%s\n
" % (HTML_PLACEHOLDER % i), html + "\n") text = text.replace(HTML_PLACEHOLDER % i, html) return text def escape(self, html): ''' Basic html escaping ''' html = html.replace('&', '&') html = html.replace('<', '<') html = html.replace('>', '>') return html.replace('"', '"') RAWHTMLTEXTPOSTPROCESSOR = RawHtmlTextPostprocessor() """ ====================================================================== ========================== MISC AUXILIARY CLASSES ==================== ====================================================================== """ class HtmlStash : """This class is used for stashing HTML objects that we extract in the beginning and replace with place-holders.""" def __init__ (self) : self.html_counter = 0 # for counting inline html segments self.rawHtmlBlocks=[] def store(self, html, safe=False) : """Saves an HTML segment for later reinsertion. Returns a placeholder string that needs to be inserted into the document. @param html: an html segment @param safe: label an html segment as safe for safemode @param inline: label a segmant as inline html @returns : a placeholder string """ self.rawHtmlBlocks.append((html, safe)) placeholder = HTML_PLACEHOLDER % self.html_counter self.html_counter += 1 return placeholder class BlockGuru : def _findHead(self, lines, fn, allowBlank=0) : """Functional magic to help determine boundaries of indented blocks. @param lines: an array of strings @param fn: a function that returns a substring of a string if the string matches the necessary criteria @param allowBlank: specifies whether it's ok to have blank lines between matching functions @returns: a list of post processes items and the unused remainder of the original list""" items = [] item = -1 i = 0 # to keep track of where we are for line in lines : if not line.strip() and not allowBlank: return items, lines[i:] if not line.strip() and allowBlank: # If we see a blank line, this _might_ be the end i += 1 # Find the next non-blank line for j in range(i, len(lines)) : if lines[j].strip() : next = lines[j] break else : # There is no more text => this is the end break # Check if the next non-blank line is still a part of the list part = fn(next) if part : items.append("") continue else : break # found end of the list part = fn(line) if part : items.append(part) i += 1 continue else : return items, lines[i:] else : i += 1 return items, lines[i:] def detabbed_fn(self, line) : """ An auxiliary method to be passed to _findHead """ m = RE.regExp['tabbed'].match(line) if m: return m.group(4) else : return None def detectTabbed(self, lines) : return self._findHead(lines, self.detabbed_fn, allowBlank = 1) def print_error(string): """Print an error string to stderr""" sys.stderr.write(string +'\n') def dequote(string) : """ Removes quotes from around a string """ if ( ( string.startswith('"') and string.endswith('"')) or (string.startswith("'") and string.endswith("'")) ) : return string[1:-1] else : return string """ ====================================================================== ========================== CORE MARKDOWN ============================= ====================================================================== This stuff is ugly, so if you are thinking of extending the syntax, see first if you can do it via pre-processors, post-processors, inline patterns or a combination of the three. """ class CorePatterns : """This class is scheduled for removal as part of a refactoring effort.""" patterns = { 'header': r'(#*)([^#]*)(#*)', # # A title 'reference-def' : r'(\ ?\ ?\ ?)\[([^\]]*)\]:\s*([^ ]*)(.*)', # [Google]: http://www.google.com/ 'containsline': r'([-]*)$|^([=]*)', # -----, =====, etc. 'ol': r'[ ]{0,3}[\d]*\.\s+(.*)', # 1. text 'ul': r'[ ]{0,3}[*+-]\s+(.*)', # "* text" 'isline1': r'(\**)', # *** 'isline2': r'(\-*)', # --- 'isline3': r'(\_*)', # ___ 'tabbed': r'((\t)|( ))(.*)', # an indented line 'quoted' : r'> ?(.*)', # a quoted block ("> ...") } def __init__ (self) : self.regExp = {} for key in self.patterns.keys() : self.regExp[key] = re.compile("^%s$" % self.patterns[key], re.DOTALL) self.regExp['containsline'] = re.compile(r'^([-]*)$|^([=]*)$', re.M) RE = CorePatterns() class Markdown: """ Markdown formatter class for creating an html document from Markdown text """ def __init__(self, source=None, # deprecated extensions=[], extension_configs=None, encoding="utf-8", safe_mode = False): """Creates a new Markdown instance. @param source: The text in Markdown format. @param encoding: The character encoding ofaround it - append the paragraph bits directly # onto parent_elem el = parent_elem else : # Otherwise make a "p" element el = self.doc.createElement("p") parent_elem.appendChild(el) for item in list : el.appendChild(item) def _processUList(self, parent_elem, lines, inList) : self._processList(parent_elem, lines, inList, listexpr='ul', tag = 'ul') def _processOList(self, parent_elem, lines, inList) : self._processList(parent_elem, lines, inList, listexpr='ol', tag = 'ol') def _processList(self, parent_elem, lines, inList, listexpr, tag) : """Given a list of document lines starting with a list item, finds the end of the list, breaks it up, and recursively processes each list item and the remainder of the text file. @param parent_elem: A dom element to which the content will be added @param lines: a list of lines @param inList: a level @returns: None""" ul = self.doc.createElement(tag) # ul might actually be '
") 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("&", "&")
code.appendChild(self.doc.createTextNode(text))
self._processSection(parent_elem, theRest, inList)
def _handleInlineWrapper (self, line, patternIndex=0) :
parts = [line]
while patternIndex < len(self.inlinePatterns) :
i = 0
while i < len(parts) :
x = parts[i]
if isinstance(x, (str, unicode)) :
result = self._applyPattern(x, \
self.inlinePatterns[patternIndex], \
patternIndex )
if result :
i -= 1
parts.remove(x)
for y in result :
parts.insert(i+1,y)
i += 1
patternIndex += 1
for i in range(len(parts)) :
x = parts[i]
if isinstance(x, (str, unicode)) :
parts[i] = self.doc.createTextNode(x)
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, patternIndex) :
""" 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)
# check if any of this nodes have children that need processing
if isinstance(node, Element):
if not node.nodeName in ["code", "pre"] :
for child in node.childNodes :
if isinstance(child, TextNode):
result = self._handleInlineWrapper(child.value, patternIndex+1)
if result:
if result == [child] :
continue
result.reverse()
#to make insertion easier
position = node.childNodes.index(child)
node.removeChild(child)
for item in result:
if isinstance(item, (str, unicode)):
if len(item) > 0:
node.insertChild(position,
self.doc.createTextNode(item))
else:
node.insertChild(position, item)
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 convert (self, source = None):
"""Return the document in XHTML format.
@returns: A serialized XHTML body."""
#try :
if source is not None: #Allow blank string
self.source = source
if not self.source :
return ""
self.source = removeBOM(self.source, self.encoding)
for pp in self.textPreprocessors :
self.source = pp.run(self.source)
doc = self._transform()
xml = doc.toxml()
#finally:
# doc.unlink()
# 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).strip()
__str__ = convert # deprecated - will be changed in 1.7 to report
# information about the MD instance
toString = __str__ # toString() method is deprecated
def __unicode__(self):
"""Return the document in XHTML format as a Unicode object.
"""
return str(self)#.decode(self.encoding)
toUnicode = __unicode__ # deprecated - will be removed in 1.7
# ====================================================================
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=encoding)
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
md = Markdown(extensions=extension_names,
extension_configs=extension_configs,
safe_mode = safe_mode)
return md.convert(text)
class Extension :
def __init__(self, configs = {}) :
self.config = configs
def getConfig(self, key) :
if self.config.has_key(key) :
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", dest="safe", default=False,
metavar="SAFE_MODE",
help="same mode ('replace', 'remove' or 'escape' 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)