bgneal@2: #!/usr/bin/env python
bgneal@2: """
bgneal@2: A simple script to convert my Blogofile restructured text posts into the format
bgneal@2: expected by Pelican.
bgneal@2: 
bgneal@2: """
bgneal@2: # Copyright (C) 2014 by Brian Neal.
bgneal@2: #
bgneal@2: # Permission is hereby granted, free of charge, to any person obtaining a copy
bgneal@2: # of this software and associated documentation files (the "Software"), to deal
bgneal@2: # in the Software without restriction, including without limitation the rights
bgneal@2: # to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
bgneal@2: # copies of the Software, and to permit persons to whom the Software is
bgneal@2: # furnished to do so, subject to the following conditions:
bgneal@2: # 
bgneal@2: # The above copyright notice and this permission notice shall be included in
bgneal@2: # all copies or substantial portions of the Software.
bgneal@2: # 
bgneal@2: # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
bgneal@2: # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
bgneal@2: # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
bgneal@2: # AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
bgneal@2: # LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
bgneal@2: # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
bgneal@2: # THE SOFTWARE.
bgneal@2: 
bgneal@2: import datetime
bgneal@2: import os
bgneal@2: import re
bgneal@2: import time
bgneal@2: 
bgneal@2: 
bgneal@2: SRC_DIR = os.path.expanduser('~/coding/python/virtualenvs/blogofile/blog/_posts')
bgneal@2: DST_DIR = os.path.expanduser('~/coding/python/venvs/blog/blog-pelican/content')
bgneal@2: 
bgneal@2: TITLE_RE = re.compile(r'^title: (?P<title>.*)$')
bgneal@2: DATE_RE = re.compile(r'^date: (?P<year>\d{4})/'
bgneal@2:                      r'(?P<month>\d{1,2})/'
bgneal@2:                      r'(?P<day>\d{1,2})\s*'
bgneal@2:                      r'(?P<time>\d{2}:\d{2}:\d{2})\s*$')
bgneal@2: CAT_RE = re.compile(r'^categories: (?P<cats>.*)$')
bgneal@2: 
bgneal@2: PELICAN_FMT = """\
bgneal@2: {title}
bgneal@2: {title_underline}
bgneal@2: 
bgneal@2: :date: {date}
bgneal@2: :category: coding
bgneal@2: :tags: {tags}
bgneal@2: :slug: {slug}
bgneal@2: :author: Brian Neal
bgneal@2: 
bgneal@2: {content}
bgneal@2: """
bgneal@2: 
bgneal@2: 
bgneal@2: class ConvertError(Exception):
bgneal@2:     """Exception class for the conversion process"""
bgneal@2: 
bgneal@2: 
bgneal@2: def slugify(s):
bgneal@2:     """Return a slug from the string s.
bgneal@2: 
bgneal@2:     This code must match what Blogofile was doing in order to keep the URLs the
bgneal@2:     same. In this case I had customized Blogfile's functionality based on a tip
bgneal@2:     by Mike Bayer: http://techspot.zzzeek.org/2010/12/06/my-blogofile-hacks/
bgneal@2: 
bgneal@2:     """
bgneal@2:     slug = s.lower()
bgneal@2: 
bgneal@2:     # convert ellipses to spaces
bgneal@2:     slug = re.sub(r'\.{2,}', ' ', slug)
bgneal@2: 
bgneal@2:     # flatten everything non alpha or . into a single -
bgneal@2:     slug = re.sub(r'[^0-9a-zA-Z\.]+', '-', slug)
bgneal@2: 
bgneal@2:     # trim off leading/trailing -
bgneal@2:     slug = re.sub(r'^-+|-+$', '', slug)
bgneal@2:     return slug
bgneal@2: 
bgneal@2: 
bgneal@2: def convert(src, dst):
bgneal@2:     """Convert Blogofile to Pelican."""
bgneal@2:     print '{} -> {}'.format(src, dst)
bgneal@2:     meta, content = parse_input(src)
bgneal@2:     write_output(meta, content, dst)
bgneal@2: 
bgneal@2: 
bgneal@2: def parse_input(src):
bgneal@2:     """Parse input Blogofile .rst input.
bgneal@2: 
bgneal@2:     Returns a 2-tuple:
bgneal@2:         meta - dictionary of Blogofile metadata
bgneal@2:         content - blog post body as a string
bgneal@2: 
bgneal@2:     """
bgneal@2:     with open(src, 'r') as fp:
bgneal@2:         lines = fp.readlines()
bgneal@2: 
bgneal@2:     # Find meta block
bgneal@2:     for i, line in enumerate(lines):
bgneal@2:         if line == '---\n':
bgneal@2:             meta_start = i
bgneal@2:             break
bgneal@2:     else:
bgneal@2:         raise ConvertError("Can't find start of meta block")
bgneal@2: 
bgneal@2:     for i, line in enumerate(lines[meta_start + 1 :]):
bgneal@2:         if line == '---\n':
bgneal@2:             meta_end = meta_start + 1 + i
bgneal@2:             break
bgneal@2:     else:
bgneal@2:         raise ConvertError("Can't find end of meta block")
bgneal@2: 
bgneal@2:     meta_lines = lines[meta_start + 1 : meta_end]
bgneal@2:     meta = {}
bgneal@2:     for line in meta_lines:
bgneal@2:         m = TITLE_RE.match(line)
bgneal@2:         if m:
bgneal@2:             meta['title'] = m.group('title').strip()
bgneal@2:             continue
bgneal@2:         m = DATE_RE.match(line)
bgneal@2:         if m:
bgneal@2:             year = int(m.group('year'))
bgneal@2:             month = int(m.group('month'))
bgneal@2:             day = int(m.group('day'))
bgneal@2:             t = time.strptime(m.group('time'), '%H:%M:%S')
bgneal@2:             meta['date'] = datetime.datetime.combine(
bgneal@2:                     datetime.date(year, month, day),
bgneal@2:                     datetime.time(t.tm_hour, t.tm_min, t.tm_sec))
bgneal@2:             continue
bgneal@2:         m = CAT_RE.match(line)
bgneal@2:         if m:
bgneal@2:             meta['categories'] = m.group('cats').replace(' ', '').split(',')
bgneal@2:             continue
bgneal@2: 
bgneal@2:     for k in ['title', 'date', 'categories']:
bgneal@2:         if k not in meta:
bgneal@2:             raise ConvertError("Missing {} in metadata".format(k))
bgneal@2: 
bgneal@2:     content = ''.join(lines[meta_end + 1:]).strip()
bgneal@2:     return meta, content
bgneal@2: 
bgneal@2: 
bgneal@2: def write_output(meta, content, dst):
bgneal@2:     """Create the Pelican style .rst file from the Blogofile metadata and
bgneal@2:     content. Output is written to the file specified by dst.
bgneal@2: 
bgneal@2:     """
bgneal@2:     title = meta['title']
bgneal@2:     date = meta['date'].strftime('%Y-%m-%d %H:%M')
bgneal@2:     tags = ', '.join(meta['categories'])
bgneal@2:     slug = slugify(title)
bgneal@2: 
bgneal@2:     post = PELICAN_FMT.format(title=title,
bgneal@2:             title_underline='#'*len(title),
bgneal@2:             date=date,
bgneal@2:             tags=tags,
bgneal@2:             slug=slug,
bgneal@2:             content=content)
bgneal@2: 
bgneal@2:     with open(dst, 'w') as fp:
bgneal@2:         fp.write(post)
bgneal@2: 
bgneal@2: 
bgneal@2: if __name__ == '__main__':
bgneal@2:     for name in os.listdir(SRC_DIR):
bgneal@2:         if name.endswith('.rst'):
bgneal@2:             src = os.path.join(SRC_DIR, name)
bgneal@2:             dst = os.path.join(DST_DIR, name)
bgneal@2: 
bgneal@2:             try:
bgneal@2:                 convert(src, dst)
bgneal@2:             except ConvertError as ex:
bgneal@2:                 print "Error converting {}: {}".format(name, ex)