From bcc437814eb90c60120d0e7a9c1ece85bf66aea3 Mon Sep 17 00:00:00 2001 From: Waylan Limberg Date: Fri, 2 May 2008 22:26:41 -0400 Subject: Updated tests in wikilink extension to match current markdown (use convert, unicode output and trimmed whitespace) and added unobstrusive meta-data support. --- mdx_wikilink.py | 70 ++++++++++++++++++++++++++++++++++++++++----------------- 1 file changed, 50 insertions(+), 20 deletions(-) diff --git a/mdx_wikilink.py b/mdx_wikilink.py index 4279c1a..46d6c3f 100644 --- a/mdx_wikilink.py +++ b/mdx_wikilink.py @@ -12,7 +12,7 @@ Basic usage: >>> text = "Some text with a WikiLink." >>> md = markdown.markdown(text, ['wikilink']) >>> md - '\\n

Some text with a WikiLink.\\n

\\n\\n\\n' + u'

Some text with a WikiLink.\\n

' To define custom settings the simple way: @@ -20,20 +20,35 @@ To define custom settings the simple way: ... ['wikilink(base_url=/wiki/,end_url=.html,html_class=foo)'] ... ) >>> md - '\\n

Some text with a WikiLink.\\n

\\n\\n\\n' + u'

Some text with a WikiLink.\\n

' Custom settings the complex way: - >>> md = markdown.Markdown(text, + >>> md = markdown.Markdown( ... extensions = ['wikilink'], ... extension_configs = {'wikilink': [ ... ('base_url', 'http://example.com/'), ... ('end_url', '.html'), ... ('html_class', '') ]}, - ... encoding='utf8', ... safe_mode = True) - >>> str(md) - '\\n

Some text with a WikiLink.\\n

\\n\\n\\n' + >>> md.convert(text) + u'

Some text with a WikiLink.\\n

' + +Use MetaData with mdx_meta.py (Note the blank html_class in MetaData): + + >>> text = """wiki_base_url: http://example.com/ + ... wiki_end_url: .html + ... wiki_html_class: + ... + ... Some text with a WikiLink.""" + >>> md = markdown.Markdown(extensions=['meta', 'wikilink']) + >>> md.convert(text) + u'

Some text with a WikiLink.\\n

' + +MetaData should not carry over to next document: + + >>> md.convert("No MetaData here.") + u'

No MetaData here.\\n

' From the command line: @@ -46,13 +61,11 @@ Contact: waylan [at] gmail [dot] com License: [BSD](http://www.opensource.org/licenses/bsd-license.php) -Version: 0.4 (Oct 14, 2006) +Version: 0.6 (May 2, 2008) Dependencies: * [Python 2.3+](http://python.org) * [Markdown 1.6+](http://www.freewisdom.org/projects/python-markdown/) -* For older dependencies use [WikiLink Version 0.3] -(http://code.limberg.name/svn/projects/py-markdown-ext/wikilinks/tags/release-0.3/) ''' import markdown @@ -68,16 +81,16 @@ class WikiLinkExtension (markdown.Extension) : # Override defaults with user settings for key, value in configs : - # self.config[key][0] = value self.setConfig(key, value) def extendMarkdown(self, md, md_globals): self.md = md - #md.registerExtension(self) #??? # append to end of inline patterns WIKILINK_RE = r'''(?P\\|\b)(?P([A-Z]+[a-z-_]+){2,})\b''' - md.inlinePatterns.append(WikiLinks(WIKILINK_RE, self.config)) + WIKILINK_PATTERN = WikiLinks(WIKILINK_RE, self.config) + WIKILINK_PATTERN.md = md + md.inlinePatterns.append(WIKILINK_PATTERN) class WikiLinks (markdown.BasePattern) : def __init__(self, pattern, config): @@ -87,19 +100,36 @@ class WikiLinks (markdown.BasePattern) : def handleMatch(self, m, doc) : if m.group('escape') == '\\': a = doc.createTextNode(m.group('camelcase')) - else : - url = '%s%s%s'% (self.config['base_url'][0], m.group('camelcase'), self.config['end_url'][0]) + else: + base_url, end_url, html_class = self._getMeta() + url = '%s%s%s'% (base_url, m.group('camelcase'), end_url) label = m.group('camelcase').replace('_', ' ') a = doc.createElement('a') a.appendChild(doc.createTextNode(label)) a.setAttribute('href', url) - if self.config['html_class'][0] : - a.setAttribute('class', self.config['html_class'][0]) + if html_class: + a.setAttribute('class', html_class) return a - + + def _getMeta(self): + """ Return meta data or config data. """ + base_url = self.config['base_url'][0] + end_url = self.config['end_url'][0] + html_class = self.config['html_class'][0] + if hasattr(self.md, 'Meta'): + if self.md.Meta.has_key('wiki_base_url'): + base_url = self.md.Meta['wiki_base_url'][0] + if self.md.Meta.has_key('wiki_end_url'): + end_url = self.md.Meta['wiki_end_url'][0] + if self.md.Meta.has_key('wiki_html_class'): + html_class = self.md.Meta['wiki_html_class'][0] + return base_url, end_url, html_class + + def makeExtension(configs=None) : return WikiLinkExtension(configs=configs) -if __name__ == "__main__": - import doctest - doctest.testmod() \ No newline at end of file +if __name__ == "__main__": + import doctest + doctest.testmod() + -- cgit v1.2.3 From 180248415c13911628708d5bcd21674609d4fe0d Mon Sep 17 00:00:00 2001 From: Waylan Limberg Date: Fri, 2 May 2008 22:35:12 -0400 Subject: Cleaned up doc strings and greatly simplified the codehilite extension. Removed support or enscript and dp highlighing engines - only pygments is supported. --- mdx_codehilite.py | 328 ++++++++++++++++++++++-------------------------------- 1 file changed, 136 insertions(+), 192 deletions(-) diff --git a/mdx_codehilite.py b/mdx_codehilite.py index ea83e9a..6f81598 100644 --- a/mdx_codehilite.py +++ b/mdx_codehilite.py @@ -1,197 +1,141 @@ #!/usr/bin/python + +""" +CodeHilite Extension for Python-Markdown +======================================= + +Adds code/syntax highlighting to standard Python-Markdown code blocks. + +By [Waylan Limberg](http://achinghead.com/). + +Project website: http://achinghead.com/markdown-wikilinks/ +Contact: waylan [at] gmail [dot] com + +License: [BSD](http://www.opensource.org/licenses/bsd-license.php) + +Version: 0.2 (April 30, 2008) + +Dependencies: +* [Python 2.3+](http://python.org/) +* [Markdown 1.7+](http://www.freewisdom.org/projects/python-markdown/) +* [Pygments](http://pygments.org/) + +""" + import markdown # --------------- CONSTANTS YOU MIGHT WANT TO MODIFY ----------------- -DEFAULT_HILITER = 'pygments' # one of 'enscript', 'dp', or 'pygments' try: TAB_LENGTH = markdown.TAB_LENGTH except AttributeError: TAB_LENGTH = 4 -# --------------- THE CODE ------------------------------------------- -# --------------- hiliter utility functions -------------------------- -def escape(txt) : - '''basic html escaping''' - txt = txt.replace('&', '&') - txt = txt.replace('<', '<') - txt = txt.replace('>', '>') - txt = txt.replace('"', '"') - return txt -def number(txt): - '''use
    for line numbering''' - # Fix Whitespace - txt = txt.replace('\t', ' '*TAB_LENGTH) - txt = txt.replace(" "*4, "    ") - txt = txt.replace(" "*3, "   ") - txt = txt.replace(" "*2, "  ") - - # Add line numbers - lines = txt.splitlines() - txt = '
      \n'
-    for line in lines:
-        txt += '\t
    1. %s
      2. \n'% line
-    txt += '
    \n' - return txt - -# ---------------- The hiliters --------------------------------------- -def enscript(src, lang=None, num=True): - ''' -Pass source code on to [enscript] (http://www.codento.com/people/mtr/genscript/) -command line utility for hiliting. - -Usage: - >>> enscript(src [, lang [, num ]] ) - - @param src: Can be a string or any object with a .readline attribute. - - @param lang: The language of code. Basic escaping only, if None. +# ------------------ The Main CodeHilite Class ---------------------- +class CodeHilite: + """ + Determine language of source code, and pass it into the pygments hilighter. - @param num: (Boolen) Turns line numbering 'on' or 'off' (on by default). + Basic Usage: + >>> code = CodeHilite(src = text) + >>> html = code.hilite() + + * src: Can be a string or any object with a .readline attribute. - @returns : A string of html. - ''' - if lang: - cmd = 'enscript --highlight=%s --color --language=html --tabsize=%d --output=-'% (lang, TAB_LENGTH) - from os import popen3 - (i, out, err) = popen3(cmd) - i.write(src) - i.close() - # check for errors - e = err.read() - if e != 'output left in -\n' : - # error - just escape - txt = escape(src) - else : - import re - pattern = re.compile(r'
    (?P.*?)
    ', re.DOTALL) - txt = pattern.search(out.read()).group('code') - # fix enscripts output - txt = txt.replace('\n', '\n').strip() - html_map = {'' : '', - '' : '', - '' : '', - '' : '', - '' : '' - } - for k, v in html_map.items() : - txt = txt.replace(k, v) - else: - txt = escape(src) - if num : - txt = number(txt) - else : - txt = '
    %s
    \n'% txt - return txt - - -def dp(src, lang=None, num=True): - ''' -Pass source code to a textarea for the [dp.SyntaxHighlighter] (http://www.dreamprojections.com/syntaxhighlighter/Default.aspx) - -Usage: - >>> dp(src [, lang [, num ]] ) - - @param src: A string. - - @param lang: The language of code. Undefined if None. - - @param num: (Boolen) Turns line numbering 'on' or 'off' (on by default). + * linenos: (Boolen) Turns line numbering 'on' or 'off' (off by default). - @returns : A string of html. - ''' - gutter = '' - if not num: - gutter = ':nogutter' - if not lang: - lang = '' - - return '
    \n'% (lang, gutter, src) + Low Level Usage: + >>> code = CodeHilite() + >>> code.src = text # String or anything with a .readline attribute + >>> code.linenos = True # True or False; Turns line numbering on or of. + >>> html = code.hilite() -def pygment(src, lang = None, num = True): - ''' -Pass code to the [Pygments](http://pygments.pocoo.org/) highliter with -optional line numbers. The output should then be styled with css to your liking. -No styles are applied by default - only styling hooks (i.e.: ). + """ -Usage: - >>> pygment(src [, lang [, num ]] ) + def __init__(self, src=None, linenos = False): + self.src = src + self.lang = None + self.linenos = linenos - @param src: Can be a string or any object with a .readline attribute. + def hilite(self): + """ + Pass code to the [Pygments](http://pygments.pocoo.org/) highliter with + optional line numbers. The output should then be styled with css to + your liking. No styles are applied by default - only styling hooks + (i.e.: ). - @param lang: The language of code. Pygments will try to guess language if None. + returns : A string of html. + + """ + + self.src = self.src.strip('\n') + + self._getLang() - @param num: (Boolen) Turns line numbering 'on' or 'off' (on by default). - - @returns : A string of html. - ''' - try: - from pygments import highlight - from pygments.lexers import get_lexer_by_name, guess_lexer, TextLexer - from pygments.formatters import HtmlFormatter - except ImportError: - # just escape and pass through - txt = escape(src) - if num: - txt = number(txt) - else : - txt = '
    %s
    \n'% txt - return txt - else: try: - lexer = get_lexer_by_name(lang) - except ValueError: + from pygments import highlight + from pygments.lexers import get_lexer_by_name, guess_lexer, TextLexer + from pygments.formatters import HtmlFormatter + except ImportError: + # just escape and pass through + txt = self._escape(self.src) + if num: + txt = self._number(txt) + else : + txt = '
    %s
    \n'% txt + return txt + else: try: - lexer = guess_lexer(src) + lexer = get_lexer_by_name(self.lang) except ValueError: - lexer = TextLexer() - formatter = HtmlFormatter(linenos=num, cssclass="codehilite") - return highlight(src, lexer, formatter) - - -# ------------------ The Main CodeHilite Class ---------------------- -class CodeHilite: - ''' -A wrapper class providing a single API for various hilighting engines. Takes source code, determines which language it containes (if not provided), and passes it into the hiliter specified. - -Basic Usage: - >>> code = CodeHilite(src = text) - >>> html = code.hilite() - - @param src: Can be a string or any object with a .readline attribute. - - @param lang: A string. Accepted values determined by hiliter used. Overrides _getLang() - - @param linenos: (Boolen) Turns line numbering 'on' or 'off' (off by default). - - @param hiliter: A string. One of 'enscript', 'dp', or 'pygments'. + try: + lexer = guess_lexer(self.src) + except ValueError: + lexer = TextLexer() + formatter = HtmlFormatter(linenos=self.linenos, cssclass="codehilite") + return highlight(self.src, lexer, formatter) + + def _escape(self, txt): + """ basic html escaping """ + txt = txt.replace('&', '&') + txt = txt.replace('<', '<') + txt = txt.replace('>', '>') + txt = txt.replace('"', '"') + return txt -Low Level Usage: - >>> code = CodeHilite() - >>> code.src = text # Can be a string or any object with a .readline attribute. - >>> code.lang = 'python' # Setting this will override _getLang() - >>> code.linenos = True # True or False; Turns line numbering on or off. - >>> code.hiliter = MyCustomHiliter # Where MyCustomHiliter is callable, takes three arguments (src, lang, linenos) and returns a string. - >>> html = code.hilite() - ''' - def __init__(self, src=None, lang=None, linenos = False, hiliter=DEFAULT_HILITER): - self.src = src - self.lang = lang - self.linenos = linenos - # map of highlighters - hl_map = { 'enscript' : enscript, 'dp' : dp, 'pygments' : pygment } - try : - self.hiliter = hl_map[hiliter] - except KeyError: - raise "Please provide a valid hiliter as a string. One of 'enscript', 'dp', or 'pygments'" + def _number(self, txt): + """ Use
      for line numbering """ + # Fix Whitespace + txt = txt.replace('\t', ' '*TAB_LENGTH) + txt = txt.replace(" "*4, "    ") + txt = txt.replace(" "*3, "   ") + txt = txt.replace(" "*2, "  ") + + # Add line numbers + lines = txt.splitlines() + txt = '
        \n'
+        for line in lines:
+            txt += '\t
      1. %s
        2. \n'% line
+        txt += '
      \n' + return txt def _getLang(self): - ''' -Determines language of a code block from shebang lines and whether said line should be removed or left in place. If the sheband line contains a path (even a single /) then it is assumed to be a real shebang lines and left alone. However, if no path is given (e.i.: #!python or :::python) then it is assumed to be a mock shebang for language identifitation of a code fragment and removed from the code block prior to processing for code highlighting. When a mock shebang (e.i: #!python) is found, line numbering is turned on. When colons are found in place of a shebang (e.i.: :::python), line numbering is left in the current state - off by default. - ''' + """ + Determines language of a code block from shebang lines and whether said + line should be removed or left in place. If the sheband line contains a + path (even a single /) then it is assumed to be a real shebang lines and + left alone. However, if no path is given (e.i.: #!python or :::python) + then it is assumed to be a mock shebang for language identifitation of a + code fragment and removed from the code block prior to processing for + code highlighting. When a mock shebang (e.i: #!python) is found, line + numbering is turned on. When colons are found in place of a shebang + (e.i.: :::python), line numbering is left in the current state - off + by default. + + """ + import re #split text into lines @@ -224,47 +168,47 @@ Determines language of a code block from shebang lines and whether said line sho self.src = "\n".join(lines).strip("\n") - def hilite(self): - '''The wrapper function which brings it all togeather''' - self.src = self.src.strip('\n') - - if not self.lang : self._getLang() - - return self.hiliter(self.src, self.lang, self.linenos) -# ------------------ The Markdown Extention ------------------------------- -class CodeHiliteExtention (markdown.Extension) : +# ------------------ The Markdown Extension ------------------------------- +class CodeHiliteExtention(markdown.Extension): def __init__(self, configs): # define default configs - self.config = {'hiliter' : [DEFAULT_HILITER, "one of 'enscript', 'dp', or 'pygments'"], - 'force_linenos' : [False, "Force line numbers - Default: False"] } + self.config = { + 'force_linenos' : [False, "Force line numbers - Default: False"] + } # Override defaults with user settings - for key, value in configs : + for key, value in configs: # self.config[key][0] = value self.setConfig(key, value) - def extendMarkdown(self, md, md_globals) : + def extendMarkdown(self, md, md_globals): def _hiliteCodeBlock(parent_elem, lines, inList): - """Overrides function of same name in standard Markdown class and - sends code blocks to a code highlighting proccessor. The result - is then stored in the HtmlStash, a placeholder is inserted into - the dom and the remainder of the text file is processed recursively. + """ + Overrides `_processCodeBlock` method in standard Markdown class + and sends code blocks to a code highlighting proccessor. The result + is then stored in the HtmlStash, a placeholder is inserted into + the dom and the remainder of the text file is processed recursively. + + * parent_elem: DOM element to which the content will be added + * lines: a list of lines + * inList: a level + + returns: None + + """ - @param parent_elem: DOM element to which the content will be added - @param lines: a list of lines - @param inList: a level - @returns: None""" detabbed, theRest = md.blockGuru.detectTabbed(lines) text = "\n".join(detabbed).rstrip()+"\n" - code = CodeHilite(text, hiliter=self.config['hiliter'][0], linenos=self.config['force_linenos'][0]) - placeholder = md.htmlStash.store(code.hilite()) + code = CodeHilite(text, linenos=self.config['force_linenos'][0]) + placeholder = md.htmlStash.store(code.hilite(), safe=True) parent_elem.appendChild(md.doc.createTextNode(placeholder)) md._processSection(parent_elem, theRest, inList) md._processCodeBlock = _hiliteCodeBlock -def makeExtension(configs=None) : +def makeExtension(configs={}): return CodeHiliteExtention(configs=configs) + -- cgit v1.2.3 From 19ea0eedcbca5189df9cf9f6ef658eade79abd79 Mon Sep 17 00:00:00 2001 From: Waylan Limberg Date: Mon, 12 May 2008 00:00:11 -0400 Subject: Added fenced_code extension and .gitignore file. --- .gitignore | 1 + mdx_fenced_code.py | 105 +++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 106 insertions(+) create mode 100644 .gitignore create mode 100644 mdx_fenced_code.py diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..0d20b64 --- /dev/null +++ b/.gitignore @@ -0,0 +1 @@ +*.pyc diff --git a/mdx_fenced_code.py b/mdx_fenced_code.py new file mode 100644 index 0000000..4b0e406 --- /dev/null +++ b/mdx_fenced_code.py @@ -0,0 +1,105 @@ +#!/usr/bin/env python + +""" +Fenced Code Extension for Python Markdown +========================================= + +This extension adds Fenced Code Blocks to Python-Markdown. + + >>> import markdown + >>> text = ''' + ... A paragraph before a fenced code block: + ... + ... ~~~ + ... Fenced code block + ... ~~~ + ... ''' + >>> html = markdown.markdown(text, extensions=['fenced_code']) + >>> html + u'

      A paragraph before a fenced code block:\\n

      \\n
      Fenced code block\\n
      '
+
+Works with safe_mode also (we check this because we are using the HtmlStash):
+
+    >>> markdown.markdown(text, extensions=['fenced_code'], safe_mode='replace')
+    u'
      A paragraph before a fenced code block:\\n
      \\n
      Fenced code block\\n
      '
+    
+Include tilde's in a code block and wrap with blank lines:
+
+    >>> text = '''
+    ... ~~~~~~~~
+    ... 
+    ... ~~~~
+    ... 
+    ... ~~~~~~~~'''
+    >>> markdown.markdown(text, extensions=['fenced_code'])
+    u'
      \\n~~~~\\n\\n
      '
+
+Multiple blocks and language tags:
+
+    >>> text = '''
+    ... ~~~~
+    ... block one
+    ... ~~~~{.python}
+    ... 
+    ... ~~~~
+    ... 
      block two
      
+    ... ~~~~{.html}'''
+    >>> markdown.markdown(text, extensions=['fenced_code'])
+    u'
      block one\\n
      \\n\\n
      <p>block two</p>\\n
      '
+
+"""
+
+import markdown, re
+
+# Global vars
+FENCED_BLOCK_RE = re.compile( \
+    r'(?P^~{3,})[ ]*\n(?P.*?)(?P=fence)[ ]*(\{\.(?P[a-zA-Z0-9_-]*)\})?[ ]*$', 
+    re.MULTILINE|re.DOTALL
+    )
+CODE_WRAP = '
      %s
      '
+LANG_TAG = ' class="%s"'
+
+
+class FencedCodeExtension(markdown.Extension):
+
+    def extendMarkdown(self, md, md_globals):
+        """ Add FencedBlockPreprocessor to the Markdown instance. """
+
+        FENCED_BLOCK_PREPROCESSOR = FencedBlockPreprocessor()
+        FENCED_BLOCK_PREPROCESSOR.md = md
+        md.textPreprocessors.insert(0, FENCED_BLOCK_PREPROCESSOR)
+
+
+class FencedBlockPreprocessor(markdown.TextPreprocessor):
+    
+    def run(self, text):
+        """ Match and store Fenced Code Blocks in the HtmlStash. """
+        while 1:
+            m = FENCED_BLOCK_RE.search(text)
+            if m:
+                lang = ''
+                if m.group('lang'):
+                    lang = LANG_TAG % m.group('lang')
+                code = CODE_WRAP % (lang, self._escape(m.group('code')))
+                placeholder = self.md.htmlStash.store(code, safe=True)
+                text = '%s\n%s\n%s'% (text[:m.start()], placeholder, text[m.end():])
+            else:
+                break
+        return text
+
+    def _escape(self, txt):
+        """ basic html escaping """
+        txt = txt.replace('&', '&')
+        txt = txt.replace('<', '<')
+        txt = txt.replace('>', '>')
+        txt = txt.replace('"', '"')
+        return txt
+
+
+def makeExtension(configs=None):
+    return FencedCodeExtension()
+
+
+if __name__ == "__main__":
+    import doctest
+    doctest.testmod()
-- 
cgit v1.2.3


From c0c88a2777a1641d4312b6aafe454f466fe104b2 Mon Sep 17 00:00:00 2001
From: Waylan Limberg 
Date: Mon, 12 May 2008 00:52:14 -0400
Subject: Added Meta-Data and HeaderId extensions. Updated setup.py to include
 all packages extensions.

---
 mdx_headerid.py | 173 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 mdx_meta.py     |  81 ++++++++++++++++++++++++++
 setup.py        |  12 +++-
 3 files changed, 265 insertions(+), 1 deletion(-)
 create mode 100644 mdx_headerid.py
 create mode 100644 mdx_meta.py

diff --git a/mdx_headerid.py b/mdx_headerid.py
new file mode 100644
index 0000000..2e658cd
--- /dev/null
+++ b/mdx_headerid.py
@@ -0,0 +1,173 @@
+#!/usr/bin/python
+
+"""
+HeaderID Extension for Python-Markdown
+======================================
+
+Adds ability to set HTML IDs for headers.
+
+Basic usage:
+
+    >>> import markdown
+    >>> text = "# Some Header # {#some_id}"
+    >>> md = markdown.markdown(text, ['headerid'])
+    >>> md
+    u'
      Some Header
      '
+
+All header IDs are unique:
+
+    >>> text = '''
+    ... #Header
+    ... #Another Header {#header}
+    ... #Third Header {#header}'''
+    >>> md = markdown.markdown(text, ['headerid'])
+    >>> md
+    u'
      Header
      \\n\\n
      Another Header
      \\n\\n
      Third Header
      '
+
+To fit within a html template's hierarchy, set the header base level:
+
+    >>> text = '''
+    ... #Some Header
+    ... ## Next Level'''
+    >>> md = markdown.markdown(text, ['headerid(level=3)'])
+    >>> md
+    u'
      Some Header
      \\n\\n
      Next Level
      '
+
+Turn off auto generated IDs:
+
+    >>> text = '''
+    ... # Some Header
+    ... # Header with ID # { #foo }'''
+    >>> md = markdown.markdown(text, ['headerid(forceid=False)'])
+    >>> md
+    u'
      Some Header
      \\n\\n
      Header with ID
      '
+
+Use with MetaData extension:
+
+    >>> text = '''header_level: 2
+    ... header_forceid: Off
+    ...
+    ... # A Header'''
+    >>> md = markdown.markdown(text, ['headerid', 'meta'])
+    >>> md
+    u'
      A Header
      '
+
+By [Waylan Limberg](http://achinghead.com/).
+
+Project website: http://achinghead.com/markdown-headerid/
+Contact: waylan [at] gmail [dot] com
+
+License: [BSD](http://www.opensource.org/licenses/bsd-license.php) 
+
+Version: 0.1 (May 2, 2008)
+
+Dependencies:
+* [Python 2.3+](http://python.org)
+* [Markdown 1.7+](http://www.freewisdom.org/projects/python-markdown/)
+
+"""
+
+import markdown
+import re
+from string import ascii_lowercase, digits, punctuation
+
+ID_CHARS = ascii_lowercase + digits + '-_'
+
+HEADER_RE = re.compile(r'''^(\#{1,6})		# group(1) = string of hashes
+                           ( [^{^#]*)		# group(2) = Header text
+                           [\#]*		    # optional closing hashes (not counted)
+                           (?:[ \t]*\{[ \t]*\#([-_:a-zA-Z0-9]+)[ \t]*\})?	# group(3) = id attr''',
+                           re.VERBOSE)
+
+IDCOUNT_RE = re.compile(r'^(.*)_([0-9]+)$')
+
+class HeaderIdExtension (markdown.Extension) :
+    def __init__(self, configs):
+        # set defaults
+        self.config = {
+                'level' : ['1', 'Base level for headers.'],
+                'forceid' : ['True', 'Force all headers to have an id.']
+            }
+
+        for key, value in configs:
+            self.setConfig(key, value)
+
+
+    def extendMarkdown(self, md, md_globals) :
+
+        md.IDs = []
+
+        def _processHeaderId(parent_elem, paragraph) :
+            ''' 
+            Overrides _processHeader of Markdown() and 
+            adds an 'id' to the header. 
+            '''
+            m = HEADER_RE.match(paragraph[0])
+            if m :
+                start_level, force_id = _get_meta()
+                level = len(m.group(1)) + start_level
+                if level > 6: level = 6
+                h = md.doc.createElement("h%d" % level)
+                parent_elem.appendChild(h)
+                for item in md._handleInline(m.group(2).strip()) :
+                    h.appendChild(item)
+                if m.group(3) :
+                    h.setAttribute('id', _unique_id(m.group(3)))
+                elif force_id:
+                    h.setAttribute('id', _create_id(m.group(2).strip()))
+            else :
+                message(CRITICAL, "We've got a problem header!")
+        
+        md._processHeader = _processHeaderId
+
+        def _get_meta():
+            ''' Return meta data suported by this ext as a tuple '''
+            level = int(self.config['level'][0]) - 1
+            force = _str2bool(self.config['forceid'][0])
+            if hasattr(md, 'Meta'):
+                if md.Meta.has_key('header_level'):
+                    level = int(md.Meta['header_level'][0]) - 1
+                if md.Meta.has_key('header_forceid'): 
+                    force = _str2bool(md.Meta['header_forceid'][0])
+            return level, force
+
+        def _str2bool(s, default=False):
+            ''' Convert a string to a booleen value. '''
+            s = str(s)
+            if s.lower() in ['0', 'f', 'false', 'off', 'no', 'n']:
+                return False
+            elif s.lower() in ['1', 't', 'true', 'on', 'yes', 'y']:
+                    return True
+            return default
+
+        def _unique_id(id):
+            ''' Ensure ID is unique. Append '_1', '_2'... if not '''
+            while id in md.IDs:
+                m = IDCOUNT_RE.match(id)
+                if m:
+                    id = '%s_%d'% (m.group(1), int(m.group(2))+1)
+                else:
+                    id = '%s_%d'% (id, 1)
+            md.IDs.append(id)
+            return id
+
+
+        def _create_id(header):
+            ''' Return ID from Header text. '''
+            h = ''
+            for c in header.lower().replace(' ', '_'):
+                if c in ID_CHARS:
+                    h += c
+                elif c not in punctuation:
+                    h += '+'
+            return _unique_id(h)
+
+            
+
+def makeExtension(configs=None) :
+    return HeaderIdExtension(configs=configs)
+
+if __name__ == "__main__":
+    import doctest
+    doctest.testmod()
+
diff --git a/mdx_meta.py b/mdx_meta.py
new file mode 100644
index 0000000..30dea8a
--- /dev/null
+++ b/mdx_meta.py
@@ -0,0 +1,81 @@
+#!usr/bin/python
+
+'''
+Meta Data Extension for Python-Markdown
+==========================================
+
+This extension adds Meta Data handling to markdown.
+
+    >>> import markdown
+    >>> text = """Title: A Test Doc.
+    ... Author: Waylan Limberg
+    ...         John Doe
+    ... Blank_Data:
+    ...
+    ... The body. This is paragraph one.
+    ... """
+    >>> md = markdown.Markdown(text, ['meta'])
+    >>> md.convert()
+    u'
      The body. This is paragraph one.\\n
      '
+    >>> md.Meta
+    {u'blank_data': [u''], u'author': [u'Waylan Limberg', u'John Doe'], u'title': [u'A Test Doc.']}
+
+Make sure text without Meta Data still works (markdown < 1.6b returns a 
      ).
+
+    >>> text = '    Some Code - not extra lines of meta data.'
+    >>> md = markdown.Markdown(text, ['meta'])
+    >>> md.convert()
+    u'
      Some Code - not extra lines of meta data.\\n
      '
+    >>> md.Meta
+    {}
+
+'''
+
+import markdown, re
+
+# Global Vars
+META_RE = re.compile(r'^[ ]{0,3}(?P[A-Za-z0-9_-]+):\s*(?P.*)')
+META_MORE_RE = re.compile(r'^[ ]{4,}(?P.*)')
+
+class MetaExtension (markdown.Extension) :
+    def __init__(self, configs):
+        pass
+
+    def extendMarkdown(self, md, md_globals) :
+        self.md = md
+
+        # Insert meta preprocessor first
+        META_PREPROCESSOR = MetaPreprocessor()
+        META_PREPROCESSOR.md = md
+        md.preprocessors.insert(0, META_PREPROCESSOR)
+        
+class MetaPreprocessor(markdown.Preprocessor) :
+    def run(self, lines) :
+        meta = {}
+        key = None
+        while 1:
+            line = lines.pop(0)
+            if line.strip() == '':
+                break # blank line - done
+            m1 = META_RE.match(line)
+            if m1:
+                key = m1.group('key').lower().strip()
+                meta[key] = [m1.group('value').strip()]
+            else:
+                m2 = META_MORE_RE.match(line)
+                if m2 and key:
+                    # Add another line to existing key
+                    meta[key].append(m2.group('value').strip())
+                else:
+                    lines.insert(0, line)
+                    break # no meta data - done
+        self.md.Meta = meta
+        return lines
+        
+
+def makeExtension(configs=None) :
+    return MetaExtension(configs=configs)
+
+if __name__ == "__main__":
+    import doctest
+    doctest.testmod()
diff --git a/setup.py b/setup.py
index de36dfc..5596eca 100644
--- a/setup.py
+++ b/setup.py
@@ -10,5 +10,15 @@ setup(
     maintainer_email = "waylan [at] gmail.com",
     url = "http://www.freewisdom.org/projects/python-markdown",
     license = "BSD License, GNU Public License (GPL)",
-    py_modules = ["markdown","mdx_footnotes", "mdx_rss"],
+    py_modules = ["markdown",
+                "mdx_codehilite",
+                "mdx_fenced_code",
+                "mdx_footnotes",
+                "mdx_headerid",
+                "mdx_imagelinks",
+                "mdx_meta",
+                "mdx_rss",
+                "mdx_tables",
+                "mdx_wikilink",
+                ],
     ) 
-- 
cgit v1.2.3


From a91e308363eeacd77270cdcb3df2109e0cfe2d18 Mon Sep 17 00:00:00 2001
From: Waylan Limberg 
Date: Wed, 14 May 2008 23:13:18 -0400
Subject: Allow hashes (#) in body of headers -- with tests. Thank you John
 Szakmeister for the bug report and patch.

---
 .gitignore              | 3 +++
 markdown.py             | 2 +-
 tests/misc/headers.html | 4 +++-
 tests/misc/headers.txt  | 2 ++
 4 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/.gitignore b/.gitignore
index 0d20b64..2ac1b58 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1 +1,4 @@
 *.pyc
+*.bak
+*.tmp
+tmp/*
diff --git a/markdown.py b/markdown.py
index acab09a..3348acc 100644
--- a/markdown.py
+++ b/markdown.py
@@ -1295,7 +1295,7 @@ class CorePatterns:
     """
 
     patterns = {
-        'header':          r'(#*)([^#]*)(#*)', # # A title
+        'header':          r'(#{1,6})[ \t]*(.*?)[ \t]*(#*)', # # A title
         'reference-def':   r'(\ ?\ ?\ ?)\[([^\]]*)\]:\s*([^ ]*)(.*)',
                            # [Google]: http://www.google.com/
         'containsline':    r'([-]*)$|^([=]*)', # -----, =====, etc.
diff --git a/tests/misc/headers.html b/tests/misc/headers.html
index 61bc266..7041eda 100644
--- a/tests/misc/headers.html
+++ b/tests/misc/headers.html
@@ -7,4 +7,6 @@
 
 
      Markdown
      
 
-
      [5]: http://foo.com/
      
\ No newline at end of file
+
      [5]: http://foo.com/
      
+
+
      Issue #1: Markdown
      
diff --git a/tests/misc/headers.txt b/tests/misc/headers.txt
index 1cb9bc0..2ddb391 100644
--- a/tests/misc/headers.txt
+++ b/tests/misc/headers.txt
@@ -7,3 +7,5 @@ Line 3
 # [Markdown](http://some.link.com/)
 
 # [5]: http://foo.com/
+
+# Issue #1: Markdown
-- 
cgit v1.2.3


From 80172e52c04683af35ab4dce1f5c285d73754824 Mon Sep 17 00:00:00 2001
From: Waylan Limberg 
Date: Sun, 18 May 2008 16:05:34 -0400
Subject: The code & pre tags in Fenced Code Blocks are now closed properly.

---
 mdx_fenced_code.py | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/mdx_fenced_code.py b/mdx_fenced_code.py
index 4b0e406..c3d9f7f 100644
--- a/mdx_fenced_code.py
+++ b/mdx_fenced_code.py
@@ -16,12 +16,12 @@ This extension adds Fenced Code Blocks to Python-Markdown.
     ... '''
     >>> html = markdown.markdown(text, extensions=['fenced_code'])
     >>> html
-    u'
      A paragraph before a fenced code block:\\n
      \\n
      Fenced code block\\n
      '
+    u'
      A paragraph before a fenced code block:\\n
      \\n
      Fenced code block\\n
      '
 
 Works with safe_mode also (we check this because we are using the HtmlStash):
 
     >>> markdown.markdown(text, extensions=['fenced_code'], safe_mode='replace')
-    u'
      A paragraph before a fenced code block:\\n
      \\n
      Fenced code block\\n
      '
+    u'
      A paragraph before a fenced code block:\\n
      \\n
      Fenced code block\\n
      '
     
 Include tilde's in a code block and wrap with blank lines:
 
@@ -32,7 +32,7 @@ Include tilde's in a code block and wrap with blank lines:
     ... 
     ... ~~~~~~~~'''
     >>> markdown.markdown(text, extensions=['fenced_code'])
-    u'
      \\n~~~~\\n\\n
      '
+    u'
      \\n~~~~\\n\\n
      '
 
 Multiple blocks and language tags:
 
@@ -45,7 +45,7 @@ Multiple blocks and language tags:
     ... 
      block two
      
     ... ~~~~{.html}'''
     >>> markdown.markdown(text, extensions=['fenced_code'])
-    u'
      block one\\n
      \\n\\n
      <p>block two</p>\\n
      '
+    u'
      block one\\n
      \\n\\n
      <p>block two</p>\\n
      '
 
 """
 
@@ -56,7 +56,7 @@ FENCED_BLOCK_RE = re.compile( \
     r'(?P^~{3,})[ ]*\n(?P.*?)(?P=fence)[ ]*(\{\.(?P[a-zA-Z0-9_-]*)\})?[ ]*$', 
     re.MULTILINE|re.DOTALL
     )
-CODE_WRAP = '
      %s
      '
+CODE_WRAP = '
      %s
      '
 LANG_TAG = ' class="%s"'
 
 
-- 
cgit v1.2.3