tools/make-man-index.py

   1 #!/usr/bin/env python3
   2 #  -*- Mode: python; coding: utf-8; indent-tabs-mode: nil -*- */
   3 # SPDX-License-Identifier: LGPL-2.1+
   4 #
   5 #  Copyright © 2013 Zbigniew Jędrzejewski-Szmek
   6
   7 import collections
   8 import sys
   9 import re
  10 from xml_helper import xml_parse, xml_print, tree
  11
  12 MDASH = ' — ' if sys.version_info.major >= 3 else ' -- '
  13
  14 TEMPLATE = '''\
  15 <refentry id="systemd.index" conditional="HAVE_PYTHON">
  16
  17   <refentryinfo>
  18     <title>systemd.index</title>
  19     <productname>systemd</productname>
  20
  21     <authorgroup>
  22       <author>
  23         <contrib>Developer</contrib>
  24         <firstname>Lennart</firstname>
  25         <surname>Poettering</surname>
  26         <email>lennart@poettering.net</email>
  27       </author>
  28     </authorgroup>
  29   </refentryinfo>
  30
  31   <refmeta>
  32     <refentrytitle>systemd.index</refentrytitle>
  33     <manvolnum>7</manvolnum>
  34   </refmeta>
  35
  36   <refnamediv>
  37     <refname>systemd.index</refname>
  38     <refpurpose>List all manpages from the systemd project</refpurpose>
  39   </refnamediv>
  40 </refentry>
  41 '''
  42
  43 SUMMARY = '''\
  44   <refsect1>
  45     <title>See Also</title>
  46     <para>
  47       <citerefentry><refentrytitle>systemd.directives</refentrytitle><manvolnum>7</manvolnum></citerefentry>
  48     </para>
  49
  50     <para id='counts' />
  51   </refsect1>
  52 '''
  53
  54 COUNTS = '\
  55 This index contains {count} entries, referring to {pages} individual manual pages.'
  56
  57
  58 def check_id(page, t):
  59     id = t.getroot().get('id')
  60     if not re.search('/' + id + '[.]', page):
  61         raise ValueError("id='{}' is not the same as page name '{}'".format(id, page))
  62
  63 def make_index(pages):
  64     index = collections.defaultdict(list)
  65     for p in pages:
  66         t = xml_parse(p)
  67         check_id(p, t)
  68         section = t.find('./refmeta/manvolnum').text
  69         refname = t.find('./refnamediv/refname').text
  70         purpose = ' '.join(t.find('./refnamediv/refpurpose').text.split())
  71         for f in t.findall('./refnamediv/refname'):
  72             infos = (f.text, section, purpose, refname)
  73             index[f.text[0].upper()].append(infos)
  74     return index
  75
  76 def add_letter(template, letter, pages):
  77     refsect1 = tree.SubElement(template, 'refsect1')
  78     title = tree.SubElement(refsect1, 'title')
  79     title.text = letter
  80     para = tree.SubElement(refsect1, 'para')
  81     for info in sorted(pages, key=lambda info: str.lower(info[0])):
  82         refname, section, purpose, realname = info
  83
  84         b = tree.SubElement(para, 'citerefentry')
  85         c = tree.SubElement(b, 'refentrytitle')
  86         c.text = refname
  87         d = tree.SubElement(b, 'manvolnum')
  88         d.text = section
  89
  90         b.tail = MDASH + purpose # + ' (' + p + ')'
  91
  92         tree.SubElement(para, 'sbr')
  93
  94 def add_summary(template, indexpages):
  95     count = 0
  96     pages = set()
  97     for group in indexpages:
  98         count += len(group)
  99         for info in group:
 100             refname, section, purpose, realname = info
 101             pages.add((realname, section))
 102
 103     refsect1 = tree.fromstring(SUMMARY)
 104     template.append(refsect1)
 105
 106     para = template.find(".//para[@id='counts']")
 107     para.text = COUNTS.format(count=count, pages=len(pages))
 108
 109 def make_page(*xml_files):
 110     template = tree.fromstring(TEMPLATE)
 111     index = make_index(xml_files)
 112
 113     for letter in sorted(index):
 114         add_letter(template, letter, index[letter])
 115
 116     add_summary(template, index.values())
 117
 118     return template
 119
 120 if __name__ == '__main__':
 121     with open(sys.argv[1], 'wb') as f:
 122         f.write(xml_print(make_page(*sys.argv[2:])))