Das deutsche Python-Forum

Code: Alles auswählen

# -*- coding: utf-8 -*-
# ---------------------------------------------------------------------------- #
#   BBCode Parser                                                              #
#   Author: Armin Ronacher <armin.ronacher@active-4.com>                       #
#       licensed under the GPL                                                 #
# ---------------------------------------------------------------------------- #
#      This program is free software; you can redistribute it and/or           #
#       modify it under the terms of the GNU General Public License            #
#     as published by the Free Software Foundation;  either version 2          #
#         of the License, or (at your option) any later version.               #
# ---------------------------------------------------------------------------- #

import re
from random import random
from md5 import md5
from time import time

LANGUAGE = {
    "SIMPLE_QUOTE":         "Zitat:",
    "AUTHOR_QUOTE":         "%s schrieb:"
}

"""
    this module implements a simple BBCode parser for pythonBB
    Here a small example for using it:     
        text = file("input.txt", "r").read()
        p = parser(text)
        text = p.parse()
        file("output.html", "w").write(text)
"""

class parser:
    def __init__(self, text):
        self.text = text
        self.table = {}
        self.parse()
       
    def parse(self):
        text = self.text
        text = self.htmlspecialchars(text)
        text = self.extract(text)
        text = self.inlineformat(text)
        text = self.paragraphs(text)
        text = self.assamble(text)
        self.table = {}
        return text
       
    def htmlspecialchars(self, text):
        rules = [
            ("&",   "&"),
            ("< ",  "<"),
            (">",   ">")
        ]
        for rule in rules:
            text = text.replace(rule[0], rule[1])
        return text
       
    def extract(self, text):
        areas = [
            ("\[code\](.*?)\[/code\](?uism)", self.do_code),
            ("\[list(=(.*?))?\](.*?)\[/list](?uism)", self.do_list)
        ]
        for area in areas:
            try:
                while re.search(area[0], text):
                    token_id = self.__create_token_id()
                    result = re.search(area[0], text)
                    self.table[token_id] = (area[1], result)
                    text = text.replace(result.group(0), token_id, 1)
            except:
                pass
        return text
           
    def inlineformat(self, text):
        rules = [
            ("\[b\](.*?)\[/b\](?uism)", r"<strong>\1</strong>"),
            ("\[i\](.*?)\[/i\](?uism)", r"<em>\1</em>"),
            ("\[u\](.*?)\[/u\](?uism)", r"<u>\1</u>"),
            ("\[url(=(.*?))?\](.*?)\[/url\](?uism)", self.do_link),
            ("\[img(=\"(.*?)\")?\](.*?)\[/img\](?uism)", self.do_image),
            ("\[quote(=\"(.*?)\")?\](.*?)\[/quote\](?uism)", self.do_quote),
        ]
        for rule in rules:
            text = re.sub(rule[0], rule[1], text)
        return text
         
    def paragraphs(self, text):
        lines = re.split("\n{2,}", text)
        text = ""
        for line in lines:
            line = line.strip().replace("\n", "<br />\n")
            text += "<p>"+line+"</p>\n\n"
        return text
       
    def assamble(self, text):
        for token in self.table:
            handler = self.table[token][0]
            content = self.table[token][1]
            block = handler(content)
            text = text.replace(token, block)
        text = re.sub("<p>([\s]*?)</p>", "", text) #removes empty paragraphs
        return text
           
    def __create_token_id(self):
        while True:
            token = "~~"+md5(str(time() + random())).hexdigest()+"~~"
            if not self.table.has_key(token):
                return token
       
    def do_link(self, matchobj):
        caption = matchobj.group(3)       
        if matchobj.group(1) == None:
            link = matchobj.group(3)
        else:
            link = matchobj.group(2)
        return "<a href=\""+link+"\">"+caption+"</a>"
   
    def do_image(self, matchobj):
        href = matchobj.group(3)
        if matchobj.group(1) == None:
            title = matchobj.group(3)
        else:
            title = matchobj.group(2)
        return "<img src=\""+href+"\" title=\""+title+"\" alt=\""+title+"\" />"
   
    def do_quote(self, matchobj):
        text = matchobj.group(3)       
        if matchobj.group(1) == None:
            title = LANGUAGE["SIMPLE_QUOTE"]
        else:
            title = matchobj.group(2)
            if not title.endswith(":"):
                title = LANGUAGE["AUTHOR_QUOTE"].replace("%s", title)
        return "</p><div class=\"quote\"><h2>"+title+"</h2><p>"+text+"</p></div><p>"
   
    def do_code(self, matchobj):
        content = matchobj.group(1)
        if content[0] == "\n": content = content[1:]
        result = "</p><code>"+content+"</code><p>"
        return result
       
    def do_list(self, matchobj):
        content = self.inlineformat(matchobj.group(3))
        content = re.sub("\[\*\](.*?)\n", r"<li>\1</li>\n", content)   

        if matchobj.group(1) == None:
            list_type = "*"
        else:
            list_type = matchobj.group(2)
        if list_type == "1":
            result == "<ol class=\"numeric\">"+content+"</ol>"
        elif list_type == "a":
            result = "<ol class=\"alpha\">"+content+"</ol>"
        elif list_type == "i":
            result = "<ol class=\"roman\">"+content+"</ol>"
        else:
            result = "<ul>"+content+"</ul>"
        result = "</p>"+result+"<p>"
        return result

Code: Alles auswählen

LIST_STYLES = {
    '1': '<ol style="list-style-type:decimal">%s</ol>',
    'a': '<ol style="list-style-type:lower-alpha">%s</ol>',
    'i': '<ol style="list-style-type:upper-roman">%s</ol>',
}
DEFAULT_LIST_STYLE = "<ul>%s</ul>" 

def do_list(self, matchobj): 
    content = self.inlineformat(matchobj.group(3)) 
    content = re.sub("\[\*\](.*?)\n", r"<li>\1</li>\n", content)    
    list_type = (matchobj.group(2) if matchobj.group(0) else None)
    result = LIST_STYLES.get(list_type, DEFAULT_LIST_STYLE)
    return result % content

Code: Alles auswählen

def parse():
    if current_token.name == "quote":
        parse_quote()
    ...

def parse_quote():
    emit("<h4>% schrieb:</h4><blockquote>" % current_token.match[1])
    next_token()
    parse()
    if current_token.name == "/quote": next_token()
    emit("</blockquote>")