1 | #!/usr/bin/env python2
|
2 | """split_doc.py."""
|
3 | from __future__ import print_function
|
4 |
|
5 | import json
|
6 | import optparse
|
7 | import re
|
8 | import sys
|
9 |
|
10 | DATE_RE = re.compile(r'(\d\d\d\d) / (\d\d) / (\d\d)', re.VERBOSE)
|
11 |
|
12 | META_RE = re.compile(r'(\S+): [ ]* (.*)', re.VERBOSE)
|
13 |
|
14 |
|
15 | def SplitDocument(default_vals, entry_f, meta_f, content_f, strict=False):
|
16 | """Split a document into metadata JSON and content Markdown.
|
17 |
|
18 | Used for blog posts and index.md / cross-ref.md.
|
19 | """
|
20 | first_line = entry_f.readline()
|
21 | if strict and first_line.strip() != '---':
|
22 | raise RuntimeError("Document should start with --- (got %r)" %
|
23 | first_line)
|
24 |
|
25 | meta = {}
|
26 |
|
27 | # TODO: if first_line is ---, then read metadata in key: value format.
|
28 | if first_line.strip() == '---':
|
29 | while True:
|
30 | line = entry_f.readline().strip()
|
31 | if line == '---':
|
32 | break
|
33 | m = META_RE.match(line)
|
34 | if not m:
|
35 | raise RuntimeError('Invalid metadata line %r' % line)
|
36 | name, value = m.groups()
|
37 |
|
38 | if name == 'date':
|
39 | m2 = DATE_RE.match(value)
|
40 | if not m2:
|
41 | raise RuntimeError('Invalid date %r' % value)
|
42 | year, month, day = m2.groups()
|
43 | meta['year'] = int(year)
|
44 | meta['month'] = int(month)
|
45 | meta['day'] = int(day)
|
46 |
|
47 | elif name == 'updated_date':
|
48 | m2 = DATE_RE.match(value)
|
49 | if not m2:
|
50 | raise RuntimeError('Invalid date %r' % value)
|
51 | year, month, day = m2.groups()
|
52 | meta['updated_year'] = int(year)
|
53 | meta['updated_month'] = int(month)
|
54 | meta['updated_day'] = int(day)
|
55 |
|
56 | else:
|
57 | meta[name] = value
|
58 |
|
59 | #print('line = %r' % line, file=sys.stderr)
|
60 | while True:
|
61 | first_nonempty = entry_f.readline()
|
62 | if first_nonempty.strip() != '':
|
63 | break
|
64 |
|
65 | else:
|
66 | if first_line:
|
67 | first_nonempty = first_line
|
68 | else:
|
69 | while True:
|
70 | first_nonempty = entry_f.readline()
|
71 | if first_nonempty.strip() != '':
|
72 | break
|
73 |
|
74 | # Invariant: we've read the first non-empty line here. Now we need to see if
|
75 | # it's the title.
|
76 |
|
77 | #print('first_nonempty = %r' % first_nonempty, file=sys.stderr)
|
78 |
|
79 | line_two = entry_f.readline()
|
80 | if re.match('=+', line_two):
|
81 | meta['title'] = first_nonempty.strip()
|
82 |
|
83 | # Fill in defaults after parsing all values.
|
84 | for name, value in default_vals.iteritems():
|
85 | if name not in meta:
|
86 | meta[name] = value
|
87 |
|
88 | json.dump(meta, meta_f, indent=2)
|
89 |
|
90 | # Read the rest of the file and write it
|
91 | contents = entry_f.read()
|
92 |
|
93 | content_f.write(first_nonempty)
|
94 | content_f.write(line_two)
|
95 |
|
96 | content_f.write(contents)
|
97 |
|
98 | comments_url = meta.get('comments_url', '')
|
99 | if comments_url:
|
100 | content_f.write("""
|
101 | [comments-url]: %s
|
102 |
|
103 | """ % comments_url)
|
104 |
|
105 |
|
106 | def Options():
|
107 | p = optparse.OptionParser('split_doc.py [options] input_file out_prefix')
|
108 | # Like awk -v
|
109 | p.add_option(
|
110 | '-v',
|
111 | dest='default_vals',
|
112 | action='append',
|
113 | default=[],
|
114 | help=
|
115 | "If the doc's own metadata doesn't define 'name', set it to this value"
|
116 | )
|
117 | p.add_option('-s',
|
118 | '--strict',
|
119 | dest='strict',
|
120 | action='store_true',
|
121 | default=False,
|
122 | help="Require metadata")
|
123 | return p
|
124 |
|
125 |
|
126 | def main(argv):
|
127 | o = Options()
|
128 | opts, argv = o.parse_args(argv)
|
129 |
|
130 | entry_path = argv[1] # e.g. blog/2016/11/01.md
|
131 | out_prefix = argv[2] # e.g _site/blog/2016/11/01
|
132 |
|
133 | meta_path = out_prefix + '_meta.json'
|
134 | content_path = out_prefix + '_content.md'
|
135 |
|
136 | default_vals = {}
|
137 | for pair in opts.default_vals:
|
138 | name, value = pair.split('=', 1)
|
139 | default_vals[name] = value
|
140 |
|
141 | with \
|
142 | open(entry_path) as entry_f, \
|
143 | open(meta_path, 'w') as meta_f, \
|
144 | open(content_path, 'w') as content_f:
|
145 | SplitDocument(default_vals,
|
146 | entry_f,
|
147 | meta_f,
|
148 | content_f,
|
149 | strict=opts.strict)
|
150 |
|
151 |
|
152 | if __name__ == '__main__':
|
153 | try:
|
154 | main(sys.argv)
|
155 | except RuntimeError as e:
|
156 | print('FATAL: %s' % e, file=sys.stderr)
|
157 | sys.exit(1)
|