1 #!/usr/bin/python 2 3 """ 4 HeaderID Extension for Python-Markdown 5 ====================================== 6 7 Adds ability to set HTML IDs for headers. 8 9 Basic usage: 10 11 >>> import markdown 12 >>> text = "# Some Header # {#some_id}" 13 >>> md = markdown.markdown(text, ['headerid']) 14 >>> md 15 u'<h1 id="some_id">Some Header</h1>' 16 17 All header IDs are unique: 18 19 >>> text = ''' 20 ... #Header 21 ... #Another Header {#header} 22 ... #Third Header {#header}''' 23 >>> md = markdown.markdown(text, ['headerid']) 24 >>> md 25 u'<h1 id="header">Header</h1>\\n<h1 id="header_1">Another Header</h1>\\n<h1 id="header_2">Third Header</h1>' 26 27 To fit within a html template's hierarchy, set the header base level: 28 29 >>> text = ''' 30 ... #Some Header 31 ... ## Next Level''' 32 >>> md = markdown.markdown(text, ['headerid(level=3)']) 33 >>> md 34 u'<h3 id="some_header">Some Header</h3>\\n<h4 id="next_level">Next Level</h4>' 35 36 Turn off auto generated IDs: 37 38 >>> text = ''' 39 ... # Some Header 40 ... # Header with ID # { #foo }''' 41 >>> md = markdown.markdown(text, ['headerid(forceid=False)']) 42 >>> md 43 u'<h1>Some Header</h1>\\n<h1 id="foo">Header with ID</h1>' 44 45 Use with MetaData extension: 46 47 >>> text = '''header_level: 2 48 ... header_forceid: Off 49 ... 50 ... # A Header''' 51 >>> md = markdown.markdown(text, ['headerid', 'meta']) 52 >>> md 53 u'<h2>A Header</h2>' 54 55 Copyright 2007-2008 [Waylan Limberg](http://achinghead.com/). 56 57 Project website: <http://www.freewisdom.org/project/python-markdown/HeaderId> 58 Contact: markdown@freewisdom.org 59 60 License: BSD (see ../docs/LICENSE for details) 61 62 Dependencies: 63 * [Python 2.3+](http://python.org) 64 * [Markdown 2.0+](http://www.freewisdom.org/projects/python-markdown/) 65 66 """ 67 68 import markdown 69 from markdown import etree 70 import re 71 from string import ascii_lowercase, digits, punctuation 72 73 ID_CHARS = ascii_lowercase + digits + '-_' 74 IDCOUNT_RE = re.compile(r'^(.*)_([0-9]+)$') 75 76 77 class HeaderIdProcessor(markdown.blockprocessors.BlockProcessor): 78 """ Replacement BlockProcessor for Header IDs. """ 79 80 # Detect a header at start of any line in block 81 RE = re.compile(r"""(^|\n) 82 (?P<level>\#{1,6}) # group('level') = string of hashes 83 (?P<header>.*?) # group('header') = Header text 84 \#* # optional closing hashes 85 (?:[ \t]*\{[ \t]*\#(?P<id>[-_:a-zA-Z0-9]+)[ \t]*\})? 86 (\n|$) # ^^ group('id') = id attribute 87 """, 88 re.VERBOSE) 89 90 IDs = [] 91 92 def test(self, parent, block): 93 return bool(self.RE.search(block)) 94 95 def run(self, parent, blocks): 96 block = blocks.pop(0) 97 m = self.RE.search(block) 98 if m: 99 before = block[:m.start()] # All lines before header 100 after = block[m.end():] # All lines after header 101 if before: 102 # As the header was not the first line of the block and the 103 # lines before the header must be parsed first, 104 # recursively parse this lines as a block. 105 self.parser.parseBlocks(parent, [before]) 106 # Create header using named groups from RE 107 start_level, force_id = self._get_meta() 108 level = len(m.group('level')) + start_level 109 if level > 6: 110 level = 6 111 h = markdown.etree.SubElement(parent, 'h%d' % level) 112 h.text = m.group('header').strip() 113 if m.group('id'): 114 h.set('id', self._unique_id(m.group('id'))) 115 elif force_id: 116 h.set('id', self._create_id(m.group('header').strip())) 117 if after: 118 # Insert remaining lines as first block for future parsing. 119 blocks.insert(0, after) 120 else: 121 # This should never happen, but just in case... 122 message(CRITICAL, "We've got a problem header!") 123 124 def _get_meta(self): 125 """ Return meta data suported by this ext as a tuple """ 126 level = int(self.config['level'][0]) - 1 127 force = self._str2bool(self.config['forceid'][0]) 128 if hasattr(self.md, 'Meta'): 129 if self.md.Meta.has_key('header_level'): 130 level = int(self.md.Meta['header_level'][0]) - 1 131 if self.md.Meta.has_key('header_forceid'): 132 force = self._str2bool(self.md.Meta['header_forceid'][0]) 133 return level, force 134 135 def _str2bool(self, s, default=False): 136 """ Convert a string to a booleen value. """ 137 s = str(s) 138 if s.lower() in ['0', 'f', 'false', 'off', 'no', 'n']: 139 return False 140 elif s.lower() in ['1', 't', 'true', 'on', 'yes', 'y']: 141 return True 142 return default 143 144 def _unique_id(self, id): 145 """ Ensure ID is unique. Append '_1', '_2'... if not """ 146 while id in self.IDs: 147 m = IDCOUNT_RE.match(id) 148 if m: 149 id = '%s_%d'% (m.group(1), int(m.group(2))+1) 150 else: 151 id = '%s_%d'% (id, 1) 152 self.IDs.append(id) 153 return id 154 155 def _create_id(self, header): 156 """ Return ID from Header text. """ 157 h = '' 158 for c in header.lower().replace(' ', '_'): 159 if c in ID_CHARS: 160 h += c 161 elif c not in punctuation: 162 h += '+' 163 return self._unique_id(h) 164 165 166 class HeaderIdExtension (markdown.Extension): 167 def __init__(self, configs): 168 # set defaults 169 self.config = { 170 'level' : ['1', 'Base level for headers.'], 171 'forceid' : ['True', 'Force all headers to have an id.'] 172 } 173 174 for key, value in configs: 175 self.setConfig(key, value) 176 177 def extendMarkdown(self, md, md_globals): 178 md.registerExtension(self) 179 self.processor = HeaderIdProcessor(md.parser) 180 self.processor.md = md 181 self.processor.config = self.config 182 # Replace existing hasheader in place. 183 md.parser.blockprocessors['hashheader'] = self.processor 184 185 def reset(self): 186 self.processor.IDs = [] 187 188 189 def makeExtension(configs=None): 190 return HeaderIdExtension(configs=configs) 191 192 if __name__ == "__main__": 193 import doctest 194 doctest.testmod() 195 196