Mercurial > public > pelican-blog

annotate tools/import_blogophile.py @ 12:5ff71680269a

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Set DISQUS_NO_ID so we don't confuse Disqus. The Pelican Bootstrap3 theme is by default adding data-disqus-identifier to my comment count links. Since my old blog didn't have these, I think Disqus is getting confused. When I removed this, some blog entries where Disqus didn't know what the comment count was suddenly began showing the correct comment counts (mainly 0 I think).
author Brian Neal <bgneal@gmail.com>
date Tue, 04 Feb 2014 18:44:59 -0600
parents 6c03ca07a16d
children
rev   line source
bgneal@2 1 #!/usr/bin/env python
bgneal@2 2 """
bgneal@2 3 A simple script to convert my Blogofile restructured text posts into the format
bgneal@2 4 expected by Pelican.
bgneal@2 5
bgneal@2 6 """
bgneal@2 7 # Copyright (C) 2014 by Brian Neal.
bgneal@2 8 #
bgneal@2 9 # Permission is hereby granted, free of charge, to any person obtaining a copy
bgneal@2 10 # of this software and associated documentation files (the "Software"), to deal
bgneal@2 11 # in the Software without restriction, including without limitation the rights
bgneal@2 12 # to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
bgneal@2 13 # copies of the Software, and to permit persons to whom the Software is
bgneal@2 14 # furnished to do so, subject to the following conditions:
bgneal@2 15 #
bgneal@2 16 # The above copyright notice and this permission notice shall be included in
bgneal@2 17 # all copies or substantial portions of the Software.
bgneal@2 18 #
bgneal@2 19 # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
bgneal@2 20 # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
bgneal@2 21 # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
bgneal@2 22 # AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
bgneal@2 23 # LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
bgneal@2 24 # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
bgneal@2 25 # THE SOFTWARE.
bgneal@2 26
bgneal@2 27 import datetime
bgneal@2 28 import os
bgneal@2 29 import re
bgneal@2 30 import time
bgneal@2 31
bgneal@2 32
bgneal@2 33 SRC_DIR = os.path.expanduser('~/coding/python/virtualenvs/blogofile/blog/_posts')
bgneal@3 34 DST_DIR = os.path.expanduser('~/coding/python/venvs/blog/blog-pelican/content/Coding')
bgneal@2 35
bgneal@2 36 TITLE_RE = re.compile(r'^title: (?P<title>.*)$')
bgneal@2 37 DATE_RE = re.compile(r'^date: (?P<year>\d{4})/'
bgneal@2 38 r'(?P<month>\d{1,2})/'
bgneal@2 39 r'(?P<day>\d{1,2})\s*'
bgneal@2 40 r'(?P<time>\d{2}:\d{2}:\d{2})\s*$')
bgneal@2 41 CAT_RE = re.compile(r'^categories: (?P<cats>.*)$')
bgneal@2 42
bgneal@2 43 PELICAN_FMT = """\
bgneal@2 44 {title}
bgneal@2 45 {title_underline}
bgneal@2 46
bgneal@2 47 :date: {date}
bgneal@2 48 :tags: {tags}
bgneal@2 49 :slug: {slug}
bgneal@2 50 :author: Brian Neal
bgneal@2 51
bgneal@2 52 {content}
bgneal@2 53 """
bgneal@2 54
bgneal@2 55
bgneal@2 56 class ConvertError(Exception):
bgneal@2 57 """Exception class for the conversion process"""
bgneal@2 58
bgneal@2 59
bgneal@2 60 def slugify(s):
bgneal@2 61 """Return a slug from the string s.
bgneal@2 62
bgneal@2 63 This code must match what Blogofile was doing in order to keep the URLs the
bgneal@2 64 same. In this case I had customized Blogfile's functionality based on a tip
bgneal@2 65 by Mike Bayer: http://techspot.zzzeek.org/2010/12/06/my-blogofile-hacks/
bgneal@2 66
bgneal@2 67 """
bgneal@2 68 slug = s.lower()
bgneal@2 69
bgneal@2 70 # convert ellipses to spaces
bgneal@2 71 slug = re.sub(r'\.{2,}', ' ', slug)
bgneal@2 72
bgneal@2 73 # flatten everything non alpha or . into a single -
bgneal@2 74 slug = re.sub(r'[^0-9a-zA-Z\.]+', '-', slug)
bgneal@2 75
bgneal@2 76 # trim off leading/trailing -
bgneal@2 77 slug = re.sub(r'^-+|-+$', '', slug)
bgneal@2 78 return slug
bgneal@2 79
bgneal@2 80
bgneal@2 81 def convert(src, dst):
bgneal@2 82 """Convert Blogofile to Pelican."""
bgneal@2 83 print '{} -> {}'.format(src, dst)
bgneal@2 84 meta, content = parse_input(src)
bgneal@2 85 write_output(meta, content, dst)
bgneal@2 86
bgneal@2 87
bgneal@2 88 def parse_input(src):
bgneal@2 89 """Parse input Blogofile .rst input.
bgneal@2 90
bgneal@2 91 Returns a 2-tuple:
bgneal@2 92 meta - dictionary of Blogofile metadata
bgneal@2 93 content - blog post body as a string
bgneal@2 94
bgneal@2 95 """
bgneal@2 96 with open(src, 'r') as fp:
bgneal@2 97 lines = fp.readlines()
bgneal@2 98
bgneal@2 99 # Find meta block
bgneal@2 100 for i, line in enumerate(lines):
bgneal@2 101 if line == '---\n':
bgneal@2 102 meta_start = i
bgneal@2 103 break
bgneal@2 104 else:
bgneal@2 105 raise ConvertError("Can't find start of meta block")
bgneal@2 106
bgneal@2 107 for i, line in enumerate(lines[meta_start + 1 :]):
bgneal@2 108 if line == '---\n':
bgneal@2 109 meta_end = meta_start + 1 + i
bgneal@2 110 break
bgneal@2 111 else:
bgneal@2 112 raise ConvertError("Can't find end of meta block")
bgneal@2 113
bgneal@2 114 meta_lines = lines[meta_start + 1 : meta_end]
bgneal@2 115 meta = {}
bgneal@2 116 for line in meta_lines:
bgneal@2 117 m = TITLE_RE.match(line)
bgneal@2 118 if m:
bgneal@2 119 meta['title'] = m.group('title').strip()
bgneal@2 120 continue
bgneal@2 121 m = DATE_RE.match(line)
bgneal@2 122 if m:
bgneal@2 123 year = int(m.group('year'))
bgneal@2 124 month = int(m.group('month'))
bgneal@2 125 day = int(m.group('day'))
bgneal@2 126 t = time.strptime(m.group('time'), '%H:%M:%S')
bgneal@2 127 meta['date'] = datetime.datetime.combine(
bgneal@2 128 datetime.date(year, month, day),
bgneal@2 129 datetime.time(t.tm_hour, t.tm_min, t.tm_sec))
bgneal@2 130 continue
bgneal@2 131 m = CAT_RE.match(line)
bgneal@2 132 if m:
bgneal@2 133 meta['categories'] = m.group('cats').replace(' ', '').split(',')
bgneal@2 134 continue
bgneal@2 135
bgneal@2 136 for k in ['title', 'date', 'categories']:
bgneal@2 137 if k not in meta:
bgneal@2 138 raise ConvertError("Missing {} in metadata".format(k))
bgneal@2 139
bgneal@2 140 content = ''.join(lines[meta_end + 1:]).strip()
bgneal@2 141 return meta, content
bgneal@2 142
bgneal@2 143
bgneal@2 144 def write_output(meta, content, dst):
bgneal@2 145 """Create the Pelican style .rst file from the Blogofile metadata and
bgneal@2 146 content. Output is written to the file specified by dst.
bgneal@2 147
bgneal@2 148 """
bgneal@2 149 title = meta['title']
bgneal@2 150 date = meta['date'].strftime('%Y-%m-%d %H:%M')
bgneal@2 151 tags = ', '.join(meta['categories'])
bgneal@2 152 slug = slugify(title)
bgneal@2 153
bgneal@2 154 post = PELICAN_FMT.format(title=title,
bgneal@2 155 title_underline='#'*len(title),
bgneal@2 156 date=date,
bgneal@2 157 tags=tags,
bgneal@2 158 slug=slug,
bgneal@2 159 content=content)
bgneal@2 160
bgneal@2 161 with open(dst, 'w') as fp:
bgneal@2 162 fp.write(post)
bgneal@2 163
bgneal@2 164
bgneal@2 165 if __name__ == '__main__':
bgneal@2 166 for name in os.listdir(SRC_DIR):
bgneal@2 167 if name.endswith('.rst'):
bgneal@2 168 src = os.path.join(SRC_DIR, name)
bgneal@2 169 dst = os.path.join(DST_DIR, name)
bgneal@2 170
bgneal@2 171 try:
bgneal@2 172 convert(src, dst)
bgneal@2 173 except ConvertError as ex:
bgneal@2 174 print "Error converting {}: {}".format(name, ex)