markdown.py

SQLAlchemy. 经典的Python ORM框架。学习必看。

                i += 1 # added on 2006.02.25
        else :
            i += 1

        # Add the dom elements
        for item in items :
            li = self.doc.createElement("li")
            ul.appendChild(li)

            self._processSection(li, item, inList + 1, looseList = looseList)

        # Process the remaining part of the section

        self._processSection(parent_elem, lines[i:], inList)


    def _linesUntil(self, lines, condition) :
        """ A utility function to break a list of lines upon the
            first line that satisfied a condition.  The condition
            argument should be a predicate function.
            """

        i = -1
        for line in lines :
            i += 1
            if condition(line) : break
        else :
            i += 1
        return lines[:i], lines[i:]

    def _processQuote(self, parent_elem, lines, inList) :
        """Given a list of document lines starting with a quote finds
           the end of the quote, unindents it and recursively
           processes the body of the quote and the remainder of the
           text file.

           @param parent_elem: DOM element to which the content will be added
           @param lines: a list of lines
           @param inList: a level
           @returns: None """

        dequoted = []
        i = 0
        for line in lines :
            m = RE.regExp['quoted'].match(line)
            if m :
                dequoted.append(m.group(1))
                i += 1
            else :
                break
        else :
            i += 1

        blockquote = self.doc.createElement('blockquote')
        parent_elem.appendChild(blockquote)

        self._processSection(blockquote, dequoted, inList)
        self._processSection(parent_elem, lines[i:], inList)




    def _processCodeBlock(self, parent_elem, lines, inList) :
        """Given a list of document lines starting with a code block
           finds the end of the block, puts it into the dom verbatim
           wrapped in ("<pre><code>") and recursively processes the
           the remainder of the text file.

           @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 = self.blockGuru.detectTabbed(lines)

        pre = self.doc.createElement('pre')
        code = self.doc.createElement('code')
        parent_elem.appendChild(pre)
        pre.appendChild(code)
        text = "\n".join(detabbed).rstrip()+"\n"
        #text = text.replace("&", "&amp;")
        code.appendChild(self.doc.createTextNode(text))
        self._processSection(parent_elem, theRest, inList)


    def _handleInlineWrapper2 (self, line) :


        parts = [line]

        #if not(line):
        #    return [self.doc.createTextNode(' ')]

        for pattern in self.inlinePatterns :

            #print
            #print self.inlinePatterns.index(pattern)

            i = 0

            #print parts
            while i < len(parts) :
                
                x = parts[i]
                #print i
                if isinstance(x, (str, unicode)) :
                    result = self._applyPattern(x, pattern)
                    #print result
                    #print result
                    #print parts, i
                    if result :
                        i -= 1
                        parts.remove(x)
                        for y in result :
                            parts.insert(i+1,y)
                
                i += 1

        for i in range(len(parts)) :
            x = parts[i]
            if isinstance(x, (str, unicode)) :
                parts[i] = self.doc.createTextNode(x)

        return parts
        


    def _handleInlineWrapper (self, line) :

        # A wrapper around _handleInline to avoid recursion

        parts = [line]

        i = 0
        
        while i < len(parts) :
            x = parts[i]
            if isinstance(x, (str, unicode)) :
                parts.remove(x)
                result = self._handleInline(x)
                for y in result :
                    parts.insert(i,y)
            else :
                i += 1

        return parts

    def _handleInline(self,  line):
        """Transform a Markdown line with inline elements to an XHTML
        fragment.

        This function uses auxiliary objects called inline patterns.
        See notes on inline patterns above.

        @param item: A block of Markdown text
        @return: A list of NanoDom nodes """

        if not(line):
            return [self.doc.createTextNode(' ')]

        for pattern in self.inlinePatterns :
            list = self._applyPattern( line, pattern)
            if list: return list

        return [self.doc.createTextNode(line)]

    def _applyPattern(self, line, pattern) :
        """ Given a pattern name, this function checks if the line
        fits the pattern, creates the necessary elements, and returns
        back a list consisting of NanoDom elements and/or strings.
        
        @param line: the text to be processed
        @param pattern: the pattern to be checked

        @returns: the appropriate newly created NanoDom element if the
                  pattern matches, None otherwise.
        """

        # match the line to pattern's pre-compiled reg exp.
        # if no match, move on.

        m = pattern.getCompiledRegExp().match(line)
        if not m :
            return None

        # if we got a match let the pattern make us a NanoDom node
        # if it doesn't, move on
        node = pattern.handleMatch(m, self.doc)

        if node :
            # Those are in the reverse order!
            return ( m.groups()[-1], # the string to the left
                     node,           # the new node
                     m.group(1))     # the string to the right of the match

        else :
            return None

    def __str__(self, source = None):
        """Return the document in XHTML format.

        @returns: A serialized XHTML body."""
        #try :

        if source :
            self.source = source
        
        doc = self._transform()
        xml = doc.toxml()

        #finally:
        #    doc.unlink()

        # Let's stick in all the raw html pieces

        for i in range(self.htmlStash.html_counter) :
            html = self.htmlStash.rawHtmlBlocks[i]
            if self.safeMode :
                html = "[HTML_REMOVED]"
                
            xml = xml.replace("<p>%s\n</p>" % (HTML_PLACEHOLDER % i),
                              html + "\n")
            xml = xml.replace(HTML_PLACEHOLDER % i,
                              html)

        # And return everything but the top level tag

        if self.stripTopLevelTags :
            xml = xml.strip()[23:-7] + "\n"

        for pp in self.textPostprocessors :
            xml = pp.run(xml)

        return self.docType + xml


    toString = __str__


    def __unicode__(self):
        """Return the document in XHTML format as a Unicode object.
        """
        return str(self)#.decode(self.encoding)


    toUnicode = __unicode__




# ====================================================================

def markdownFromFile(input = None,
                     output = None,
                     extensions = [],
                     encoding = None,
                     message_threshold = CRITICAL,
                     safe = False) :

    global MESSAGE_THRESHOLD
    MESSAGE_THRESHOLD = message_threshold

    message(VERBOSE, "input file: %s" % input)


    if not encoding :
        encoding = "utf-8"

    input_file = codecs.open(input, mode="r", encoding="utf-8")
    text = input_file.read()
    input_file.close()

    new_text = markdown(text, extensions, encoding, safe_mode = safe)

    if output :
        output_file = codecs.open(output, "w", encoding=encoding)
        output_file.write(new_text)
        output_file.close()

    else :
        sys.stdout.write(new_text.encode(encoding))

def markdown(text,
             extensions = [],
             encoding = None,
             safe_mode = False) :
    
    message(VERBOSE, "in markdown.markdown(), received text:\n%s" % text)

    extension_names = []
    extension_configs = {}
    
    for ext in extensions :
        pos = ext.find("(") 
        if pos == -1 :
            extension_names.append(ext)
        else :
            name = ext[:pos]
            extension_names.append(name)
            pairs = [x.split("=") for x in ext[pos+1:-1].split(",")]
            configs = [(x.strip(), y.strip()) for (x, y) in pairs]
            extension_configs[name] = configs
            #print configs

    md = Markdown(text, extensions=extension_names,
                  extension_configs=extension_configs,
                  safe_mode = safe_mode)

    return md.toString()
        

class Extension :

    def __init__(self, configs = {}) :
        self.config = configs

    def getConfig(self, key) :
        if self.config.has_key(key) :
            #print self.config[key][0]
            return self.config[key][0]
        else :
            return ""

    def getConfigInfo(self) :
        return [(key, self.config[key][1]) for key in self.config.keys()]

    def setConfig(self, key, value) :
        self.config[key][0] = value


OPTPARSE_WARNING = """
Python 2.3 or higher required for advanced command line options.
For lower versions of Python use:

      %s INPUT_FILE > OUTPUT_FILE
    
""" % EXECUTABLE_NAME_FOR_USAGE

def parse_options() :

    try :
        optparse = __import__("optparse")
    except :
        if len(sys.argv) == 2 :
            return {'input' : sys.argv[1],
                    'output' : None,
                    'message_threshold' : CRITICAL,
                    'safe' : False,
                    'extensions' : [],
                    'encoding' : None }

        else :
            print OPTPARSE_WARNING
            return None

    parser = optparse.OptionParser(usage="%prog INPUTFILE [options]")

    parser.add_option("-f", "--file", dest="filename",
                      help="write output to OUTPUT_FILE",
                      metavar="OUTPUT_FILE")
    parser.add_option("-e", "--encoding", dest="encoding",
                      help="encoding for input and output files",)
    parser.add_option("-q", "--quiet", default = CRITICAL,
                      action="store_const", const=NONE, dest="verbose",
                      help="suppress all messages")
    parser.add_option("-v", "--verbose",
                      action="store_const", const=INFO, dest="verbose",
                      help="print info messages")
    parser.add_option("-s", "--safe",
                      action="store_const", const=True, dest="safe",
                      help="same mode (strip user's HTML tag)")
    
    parser.add_option("--noisy",
                      action="store_const", const=VERBOSE, dest="verbose",
                      help="print debug messages")
    parser.add_option("-x", "--extension", action="append", dest="extensions",
                      help = "load extension EXTENSION", metavar="EXTENSION")

    (options, args) = parser.parse_args()

    if not len(args) == 1 :
        parser.print_help()
        return None
    else :
        input_file = args[0]

    if not options.extensions :
        options.extensions = []

    return {'input' : input_file,
            'output' : options.filename,
            'message_threshold' : options.verbose,
            'safe' : options.safe,
            'extensions' : options.extensions,
            'encoding' : options.encoding }

if __name__ == '__main__':
    """ Run Markdown from the command line. """

    options = parse_options()

    #if os.access(inFile, os.R_OK):

    if not options :
        sys.exit(0)
    
    markdownFromFile(**options)