From 3e7b8b209060988ed020f0eda33b1a2f7c292be7 Mon Sep 17 00:00:00 2001 From: Rob Austein Date: Mon, 18 Mar 2019 20:44:57 +0000 Subject: Initial wiki dump and initial tools --- tools/trac2md.py | 192 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 192 insertions(+) create mode 100644 tools/trac2md.py (limited to 'tools/trac2md.py') diff --git a/tools/trac2md.py b/tools/trac2md.py new file mode 100644 index 0000000..40c09d4 --- /dev/null +++ b/tools/trac2md.py @@ -0,0 +1,192 @@ +#!/usr/bin/python + +# Untested code from https://www.snip2code.com/Snippet/1704331/Convert-trac-markup-to-Markdown/ + +# This code mostly taken from patches to pagure_importer by mreynolds + +import sys +import re +import time +import requests +import shutil +import os +from base64 import b64decode +from datetime import datetime + +wikilink_pattern = re.compile('\[http(.*)\]') +wikilink_extract = re.compile('\[(.*)\]') +wikiheading1_pattern = re.compile('^= (.*) =$') +wikiheading2_pattern = re.compile('^== (.*) ==$') +wikiheading3_pattern = re.compile('^=== (.*) ===$') +strikethrough_pattern = re.compile('~~(.*)~~') + +def to_timestamp(tm): + ''' Convert to timestamp which can be jsonified ''' + + tm = tm.replace('+00:00', '') + date = datetime.strptime(tm, '%Y-%m-%dT%H:%M:%S') + ts = str(time.mktime(date.timetuple()))[:-2] # Strip the .0 + return ts + + +def strip_wikilink(content): + ''' Need to remove wiki link format from custom fields. They come in a + variety of forms that can be comma or whitespace separated. They can also + include link names which must also be removed. + + [https://bugzilla.redhat.com/show_bug.cgi?id=772777] + [https://bugzilla.com/123456789], [http://bugzilla.com/7777777 7777777] + [https://bugzilla.com/6666666 6666666] + ''' + + links = [] + if wikilink_pattern.search(content): + # Looks like we have a link in here + links = [] + mylist = re.findall(r'\[([^]]*)\]', content) + for i in mylist: + links.append(i.split(' ', 1)[0]) + return ', '.join(links) + else: + return content + + +def convert_headers(line): + ''' Convert wikiformat headers + ''' + level_count = 1 + for header in [wikiheading1_pattern, + wikiheading2_pattern, + wikiheading3_pattern]: + try: + level = header.search(line).group(1) + if level: + line = "%s %s" % ('#' * level_count, level) + break # No need to check other heading levels + except: + # Try the next heading level + pass + level_count += 1 + + return line + + +def convert_wikilinks(line): + ''' Convert wikiformat links + ''' + if wikilink_pattern.search(line): + try: + result = wikilink_extract.search(line).group(1) + if result: + parts = result.split(' ', 1) + if len(parts) == 1: + mdlink = '[%s](%s)' % (parts[0], parts[0]) + elif len(parts) == 2: + mdlink = '[%s](%s)' % (parts[1], parts[0]) + line = line.replace('[' + result + ']', mdlink) + except: + # Not a link, not a problem + pass + + return line + + +def convert_strike(line): + ''' Convert wikiformat striked text + ''' + striked_result = strikethrough_pattern.search(line) + if striked_result: + try: + striked_text = striked_result.group(1) + if striked_text: + orig_text = '~~%s~~' % striked_text + new_text = '%s' % striked_text + line = line.replace(orig_text, new_text) + except: + # Not striked + pass + return line + +def WikiToMD(content): + ''' Convert wiki/RST format to Markdown. Code blocks, bold/italics, + wiki links, lists, striked text, and headers. ''' + + code_block = False + in_list = False + nested_level = 0 + prev_indent = 0 + new_content = "" + + for line in content.split('\n'): + line = line.replace("\r", "") + if "{{{" in line: + code_block = True + line = line.replace("{{{", "```") + if "}}}" in line: + code_block = False + line = line.replace("}}}", "```") + if not code_block: + # + # Convert bullet lists. The start and end of a list needs + # an empty line. wikiformat uses both '*' and '-' for its + # lists. However, markdown only supports '-'. + # + if line.startswith('* '): + if not in_list: + new_content = "%s\n" % (new_content) + in_list = True + line = line[1:] + line = '-%s' % (line) + elif line.startswith('- '): + # No need to modify the line, just add the new line + if not in_list: + new_content = "%s\n" % (new_content) + in_list = True + elif line.startswith(' '): + # Check for nested lists + nested_line = line.lstrip(' ') + if nested_line.startswith('* ') or \ + nested_line.startswith('- '): + # Adjust the nested list level as needed + indent = len(line) - len(nested_line) + if indent > prev_indent: + nested_level += 1 + elif indent < prev_indent: + nested_level -= 1 + prev_indent = indent + + # Set the proper indentation for markdown + line = ('%s-%s' % (' ' * nested_level, + nested_line[1:])) + else: + if in_list: + # Add the closing empty line + new_content = "%s\n" % (new_content) + in_list = False + nested_level = 0 + prev_indent = 0 + + # Convert headers + line = convert_headers(line) + + # Convert wiki links + line = convert_wikilinks(line) + + # Convert striked through text + line = convert_strike(line) + + # Convert bold and italic text (do this last) + line = line.replace("'''", "**") # Convert bold text + line = line.replace("''", "*") # Convert italic text + + new_content = "%s%s\n" % (new_content, line) + + return new_content + +for f in sys.argv[1:]: + d = WikiToMD(open(f, "r").read()) + newf = f.replace(".trac", ".md") + with open(newf, "w") as fp: + fp.write(d) + pass + pass -- cgit v1.2.3