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(-) (limited to 'mdx_codehilite.py') 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