0])):
  72         refname, section, purpose, _realname = info
  73
  74         b = tree.SubElement(para, 'citerefentry')
  75         c = tree.SubElement(b, 'refentrytitle')
  76         c.text = refname
  77         d = tree.SubElement(b, 'manvolnum')
  78         d.text = section
  79
  80         b.tail = MDASH + purpose # + ' (' + p + ')'
  81
  82         tree.SubElement(para, 'sbr')
  83
  84 def add_summary(template, indexpages):
  85     count = 0
  86     pages = set()
  87     for group in indexpages:
  88         count += len(group)
  89         for info in group:
  90             _refname, section, _purpose, realname = info
  91             pages.add((realname, section))
  92
  93     refsect1 = tree.fromstring(SUMMARY)
  94     template.append(refsect1)
  95
  96     para = template.find(".//para[@id='counts']")
  97     para.text = COUNTS.format(count=count, pages=len(pages))
  98
  99 def make_page(*xml_files):
 100     template = tree.fromstring(TEMPLATE)
 101     index = make_index(xml_files)
 102
 103     for letter in sorted(index):
 104         add_letter(template, letter, index[letter])
 105
 106     add_summary(template, index.values())
 107
 108     return template
 109
 110 if __name__ == '__main__':
 111     with open(sys.argv[1], 'wb') as file:
 112         file.write(xml_print(make_page(*sys.argv[2:])))