|
1
|
|
|
#!/usr/bin/env python |
|
2
|
|
|
|
|
3
|
|
|
""" |
|
4
|
|
|
Pandoc filter to create lists of all kinds |
|
5
|
|
|
""" |
|
6
|
|
|
|
|
7
|
|
|
from pandocfilters import toJSONFilters, walk, Str, Plain, Link, BulletList, Para, RawInline |
|
8
|
|
|
from functools import reduce |
|
9
|
|
|
from copy import deepcopy |
|
10
|
|
|
import io |
|
11
|
|
|
import sys |
|
12
|
|
|
import codecs |
|
13
|
|
|
import json |
|
14
|
|
|
import re |
|
15
|
|
|
import unicodedata |
|
16
|
|
|
import subprocess |
|
17
|
|
|
|
|
18
|
|
|
collections = {} |
|
19
|
|
|
headers = [0, 0, 0, 0, 0, 0] |
|
20
|
|
|
headers2 = [0, 0, 0, 0, 0, 0] |
|
21
|
|
|
|
|
22
|
|
|
def stringify(x, format): |
|
23
|
|
|
"""Walks the tree x and returns concatenated string content, |
|
24
|
|
|
leaving out all formatting. |
|
25
|
|
|
""" |
|
26
|
|
|
result = [] |
|
27
|
|
|
|
|
28
|
|
|
def go(key, val, format, meta): |
|
29
|
|
|
if key in ['Str', 'MetaString']: |
|
30
|
|
|
result.append(val) |
|
31
|
|
|
elif key == 'Code': |
|
32
|
|
|
result.append(val[1]) |
|
33
|
|
|
elif key == 'Math': |
|
34
|
|
|
# Modified from the stringify function in the pandocfilter package |
|
35
|
|
|
if format == 'latex': |
|
36
|
|
|
result.append('$' + val[1] + '$') |
|
37
|
|
|
else: |
|
38
|
|
|
result.append(val[1]) |
|
39
|
|
|
elif key == 'LineBreak': |
|
40
|
|
|
result.append(" ") |
|
41
|
|
|
elif key == 'Space': |
|
42
|
|
|
result.append(" ") |
|
43
|
|
|
elif key == 'Note': |
|
44
|
|
|
# Do not stringify value from Note node |
|
45
|
|
|
del val[:] |
|
46
|
|
|
|
|
47
|
|
|
walk(x, go, format, {}) |
|
48
|
|
|
return ''.join(result) |
|
49
|
|
|
|
|
50
|
|
|
def collect(key, value, format, meta): |
|
51
|
|
|
global headers |
|
52
|
|
|
|
|
53
|
|
|
# Is it a header? Keep the correct numbered headers in the headers array |
|
54
|
|
|
if key == 'Header': |
|
55
|
|
|
[level, [id, classes, attributes], content] = value |
|
56
|
|
|
if 'unnumbered' not in classes: |
|
57
|
|
|
headers[level - 1] = headers[level - 1] + 1 |
|
58
|
|
|
for index in range(level, 6): |
|
59
|
|
|
headers[index] = 0 |
|
60
|
|
|
|
|
61
|
|
|
# Is it a link with a right tag? |
|
62
|
|
|
elif key == 'Span': |
|
63
|
|
|
|
|
64
|
|
|
# Get the Span |
|
65
|
|
|
[[anchor, classes, other], text] = value |
|
66
|
|
|
|
|
67
|
|
|
# Is the anchor correct? |
|
68
|
|
|
result = re.match('^([a-zA-Z][\w.-]*):([\w.-]+)$', anchor) |
|
69
|
|
|
if result: |
|
70
|
|
|
global collections |
|
71
|
|
|
|
|
72
|
|
|
# Compute the name |
|
73
|
|
|
name = result.group(1) |
|
74
|
|
|
|
|
75
|
|
|
# Compute the identifier |
|
76
|
|
|
identifier = result.group(2) |
|
77
|
|
|
|
|
78
|
|
|
# Store the new item |
|
79
|
|
|
string = stringify(deepcopy(text), format) |
|
80
|
|
|
|
|
81
|
|
|
# Prepare the names |
|
82
|
|
|
names = [] |
|
83
|
|
|
|
|
84
|
|
|
# Add the atomic name to the list |
|
85
|
|
|
names.append(name) |
|
86
|
|
|
|
|
87
|
|
|
# Prepare the latex output |
|
88
|
|
|
if format == 'latex': |
|
89
|
|
|
latex = '\\phantomsection\\addcontentsline{' + name + '}{figure}{' + string + '}' |
|
90
|
|
|
|
|
91
|
|
|
# Loop on all the headers |
|
92
|
|
|
for i in [0, 1, 2, 3, 4, 5]: |
|
93
|
|
|
if headers[i] > 0: |
|
94
|
|
|
# Add an alternate name to the list |
|
95
|
|
|
altName = name + ':' + '.'.join(map(str, headers[:i+1])) |
|
96
|
|
|
names.append(altName) |
|
97
|
|
|
if format == 'latex': |
|
98
|
|
|
# Complete the latex output |
|
99
|
|
|
latex = latex + '\\phantomsection\\addcontentsline{' + altName + '}{figure}{' + string + '}' |
|
100
|
|
|
latex = latex + '\\phantomsection\\addcontentsline{' + altName + '_}{figure}{' + string + '}' |
|
101
|
|
|
else: |
|
102
|
|
|
break |
|
103
|
|
|
|
|
104
|
|
|
for name in names: |
|
105
|
|
|
# Prepare the new collections if needed |
|
106
|
|
|
if name not in collections: |
|
107
|
|
|
collections[name] = [] |
|
108
|
|
|
collections[name].append({'identifier': identifier, 'text': string}) |
|
109
|
|
|
|
|
110
|
|
|
# Special case for LaTeX output |
|
111
|
|
|
if format == 'latex': |
|
112
|
|
|
text.insert(0, RawInline('tex', latex)) |
|
113
|
|
|
value[1] = text |
|
114
|
|
|
|
|
115
|
|
|
def listof(key, value, format, meta): |
|
116
|
|
|
global headers2 |
|
117
|
|
|
|
|
118
|
|
|
# Is it a header? |
|
119
|
|
|
if key == 'Header': |
|
120
|
|
|
[level, [id, classes, attributes], content] = value |
|
121
|
|
|
if 'unnumbered' not in classes: |
|
122
|
|
|
headers2[level - 1] = headers2[level - 1] + 1 |
|
123
|
|
|
for index in range(level, 6): |
|
124
|
|
|
headers2[index] = 0 |
|
125
|
|
|
|
|
126
|
|
|
# Is it a paragraph with only one string? |
|
127
|
|
|
if key == 'Para' and len(value) == 1 and value[0]['t'] == 'Str': |
|
128
|
|
|
|
|
129
|
|
|
# Is it {tag}? |
|
130
|
|
|
result = re.match('^{(?P<name>(?P<prefix>[a-zA-Z][\w.-]*)(?P<section>\:((?P<sharp>#(\.#)*)|(\d+(\.\d+)*)))?)}$', value[0]['c']) |
|
131
|
|
|
if result: |
|
132
|
|
|
|
|
133
|
|
|
prefix = result.group('prefix') |
|
134
|
|
|
|
|
135
|
|
|
# Get the collection name |
|
136
|
|
|
if result.group('sharp') == None: |
|
137
|
|
|
name = result.group('name') |
|
138
|
|
|
else: |
|
139
|
|
|
level = (len(result.group('sharp')) - 1) // 2 + 1 |
|
140
|
|
|
name = prefix + ':' + '.'.join(map(str, headers2[:level])) |
|
141
|
|
|
|
|
142
|
|
|
# Is it an existing collection |
|
143
|
|
|
if name in collections: |
|
144
|
|
|
|
|
145
|
|
|
if format == 'latex': |
|
146
|
|
|
# Special case for LaTeX output |
|
147
|
|
|
if 'toccolor' in meta: |
|
148
|
|
|
linkcolor = '\\hypersetup{linkcolor=' + stringify(meta['toccolor']['c'], format) + '}' |
|
149
|
|
|
else: |
|
150
|
|
|
linkcolor = '\\hypersetup{linkcolor=black}' |
|
151
|
|
|
if result.group('sharp') == None: |
|
152
|
|
|
suffix = '' |
|
153
|
|
|
else: |
|
154
|
|
|
suffix = '_' |
|
155
|
|
|
return Para([RawInline('tex', linkcolor + '\\makeatletter\\@starttoc{' + name + suffix + '}\\makeatother')]) |
|
156
|
|
|
|
|
157
|
|
|
else: |
|
158
|
|
|
# Prepare the list |
|
159
|
|
|
elements = [] |
|
160
|
|
|
|
|
161
|
|
|
# Loop on the collection |
|
162
|
|
|
for value in collections[name]: |
|
163
|
|
|
|
|
164
|
|
|
# Add an item to the list |
|
165
|
|
|
if pandocVersion() < '1.16': |
|
166
|
|
|
# pandoc 1.15 |
|
167
|
|
|
link = Link([Str(value['text'])], ['#' + prefix + ':' + value['identifier'], '']) |
|
168
|
|
|
else: |
|
169
|
|
|
# pandoc 1.16 |
|
170
|
|
|
link = Link(['', [], []], [Str(value['text'])], ['#' + prefix + ':' + value['identifier'], '']) |
|
171
|
|
|
|
|
172
|
|
|
elements.append([Plain([link])]) |
|
173
|
|
|
|
|
174
|
|
|
# Return a bullet list |
|
175
|
|
|
return BulletList(elements) |
|
176
|
|
|
|
|
177
|
|
|
# Special case where the paragraph start with '{{...' |
|
178
|
|
|
elif re.match('^{{[a-zA-Z][\w.-]*}$', value[0]['c']): |
|
179
|
|
|
value[0]['c'] = value[0]['c'][1:] |
|
180
|
|
|
|
|
181
|
|
|
def pandocVersion(): |
|
182
|
|
|
if not hasattr(pandocVersion, 'value'): |
|
183
|
|
|
p = subprocess.Popen(['pandoc', '-v'], stdout=subprocess.PIPE,stderr=subprocess.PIPE) |
|
184
|
|
|
out, err = p.communicate() |
|
185
|
|
|
pandocVersion.value = re.search(b'pandoc (?P<version>.*)', out).group('version').decode('utf-8') |
|
186
|
|
|
return pandocVersion.value |
|
187
|
|
|
|
|
188
|
|
|
|
|
189
|
|
|
def main(): |
|
190
|
|
|
toJSONFilters([collect, listof]) |
|
191
|
|
|
|
|
192
|
|
|
if __name__ == '__main__': |
|
193
|
|
|
main() |
|
194
|
|
|
|
|
195
|
|
|
|
chdemko /
pandoc-listof
