tools/checkdocs.py

   1 #!/usr/bin/env python3
   2
   3 # The MIT License (MIT)
   4 #
   5 # Copyright (c) 2014 Philippe Proulx <eepp.ca>
   6 # Copyright (c) 2014 The LTTng Project <lttng.org>
   7 #
   8 # Permission is hereby granted, free of charge, to any person obtaining a copy
   9 # of this software and associated documentation files (the "Software"), to deal
  10 # in the Software without restriction, including without limitation the rights
  11 # to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
  12 # copies of the Software, and to permit persons to whom the Software is
  13 # furnished to do so, subject to the following conditions:
  14 #
  15 # The above copyright notice and this permission notice shall be included in
  16 # all copies or substantial portions of the Software.
  17 #
  18 # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  19 # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  20 # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
  21 # AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  22 # LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
  23 # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
  24 # THE SOFTWARE.
  25
  26 import re
  27 import os
  28 import sys
  29 from termcolor import colored
  30
  31
  32 TOC_PATH = 'toc/docs.yml'
  33 CONTENTS_ROOT_PATH = 'contents'
  34
  35
  36 def _perror(filename, msg):
  37     s = '{}  {} {}'.format(filename, colored('Error:', 'red'),
  38                            colored(msg, 'red', attrs=['bold']))
  39     print(s)
  40
  41
  42 def _pwarn(filename, msg):
  43     s = '{}  {} {}'.format(filename, colored('Warning:', 'yellow'),
  44                            colored(msg, 'yellow', attrs=['bold']))
  45     print(s)
  46
  47
  48 def _get_files(root):
  49     files = []
  50
  51     for dirpath, dirnames, filenames in os.walk(root):
  52         for f in filenames:
  53             files.append(os.path.join(dirpath, f))
  54
  55     return sorted(files)
  56
  57
  58 def _get_toc_ids(path):
  59     p = re.compile(r'id\s*:\s*(.+)$', flags=re.M)
  60
  61     with open(path) as f:
  62         orig_ids = p.findall(f.read())
  63
  64     ids = set(orig_ids)
  65
  66     if len(ids) != len(orig_ids):
  67         _perror(path, 'Duplicate IDs')
  68         return None
  69
  70     return ids
  71
  72
  73 def _check_file_links(toc_ids, path, c):
  74     ilinkp = re.compile(r'\[[^\]]+\]\(([^)]+)\)', flags=re.M)
  75     elinkp = re.compile(r'<a(?:\s+[^>]+|\s*)>')
  76
  77     ret = True
  78
  79     ilinks = ilinkp.findall(c)
  80     elinks = elinkp.findall(c)
  81
  82     for link in ilinks:
  83         if not link.startswith('#doc-'):
  84             s = 'Internal link does not start with "#doc-": "{}"'.format(link)
  85             _perror(path, s)
  86             ret = False
  87             continue
  88
  89         sid = link[5:]
  90
  91         if sid not in toc_ids:
  92             _perror(path, 'Dead internal link: "{}"'.format(link))
  93             ret = False
  94
  95     hrefp = re.compile(r'href="([^"]+)"')
  96     classesp = re.compile(r'class="([^"]+)"')
  97
  98     for link in elinks:
  99         href = hrefp.search(link)
 100         classes = classesp.search(link)
 101
 102         if classes is None:
 103             _pwarn(path, 'External link has no "ext" class: "{}"'.format(link))
 104             classes = []
 105         else:
 106             classes = classes.group(1).split(' ')
 107
 108             if 'int' not in classes and 'ext' not in classes:
 109                 _pwarn(path, 'External link has no "ext" class: "{}"'.format(link))
 110
 111         if href is not None:
 112             if href.group(1).startswith('#') and 'int' not in classes:
 113                 _pwarn(path, 'External link starts with #: "{}"'.format(href.group(1)))
 114         else:
 115             _perror(path, 'External link with no "href": "{}"'.format(link))
 116             ret = False
 117
 118     return ret
 119
 120
 121 def _check_contents(toc_ids, contents_files):
 122     ret = True
 123
 124     for path in contents_files:
 125         with open(path) as f:
 126             c = f.read()
 127
 128         ret &= _check_file_links(toc_ids, path, c)
 129
 130     return ret
 131
 132
 133 def _check_non_md(files):
 134     ret = True
 135
 136     for f in files:
 137         if not f.endswith('.md'):
 138             _perror(f, 'Wrong, non-Markdown file')
 139             ret = False
 140
 141     return ret
 142
 143
 144 def checkdocs():
 145     toc_ids = _get_toc_ids(TOC_PATH)
 146
 147     if toc_ids is None:
 148         return False
 149
 150     contents_files = _get_files(CONTENTS_ROOT_PATH)
 151
 152     if not _check_non_md(contents_files):
 153         return False
 154
 155     if not _check_contents(toc_ids, contents_files):
 156         return False
 157
 158     return True
 159
 160
 161 if __name__ == '__main__':
 162     sys.exit(0 if checkdocs() else 1)