Show More
@@ -1,832 +1,832 | |||||
1 | # minirst.py - minimal reStructuredText parser |
|
1 | # minirst.py - minimal reStructuredText parser | |
2 | # |
|
2 | # | |
3 | # Copyright 2009, 2010 Matt Mackall <mpm@selenic.com> and others |
|
3 | # Copyright 2009, 2010 Matt Mackall <mpm@selenic.com> and others | |
4 | # |
|
4 | # | |
5 | # This software may be used and distributed according to the terms of the |
|
5 | # This software may be used and distributed according to the terms of the | |
6 | # GNU General Public License version 2 or any later version. |
|
6 | # GNU General Public License version 2 or any later version. | |
7 |
|
7 | |||
8 | """simplified reStructuredText parser. |
|
8 | """simplified reStructuredText parser. | |
9 |
|
9 | |||
10 | This parser knows just enough about reStructuredText to parse the |
|
10 | This parser knows just enough about reStructuredText to parse the | |
11 | Mercurial docstrings. |
|
11 | Mercurial docstrings. | |
12 |
|
12 | |||
13 | It cheats in a major way: nested blocks are not really nested. They |
|
13 | It cheats in a major way: nested blocks are not really nested. They | |
14 | are just indented blocks that look like they are nested. This relies |
|
14 | are just indented blocks that look like they are nested. This relies | |
15 | on the user to keep the right indentation for the blocks. |
|
15 | on the user to keep the right indentation for the blocks. | |
16 |
|
16 | |||
17 | Remember to update https://mercurial-scm.org/wiki/HelpStyleGuide |
|
17 | Remember to update https://mercurial-scm.org/wiki/HelpStyleGuide | |
18 | when adding support for new constructs. |
|
18 | when adding support for new constructs. | |
19 | """ |
|
19 | """ | |
20 |
|
20 | |||
21 | from __future__ import absolute_import |
|
21 | from __future__ import absolute_import | |
22 |
|
22 | |||
23 | import cgi |
|
23 | import cgi | |
24 | import re |
|
24 | import re | |
25 |
|
25 | |||
26 | from .i18n import _ |
|
26 | from .i18n import _ | |
27 | from . import ( |
|
27 | from . import ( | |
28 | encoding, |
|
28 | encoding, | |
29 | pycompat, |
|
29 | pycompat, | |
30 | util, |
|
30 | util, | |
31 | ) |
|
31 | ) | |
32 |
|
32 | |||
33 | def section(s): |
|
33 | def section(s): | |
34 | return "%s\n%s\n\n" % (s, "\"" * encoding.colwidth(s)) |
|
34 | return "%s\n%s\n\n" % (s, "\"" * encoding.colwidth(s)) | |
35 |
|
35 | |||
36 | def subsection(s): |
|
36 | def subsection(s): | |
37 | return "%s\n%s\n\n" % (s, '=' * encoding.colwidth(s)) |
|
37 | return "%s\n%s\n\n" % (s, '=' * encoding.colwidth(s)) | |
38 |
|
38 | |||
39 | def subsubsection(s): |
|
39 | def subsubsection(s): | |
40 | return "%s\n%s\n\n" % (s, "-" * encoding.colwidth(s)) |
|
40 | return "%s\n%s\n\n" % (s, "-" * encoding.colwidth(s)) | |
41 |
|
41 | |||
42 | def subsubsubsection(s): |
|
42 | def subsubsubsection(s): | |
43 | return "%s\n%s\n\n" % (s, "." * encoding.colwidth(s)) |
|
43 | return "%s\n%s\n\n" % (s, "." * encoding.colwidth(s)) | |
44 |
|
44 | |||
45 | def replace(text, substs): |
|
45 | def replace(text, substs): | |
46 | ''' |
|
46 | ''' | |
47 | Apply a list of (find, replace) pairs to a text. |
|
47 | Apply a list of (find, replace) pairs to a text. | |
48 |
|
48 | |||
49 | >>> replace("foo bar", [('f', 'F'), ('b', 'B')]) |
|
49 | >>> replace("foo bar", [('f', 'F'), ('b', 'B')]) | |
50 | 'Foo Bar' |
|
50 | 'Foo Bar' | |
51 | >>> encoding.encoding = 'latin1' |
|
51 | >>> encoding.encoding = 'latin1' | |
52 | >>> replace('\\x81\\\\', [('\\\\', '/')]) |
|
52 | >>> replace('\\x81\\\\', [('\\\\', '/')]) | |
53 | '\\x81/' |
|
53 | '\\x81/' | |
54 | >>> encoding.encoding = 'shiftjis' |
|
54 | >>> encoding.encoding = 'shiftjis' | |
55 | >>> replace('\\x81\\\\', [('\\\\', '/')]) |
|
55 | >>> replace('\\x81\\\\', [('\\\\', '/')]) | |
56 | '\\x81\\\\' |
|
56 | '\\x81\\\\' | |
57 | ''' |
|
57 | ''' | |
58 |
|
58 | |||
59 | # some character encodings (cp932 for Japanese, at least) use |
|
59 | # some character encodings (cp932 for Japanese, at least) use | |
60 | # ASCII characters other than control/alphabet/digit as a part of |
|
60 | # ASCII characters other than control/alphabet/digit as a part of | |
61 | # multi-bytes characters, so direct replacing with such characters |
|
61 | # multi-bytes characters, so direct replacing with such characters | |
62 | # on strings in local encoding causes invalid byte sequences. |
|
62 | # on strings in local encoding causes invalid byte sequences. | |
63 | utext = text.decode(pycompat.sysstr(encoding.encoding)) |
|
63 | utext = text.decode(pycompat.sysstr(encoding.encoding)) | |
64 | for f, t in substs: |
|
64 | for f, t in substs: | |
65 | utext = utext.replace(f.decode("ascii"), t.decode("ascii")) |
|
65 | utext = utext.replace(f.decode("ascii"), t.decode("ascii")) | |
66 | return utext.encode(pycompat.sysstr(encoding.encoding)) |
|
66 | return utext.encode(pycompat.sysstr(encoding.encoding)) | |
67 |
|
67 | |||
68 | _blockre = re.compile(br"\n(?:\s*\n)+") |
|
68 | _blockre = re.compile(br"\n(?:\s*\n)+") | |
69 |
|
69 | |||
70 | def findblocks(text): |
|
70 | def findblocks(text): | |
71 | """Find continuous blocks of lines in text. |
|
71 | """Find continuous blocks of lines in text. | |
72 |
|
72 | |||
73 | Returns a list of dictionaries representing the blocks. Each block |
|
73 | Returns a list of dictionaries representing the blocks. Each block | |
74 | has an 'indent' field and a 'lines' field. |
|
74 | has an 'indent' field and a 'lines' field. | |
75 | """ |
|
75 | """ | |
76 | blocks = [] |
|
76 | blocks = [] | |
77 | for b in _blockre.split(text.lstrip('\n').rstrip()): |
|
77 | for b in _blockre.split(text.lstrip('\n').rstrip()): | |
78 | lines = b.splitlines() |
|
78 | lines = b.splitlines() | |
79 | if lines: |
|
79 | if lines: | |
80 | indent = min((len(l) - len(l.lstrip())) for l in lines) |
|
80 | indent = min((len(l) - len(l.lstrip())) for l in lines) | |
81 | lines = [l[indent:] for l in lines] |
|
81 | lines = [l[indent:] for l in lines] | |
82 | blocks.append({'indent': indent, 'lines': lines}) |
|
82 | blocks.append({'indent': indent, 'lines': lines}) | |
83 | return blocks |
|
83 | return blocks | |
84 |
|
84 | |||
85 | def findliteralblocks(blocks): |
|
85 | def findliteralblocks(blocks): | |
86 | """Finds literal blocks and adds a 'type' field to the blocks. |
|
86 | """Finds literal blocks and adds a 'type' field to the blocks. | |
87 |
|
87 | |||
88 | Literal blocks are given the type 'literal', all other blocks are |
|
88 | Literal blocks are given the type 'literal', all other blocks are | |
89 | given type the 'paragraph'. |
|
89 | given type the 'paragraph'. | |
90 | """ |
|
90 | """ | |
91 | i = 0 |
|
91 | i = 0 | |
92 | while i < len(blocks): |
|
92 | while i < len(blocks): | |
93 | # Searching for a block that looks like this: |
|
93 | # Searching for a block that looks like this: | |
94 | # |
|
94 | # | |
95 | # +------------------------------+ |
|
95 | # +------------------------------+ | |
96 | # | paragraph | |
|
96 | # | paragraph | | |
97 | # | (ends with "::") | |
|
97 | # | (ends with "::") | | |
98 | # +------------------------------+ |
|
98 | # +------------------------------+ | |
99 | # +---------------------------+ |
|
99 | # +---------------------------+ | |
100 | # | indented literal block | |
|
100 | # | indented literal block | | |
101 | # +---------------------------+ |
|
101 | # +---------------------------+ | |
102 | blocks[i]['type'] = 'paragraph' |
|
102 | blocks[i]['type'] = 'paragraph' | |
103 | if blocks[i]['lines'][-1].endswith('::') and i + 1 < len(blocks): |
|
103 | if blocks[i]['lines'][-1].endswith('::') and i + 1 < len(blocks): | |
104 | indent = blocks[i]['indent'] |
|
104 | indent = blocks[i]['indent'] | |
105 | adjustment = blocks[i + 1]['indent'] - indent |
|
105 | adjustment = blocks[i + 1]['indent'] - indent | |
106 |
|
106 | |||
107 | if blocks[i]['lines'] == ['::']: |
|
107 | if blocks[i]['lines'] == ['::']: | |
108 | # Expanded form: remove block |
|
108 | # Expanded form: remove block | |
109 | del blocks[i] |
|
109 | del blocks[i] | |
110 | i -= 1 |
|
110 | i -= 1 | |
111 | elif blocks[i]['lines'][-1].endswith(' ::'): |
|
111 | elif blocks[i]['lines'][-1].endswith(' ::'): | |
112 | # Partially minimized form: remove space and both |
|
112 | # Partially minimized form: remove space and both | |
113 | # colons. |
|
113 | # colons. | |
114 | blocks[i]['lines'][-1] = blocks[i]['lines'][-1][:-3] |
|
114 | blocks[i]['lines'][-1] = blocks[i]['lines'][-1][:-3] | |
115 | elif len(blocks[i]['lines']) == 1 and \ |
|
115 | elif len(blocks[i]['lines']) == 1 and \ | |
116 | blocks[i]['lines'][0].lstrip(' ').startswith('.. ') and \ |
|
116 | blocks[i]['lines'][0].lstrip(' ').startswith('.. ') and \ | |
117 | blocks[i]['lines'][0].find(' ', 3) == -1: |
|
117 | blocks[i]['lines'][0].find(' ', 3) == -1: | |
118 | # directive on its own line, not a literal block |
|
118 | # directive on its own line, not a literal block | |
119 | i += 1 |
|
119 | i += 1 | |
120 | continue |
|
120 | continue | |
121 | else: |
|
121 | else: | |
122 | # Fully minimized form: remove just one colon. |
|
122 | # Fully minimized form: remove just one colon. | |
123 | blocks[i]['lines'][-1] = blocks[i]['lines'][-1][:-1] |
|
123 | blocks[i]['lines'][-1] = blocks[i]['lines'][-1][:-1] | |
124 |
|
124 | |||
125 | # List items are formatted with a hanging indent. We must |
|
125 | # List items are formatted with a hanging indent. We must | |
126 | # correct for this here while we still have the original |
|
126 | # correct for this here while we still have the original | |
127 | # information on the indentation of the subsequent literal |
|
127 | # information on the indentation of the subsequent literal | |
128 | # blocks available. |
|
128 | # blocks available. | |
129 | m = _bulletre.match(blocks[i]['lines'][0]) |
|
129 | m = _bulletre.match(blocks[i]['lines'][0]) | |
130 | if m: |
|
130 | if m: | |
131 | indent += m.end() |
|
131 | indent += m.end() | |
132 | adjustment -= m.end() |
|
132 | adjustment -= m.end() | |
133 |
|
133 | |||
134 | # Mark the following indented blocks. |
|
134 | # Mark the following indented blocks. | |
135 | while i + 1 < len(blocks) and blocks[i + 1]['indent'] > indent: |
|
135 | while i + 1 < len(blocks) and blocks[i + 1]['indent'] > indent: | |
136 | blocks[i + 1]['type'] = 'literal' |
|
136 | blocks[i + 1]['type'] = 'literal' | |
137 | blocks[i + 1]['indent'] -= adjustment |
|
137 | blocks[i + 1]['indent'] -= adjustment | |
138 | i += 1 |
|
138 | i += 1 | |
139 | i += 1 |
|
139 | i += 1 | |
140 | return blocks |
|
140 | return blocks | |
141 |
|
141 | |||
142 | _bulletre = re.compile(br'(\*|-|[0-9A-Za-z]+\.|\(?[0-9A-Za-z]+\)|\|) ') |
|
142 | _bulletre = re.compile(br'(\*|-|[0-9A-Za-z]+\.|\(?[0-9A-Za-z]+\)|\|) ') | |
143 | _optionre = re.compile(br'^(-([a-zA-Z0-9]), )?(--[a-z0-9-]+)' |
|
143 | _optionre = re.compile(br'^(-([a-zA-Z0-9]), )?(--[a-z0-9-]+)' | |
144 | br'((.*) +)(.*)$') |
|
144 | br'((.*) +)(.*)$') | |
145 | _fieldre = re.compile(br':(?![: ])([^:]*)(?<! ):[ ]+(.*)') |
|
145 | _fieldre = re.compile(br':(?![: ])([^:]*)(?<! ):[ ]+(.*)') | |
146 | _definitionre = re.compile(br'[^ ]') |
|
146 | _definitionre = re.compile(br'[^ ]') | |
147 | _tablere = re.compile(br'(=+\s+)*=+') |
|
147 | _tablere = re.compile(br'(=+\s+)*=+') | |
148 |
|
148 | |||
149 | def splitparagraphs(blocks): |
|
149 | def splitparagraphs(blocks): | |
150 | """Split paragraphs into lists.""" |
|
150 | """Split paragraphs into lists.""" | |
151 | # Tuples with (list type, item regexp, single line items?). Order |
|
151 | # Tuples with (list type, item regexp, single line items?). Order | |
152 | # matters: definition lists has the least specific regexp and must |
|
152 | # matters: definition lists has the least specific regexp and must | |
153 | # come last. |
|
153 | # come last. | |
154 | listtypes = [('bullet', _bulletre, True), |
|
154 | listtypes = [('bullet', _bulletre, True), | |
155 | ('option', _optionre, True), |
|
155 | ('option', _optionre, True), | |
156 | ('field', _fieldre, True), |
|
156 | ('field', _fieldre, True), | |
157 | ('definition', _definitionre, False)] |
|
157 | ('definition', _definitionre, False)] | |
158 |
|
158 | |||
159 | def match(lines, i, itemre, singleline): |
|
159 | def match(lines, i, itemre, singleline): | |
160 | """Does itemre match an item at line i? |
|
160 | """Does itemre match an item at line i? | |
161 |
|
161 | |||
162 | A list item can be followed by an indented line or another list |
|
162 | A list item can be followed by an indented line or another list | |
163 | item (but only if singleline is True). |
|
163 | item (but only if singleline is True). | |
164 | """ |
|
164 | """ | |
165 | line1 = lines[i] |
|
165 | line1 = lines[i] | |
166 | line2 = i + 1 < len(lines) and lines[i + 1] or '' |
|
166 | line2 = i + 1 < len(lines) and lines[i + 1] or '' | |
167 | if not itemre.match(line1): |
|
167 | if not itemre.match(line1): | |
168 | return False |
|
168 | return False | |
169 | if singleline: |
|
169 | if singleline: | |
170 | return line2 == '' or line2[0] == ' ' or itemre.match(line2) |
|
170 | return line2 == '' or line2[0] == ' ' or itemre.match(line2) | |
171 | else: |
|
171 | else: | |
172 | return line2.startswith(' ') |
|
172 | return line2.startswith(' ') | |
173 |
|
173 | |||
174 | i = 0 |
|
174 | i = 0 | |
175 | while i < len(blocks): |
|
175 | while i < len(blocks): | |
176 | if blocks[i]['type'] == 'paragraph': |
|
176 | if blocks[i]['type'] == 'paragraph': | |
177 | lines = blocks[i]['lines'] |
|
177 | lines = blocks[i]['lines'] | |
178 | for type, itemre, singleline in listtypes: |
|
178 | for type, itemre, singleline in listtypes: | |
179 | if match(lines, 0, itemre, singleline): |
|
179 | if match(lines, 0, itemre, singleline): | |
180 | items = [] |
|
180 | items = [] | |
181 | for j, line in enumerate(lines): |
|
181 | for j, line in enumerate(lines): | |
182 | if match(lines, j, itemre, singleline): |
|
182 | if match(lines, j, itemre, singleline): | |
183 | items.append({'type': type, 'lines': [], |
|
183 | items.append({'type': type, 'lines': [], | |
184 | 'indent': blocks[i]['indent']}) |
|
184 | 'indent': blocks[i]['indent']}) | |
185 | items[-1]['lines'].append(line) |
|
185 | items[-1]['lines'].append(line) | |
186 | blocks[i:i + 1] = items |
|
186 | blocks[i:i + 1] = items | |
187 | break |
|
187 | break | |
188 | i += 1 |
|
188 | i += 1 | |
189 | return blocks |
|
189 | return blocks | |
190 |
|
190 | |||
191 | _fieldwidth = 14 |
|
191 | _fieldwidth = 14 | |
192 |
|
192 | |||
193 | def updatefieldlists(blocks): |
|
193 | def updatefieldlists(blocks): | |
194 | """Find key for field lists.""" |
|
194 | """Find key for field lists.""" | |
195 | i = 0 |
|
195 | i = 0 | |
196 | while i < len(blocks): |
|
196 | while i < len(blocks): | |
197 | if blocks[i]['type'] != 'field': |
|
197 | if blocks[i]['type'] != 'field': | |
198 | i += 1 |
|
198 | i += 1 | |
199 | continue |
|
199 | continue | |
200 |
|
200 | |||
201 | j = i |
|
201 | j = i | |
202 | while j < len(blocks) and blocks[j]['type'] == 'field': |
|
202 | while j < len(blocks) and blocks[j]['type'] == 'field': | |
203 | m = _fieldre.match(blocks[j]['lines'][0]) |
|
203 | m = _fieldre.match(blocks[j]['lines'][0]) | |
204 | key, rest = m.groups() |
|
204 | key, rest = m.groups() | |
205 | blocks[j]['lines'][0] = rest |
|
205 | blocks[j]['lines'][0] = rest | |
206 | blocks[j]['key'] = key |
|
206 | blocks[j]['key'] = key | |
207 | j += 1 |
|
207 | j += 1 | |
208 |
|
208 | |||
209 | i = j + 1 |
|
209 | i = j + 1 | |
210 |
|
210 | |||
211 | return blocks |
|
211 | return blocks | |
212 |
|
212 | |||
213 | def updateoptionlists(blocks): |
|
213 | def updateoptionlists(blocks): | |
214 | i = 0 |
|
214 | i = 0 | |
215 | while i < len(blocks): |
|
215 | while i < len(blocks): | |
216 | if blocks[i]['type'] != 'option': |
|
216 | if blocks[i]['type'] != 'option': | |
217 | i += 1 |
|
217 | i += 1 | |
218 | continue |
|
218 | continue | |
219 |
|
219 | |||
220 | optstrwidth = 0 |
|
220 | optstrwidth = 0 | |
221 | j = i |
|
221 | j = i | |
222 | while j < len(blocks) and blocks[j]['type'] == 'option': |
|
222 | while j < len(blocks) and blocks[j]['type'] == 'option': | |
223 | m = _optionre.match(blocks[j]['lines'][0]) |
|
223 | m = _optionre.match(blocks[j]['lines'][0]) | |
224 |
|
224 | |||
225 | shortoption = m.group(2) |
|
225 | shortoption = m.group(2) | |
226 | group3 = m.group(3) |
|
226 | group3 = m.group(3) | |
227 | longoption = group3[2:].strip() |
|
227 | longoption = group3[2:].strip() | |
228 | desc = m.group(6).strip() |
|
228 | desc = m.group(6).strip() | |
229 | longoptionarg = m.group(5).strip() |
|
229 | longoptionarg = m.group(5).strip() | |
230 | blocks[j]['lines'][0] = desc |
|
230 | blocks[j]['lines'][0] = desc | |
231 |
|
231 | |||
232 | noshortop = '' |
|
232 | noshortop = '' | |
233 | if not shortoption: |
|
233 | if not shortoption: | |
234 | noshortop = ' ' |
|
234 | noshortop = ' ' | |
235 |
|
235 | |||
236 | opt = "%s%s" % (shortoption and "-%s " % shortoption or '', |
|
236 | opt = "%s%s" % (shortoption and "-%s " % shortoption or '', | |
237 | ("%s--%s %s") % (noshortop, longoption, |
|
237 | ("%s--%s %s") % (noshortop, longoption, | |
238 | longoptionarg)) |
|
238 | longoptionarg)) | |
239 | opt = opt.rstrip() |
|
239 | opt = opt.rstrip() | |
240 | blocks[j]['optstr'] = opt |
|
240 | blocks[j]['optstr'] = opt | |
241 | optstrwidth = max(optstrwidth, encoding.colwidth(opt)) |
|
241 | optstrwidth = max(optstrwidth, encoding.colwidth(opt)) | |
242 | j += 1 |
|
242 | j += 1 | |
243 |
|
243 | |||
244 | for block in blocks[i:j]: |
|
244 | for block in blocks[i:j]: | |
245 | block['optstrwidth'] = optstrwidth |
|
245 | block['optstrwidth'] = optstrwidth | |
246 | i = j + 1 |
|
246 | i = j + 1 | |
247 | return blocks |
|
247 | return blocks | |
248 |
|
248 | |||
249 | def prunecontainers(blocks, keep): |
|
249 | def prunecontainers(blocks, keep): | |
250 | """Prune unwanted containers. |
|
250 | """Prune unwanted containers. | |
251 |
|
251 | |||
252 | The blocks must have a 'type' field, i.e., they should have been |
|
252 | The blocks must have a 'type' field, i.e., they should have been | |
253 | run through findliteralblocks first. |
|
253 | run through findliteralblocks first. | |
254 | """ |
|
254 | """ | |
255 | pruned = [] |
|
255 | pruned = [] | |
256 | i = 0 |
|
256 | i = 0 | |
257 | while i + 1 < len(blocks): |
|
257 | while i + 1 < len(blocks): | |
258 | # Searching for a block that looks like this: |
|
258 | # Searching for a block that looks like this: | |
259 | # |
|
259 | # | |
260 | # +-------+---------------------------+ |
|
260 | # +-------+---------------------------+ | |
261 | # | ".. container ::" type | |
|
261 | # | ".. container ::" type | | |
262 | # +---+ | |
|
262 | # +---+ | | |
263 | # | blocks | |
|
263 | # | blocks | | |
264 | # +-------------------------------+ |
|
264 | # +-------------------------------+ | |
265 | if (blocks[i]['type'] == 'paragraph' and |
|
265 | if (blocks[i]['type'] == 'paragraph' and | |
266 | blocks[i]['lines'][0].startswith('.. container::')): |
|
266 | blocks[i]['lines'][0].startswith('.. container::')): | |
267 | indent = blocks[i]['indent'] |
|
267 | indent = blocks[i]['indent'] | |
268 | adjustment = blocks[i + 1]['indent'] - indent |
|
268 | adjustment = blocks[i + 1]['indent'] - indent | |
269 | containertype = blocks[i]['lines'][0][15:] |
|
269 | containertype = blocks[i]['lines'][0][15:] | |
270 | prune = True |
|
270 | prune = True | |
271 | for c in keep: |
|
271 | for c in keep: | |
272 | if c in containertype.split('.'): |
|
272 | if c in containertype.split('.'): | |
273 | prune = False |
|
273 | prune = False | |
274 | if prune: |
|
274 | if prune: | |
275 | pruned.append(containertype) |
|
275 | pruned.append(containertype) | |
276 |
|
276 | |||
277 | # Always delete "..container:: type" block |
|
277 | # Always delete "..container:: type" block | |
278 | del blocks[i] |
|
278 | del blocks[i] | |
279 | j = i |
|
279 | j = i | |
280 | i -= 1 |
|
280 | i -= 1 | |
281 | while j < len(blocks) and blocks[j]['indent'] > indent: |
|
281 | while j < len(blocks) and blocks[j]['indent'] > indent: | |
282 | if prune: |
|
282 | if prune: | |
283 | del blocks[j] |
|
283 | del blocks[j] | |
284 | else: |
|
284 | else: | |
285 | blocks[j]['indent'] -= adjustment |
|
285 | blocks[j]['indent'] -= adjustment | |
286 | j += 1 |
|
286 | j += 1 | |
287 | i += 1 |
|
287 | i += 1 | |
288 | return blocks, pruned |
|
288 | return blocks, pruned | |
289 |
|
289 | |||
290 | _sectionre = re.compile(br"""^([-=`:.'"~^_*+#])\1+$""") |
|
290 | _sectionre = re.compile(br"""^([-=`:.'"~^_*+#])\1+$""") | |
291 |
|
291 | |||
292 | def findtables(blocks): |
|
292 | def findtables(blocks): | |
293 | '''Find simple tables |
|
293 | '''Find simple tables | |
294 |
|
294 | |||
295 | Only simple one-line table elements are supported |
|
295 | Only simple one-line table elements are supported | |
296 | ''' |
|
296 | ''' | |
297 |
|
297 | |||
298 | for block in blocks: |
|
298 | for block in blocks: | |
299 | # Searching for a block that looks like this: |
|
299 | # Searching for a block that looks like this: | |
300 | # |
|
300 | # | |
301 | # === ==== === |
|
301 | # === ==== === | |
302 | # A B C |
|
302 | # A B C | |
303 | # === ==== === <- optional |
|
303 | # === ==== === <- optional | |
304 | # 1 2 3 |
|
304 | # 1 2 3 | |
305 | # x y z |
|
305 | # x y z | |
306 | # === ==== === |
|
306 | # === ==== === | |
307 | if (block['type'] == 'paragraph' and |
|
307 | if (block['type'] == 'paragraph' and | |
308 | len(block['lines']) > 2 and |
|
308 | len(block['lines']) > 2 and | |
309 | _tablere.match(block['lines'][0]) and |
|
309 | _tablere.match(block['lines'][0]) and | |
310 | block['lines'][0] == block['lines'][-1]): |
|
310 | block['lines'][0] == block['lines'][-1]): | |
311 | block['type'] = 'table' |
|
311 | block['type'] = 'table' | |
312 | block['header'] = False |
|
312 | block['header'] = False | |
313 | div = block['lines'][0] |
|
313 | div = block['lines'][0] | |
314 |
|
314 | |||
315 | # column markers are ASCII so we can calculate column |
|
315 | # column markers are ASCII so we can calculate column | |
316 | # position in bytes |
|
316 | # position in bytes | |
317 | columns = [x for x in xrange(len(div)) |
|
317 | columns = [x for x in xrange(len(div)) | |
318 | if div[x] == '=' and (x == 0 or div[x - 1] == ' ')] |
|
318 | if div[x] == '=' and (x == 0 or div[x - 1] == ' ')] | |
319 | rows = [] |
|
319 | rows = [] | |
320 | for l in block['lines'][1:-1]: |
|
320 | for l in block['lines'][1:-1]: | |
321 | if l == div: |
|
321 | if l == div: | |
322 | block['header'] = True |
|
322 | block['header'] = True | |
323 | continue |
|
323 | continue | |
324 | row = [] |
|
324 | row = [] | |
325 | # we measure columns not in bytes or characters but in |
|
325 | # we measure columns not in bytes or characters but in | |
326 | # colwidth which makes things tricky |
|
326 | # colwidth which makes things tricky | |
327 | pos = columns[0] # leading whitespace is bytes |
|
327 | pos = columns[0] # leading whitespace is bytes | |
328 | for n, start in enumerate(columns): |
|
328 | for n, start in enumerate(columns): | |
329 | if n + 1 < len(columns): |
|
329 | if n + 1 < len(columns): | |
330 | width = columns[n + 1] - start |
|
330 | width = columns[n + 1] - start | |
331 | v = encoding.getcols(l, pos, width) # gather columns |
|
331 | v = encoding.getcols(l, pos, width) # gather columns | |
332 | pos += len(v) # calculate byte position of end |
|
332 | pos += len(v) # calculate byte position of end | |
333 | row.append(v.strip()) |
|
333 | row.append(v.strip()) | |
334 | else: |
|
334 | else: | |
335 | row.append(l[pos:].strip()) |
|
335 | row.append(l[pos:].strip()) | |
336 | rows.append(row) |
|
336 | rows.append(row) | |
337 |
|
337 | |||
338 | block['table'] = rows |
|
338 | block['table'] = rows | |
339 |
|
339 | |||
340 | return blocks |
|
340 | return blocks | |
341 |
|
341 | |||
342 | def findsections(blocks): |
|
342 | def findsections(blocks): | |
343 | """Finds sections. |
|
343 | """Finds sections. | |
344 |
|
344 | |||
345 | The blocks must have a 'type' field, i.e., they should have been |
|
345 | The blocks must have a 'type' field, i.e., they should have been | |
346 | run through findliteralblocks first. |
|
346 | run through findliteralblocks first. | |
347 | """ |
|
347 | """ | |
348 | for block in blocks: |
|
348 | for block in blocks: | |
349 | # Searching for a block that looks like this: |
|
349 | # Searching for a block that looks like this: | |
350 | # |
|
350 | # | |
351 | # +------------------------------+ |
|
351 | # +------------------------------+ | |
352 | # | Section title | |
|
352 | # | Section title | | |
353 | # | ------------- | |
|
353 | # | ------------- | | |
354 | # +------------------------------+ |
|
354 | # +------------------------------+ | |
355 | if (block['type'] == 'paragraph' and |
|
355 | if (block['type'] == 'paragraph' and | |
356 | len(block['lines']) == 2 and |
|
356 | len(block['lines']) == 2 and | |
357 | encoding.colwidth(block['lines'][0]) == len(block['lines'][1]) and |
|
357 | encoding.colwidth(block['lines'][0]) == len(block['lines'][1]) and | |
358 | _sectionre.match(block['lines'][1])): |
|
358 | _sectionre.match(block['lines'][1])): | |
359 | block['underline'] = block['lines'][1][0] |
|
359 | block['underline'] = block['lines'][1][0] | |
360 | block['type'] = 'section' |
|
360 | block['type'] = 'section' | |
361 | del block['lines'][1] |
|
361 | del block['lines'][1] | |
362 | return blocks |
|
362 | return blocks | |
363 |
|
363 | |||
364 | def inlineliterals(blocks): |
|
364 | def inlineliterals(blocks): | |
365 | substs = [('``', '"')] |
|
365 | substs = [('``', '"')] | |
366 | for b in blocks: |
|
366 | for b in blocks: | |
367 | if b['type'] in ('paragraph', 'section'): |
|
367 | if b['type'] in ('paragraph', 'section'): | |
368 | b['lines'] = [replace(l, substs) for l in b['lines']] |
|
368 | b['lines'] = [replace(l, substs) for l in b['lines']] | |
369 | return blocks |
|
369 | return blocks | |
370 |
|
370 | |||
371 | def hgrole(blocks): |
|
371 | def hgrole(blocks): | |
372 | substs = [(':hg:`', "'hg "), ('`', "'")] |
|
372 | substs = [(':hg:`', "'hg "), ('`', "'")] | |
373 | for b in blocks: |
|
373 | for b in blocks: | |
374 | if b['type'] in ('paragraph', 'section'): |
|
374 | if b['type'] in ('paragraph', 'section'): | |
375 | # Turn :hg:`command` into "hg command". This also works |
|
375 | # Turn :hg:`command` into "hg command". This also works | |
376 | # when there is a line break in the command and relies on |
|
376 | # when there is a line break in the command and relies on | |
377 | # the fact that we have no stray back-quotes in the input |
|
377 | # the fact that we have no stray back-quotes in the input | |
378 | # (run the blocks through inlineliterals first). |
|
378 | # (run the blocks through inlineliterals first). | |
379 | b['lines'] = [replace(l, substs) for l in b['lines']] |
|
379 | b['lines'] = [replace(l, substs) for l in b['lines']] | |
380 | return blocks |
|
380 | return blocks | |
381 |
|
381 | |||
382 | def addmargins(blocks): |
|
382 | def addmargins(blocks): | |
383 | """Adds empty blocks for vertical spacing. |
|
383 | """Adds empty blocks for vertical spacing. | |
384 |
|
384 | |||
385 | This groups bullets, options, and definitions together with no vertical |
|
385 | This groups bullets, options, and definitions together with no vertical | |
386 | space between them, and adds an empty block between all other blocks. |
|
386 | space between them, and adds an empty block between all other blocks. | |
387 | """ |
|
387 | """ | |
388 | i = 1 |
|
388 | i = 1 | |
389 | while i < len(blocks): |
|
389 | while i < len(blocks): | |
390 | if (blocks[i]['type'] == blocks[i - 1]['type'] and |
|
390 | if (blocks[i]['type'] == blocks[i - 1]['type'] and | |
391 | blocks[i]['type'] in ('bullet', 'option', 'field')): |
|
391 | blocks[i]['type'] in ('bullet', 'option', 'field')): | |
392 | i += 1 |
|
392 | i += 1 | |
393 | elif not blocks[i - 1]['lines']: |
|
393 | elif not blocks[i - 1]['lines']: | |
394 | # no lines in previous block, do not separate |
|
394 | # no lines in previous block, do not separate | |
395 | i += 1 |
|
395 | i += 1 | |
396 | else: |
|
396 | else: | |
397 | blocks.insert(i, {'lines': [''], 'indent': 0, 'type': 'margin'}) |
|
397 | blocks.insert(i, {'lines': [''], 'indent': 0, 'type': 'margin'}) | |
398 | i += 2 |
|
398 | i += 2 | |
399 | return blocks |
|
399 | return blocks | |
400 |
|
400 | |||
401 | def prunecomments(blocks): |
|
401 | def prunecomments(blocks): | |
402 | """Remove comments.""" |
|
402 | """Remove comments.""" | |
403 | i = 0 |
|
403 | i = 0 | |
404 | while i < len(blocks): |
|
404 | while i < len(blocks): | |
405 | b = blocks[i] |
|
405 | b = blocks[i] | |
406 | if b['type'] == 'paragraph' and (b['lines'][0].startswith('.. ') or |
|
406 | if b['type'] == 'paragraph' and (b['lines'][0].startswith('.. ') or | |
407 | b['lines'] == ['..']): |
|
407 | b['lines'] == ['..']): | |
408 | del blocks[i] |
|
408 | del blocks[i] | |
409 | if i < len(blocks) and blocks[i]['type'] == 'margin': |
|
409 | if i < len(blocks) and blocks[i]['type'] == 'margin': | |
410 | del blocks[i] |
|
410 | del blocks[i] | |
411 | else: |
|
411 | else: | |
412 | i += 1 |
|
412 | i += 1 | |
413 | return blocks |
|
413 | return blocks | |
414 |
|
414 | |||
415 |
|
415 | |||
416 | _admonitions = set([ |
|
416 | _admonitions = set([ | |
417 | 'admonition', |
|
417 | 'admonition', | |
418 | 'attention', |
|
418 | 'attention', | |
419 | 'caution', |
|
419 | 'caution', | |
420 | 'danger', |
|
420 | 'danger', | |
421 | 'error', |
|
421 | 'error', | |
422 | 'hint', |
|
422 | 'hint', | |
423 | 'important', |
|
423 | 'important', | |
424 | 'note', |
|
424 | 'note', | |
425 | 'tip', |
|
425 | 'tip', | |
426 | 'warning', |
|
426 | 'warning', | |
427 | ]) |
|
427 | ]) | |
428 |
|
428 | |||
429 | def findadmonitions(blocks, admonitions=None): |
|
429 | def findadmonitions(blocks, admonitions=None): | |
430 | """ |
|
430 | """ | |
431 | Makes the type of the block an admonition block if |
|
431 | Makes the type of the block an admonition block if | |
432 | the first line is an admonition directive |
|
432 | the first line is an admonition directive | |
433 | """ |
|
433 | """ | |
434 | admonitions = admonitions or _admonitions |
|
434 | admonitions = admonitions or _admonitions | |
435 |
|
435 | |||
436 | admonitionre = re.compile(br'\.\. (%s)::' % '|'.join(sorted(admonitions)), |
|
436 | admonitionre = re.compile(br'\.\. (%s)::' % '|'.join(sorted(admonitions)), | |
437 | flags=re.IGNORECASE) |
|
437 | flags=re.IGNORECASE) | |
438 |
|
438 | |||
439 | i = 0 |
|
439 | i = 0 | |
440 | while i < len(blocks): |
|
440 | while i < len(blocks): | |
441 | m = admonitionre.match(blocks[i]['lines'][0]) |
|
441 | m = admonitionre.match(blocks[i]['lines'][0]) | |
442 | if m: |
|
442 | if m: | |
443 | blocks[i]['type'] = 'admonition' |
|
443 | blocks[i]['type'] = 'admonition' | |
444 | admonitiontitle = blocks[i]['lines'][0][3:m.end() - 2].lower() |
|
444 | admonitiontitle = blocks[i]['lines'][0][3:m.end() - 2].lower() | |
445 |
|
445 | |||
446 | firstline = blocks[i]['lines'][0][m.end() + 1:] |
|
446 | firstline = blocks[i]['lines'][0][m.end() + 1:] | |
447 | if firstline: |
|
447 | if firstline: | |
448 | blocks[i]['lines'].insert(1, ' ' + firstline) |
|
448 | blocks[i]['lines'].insert(1, ' ' + firstline) | |
449 |
|
449 | |||
450 | blocks[i]['admonitiontitle'] = admonitiontitle |
|
450 | blocks[i]['admonitiontitle'] = admonitiontitle | |
451 | del blocks[i]['lines'][0] |
|
451 | del blocks[i]['lines'][0] | |
452 | i = i + 1 |
|
452 | i = i + 1 | |
453 | return blocks |
|
453 | return blocks | |
454 |
|
454 | |||
455 | _admonitiontitles = {'attention': _('Attention:'), |
|
455 | _admonitiontitles = {'attention': _('Attention:'), | |
456 | 'caution': _('Caution:'), |
|
456 | 'caution': _('Caution:'), | |
457 | 'danger': _('!Danger!') , |
|
457 | 'danger': _('!Danger!') , | |
458 | 'error': _('Error:'), |
|
458 | 'error': _('Error:'), | |
459 | 'hint': _('Hint:'), |
|
459 | 'hint': _('Hint:'), | |
460 | 'important': _('Important:'), |
|
460 | 'important': _('Important:'), | |
461 | 'note': _('Note:'), |
|
461 | 'note': _('Note:'), | |
462 | 'tip': _('Tip:'), |
|
462 | 'tip': _('Tip:'), | |
463 | 'warning': _('Warning!')} |
|
463 | 'warning': _('Warning!')} | |
464 |
|
464 | |||
465 | def formatoption(block, width): |
|
465 | def formatoption(block, width): | |
466 | desc = ' '.join(map(str.strip, block['lines'])) |
|
466 | desc = ' '.join(map(str.strip, block['lines'])) | |
467 | colwidth = encoding.colwidth(block['optstr']) |
|
467 | colwidth = encoding.colwidth(block['optstr']) | |
468 | usablewidth = width - 1 |
|
468 | usablewidth = width - 1 | |
469 | hanging = block['optstrwidth'] |
|
469 | hanging = block['optstrwidth'] | |
470 | initindent = '%s%s ' % (block['optstr'], ' ' * ((hanging - colwidth))) |
|
470 | initindent = '%s%s ' % (block['optstr'], ' ' * ((hanging - colwidth))) | |
471 | hangindent = ' ' * (encoding.colwidth(initindent) + 1) |
|
471 | hangindent = ' ' * (encoding.colwidth(initindent) + 1) | |
472 | return ' %s\n' % (util.wrap(desc, usablewidth, |
|
472 | return ' %s\n' % (util.wrap(desc, usablewidth, | |
473 | initindent=initindent, |
|
473 | initindent=initindent, | |
474 | hangindent=hangindent)) |
|
474 | hangindent=hangindent)) | |
475 |
|
475 | |||
476 | def formatblock(block, width): |
|
476 | def formatblock(block, width): | |
477 | """Format a block according to width.""" |
|
477 | """Format a block according to width.""" | |
478 | if width <= 0: |
|
478 | if width <= 0: | |
479 | width = 78 |
|
479 | width = 78 | |
480 | indent = ' ' * block['indent'] |
|
480 | indent = ' ' * block['indent'] | |
481 | if block['type'] == 'admonition': |
|
481 | if block['type'] == 'admonition': | |
482 | admonition = _admonitiontitles[block['admonitiontitle']] |
|
482 | admonition = _admonitiontitles[block['admonitiontitle']] | |
483 | if not block['lines']: |
|
483 | if not block['lines']: | |
484 | return indent + admonition + '\n' |
|
484 | return indent + admonition + '\n' | |
485 | hang = len(block['lines'][-1]) - len(block['lines'][-1].lstrip()) |
|
485 | hang = len(block['lines'][-1]) - len(block['lines'][-1].lstrip()) | |
486 |
|
486 | |||
487 | defindent = indent + hang * ' ' |
|
487 | defindent = indent + hang * ' ' | |
488 | text = ' '.join(map(str.strip, block['lines'])) |
|
488 | text = ' '.join(map(str.strip, block['lines'])) | |
489 | return '%s\n%s\n' % (indent + admonition, |
|
489 | return '%s\n%s\n' % (indent + admonition, | |
490 | util.wrap(text, width=width, |
|
490 | util.wrap(text, width=width, | |
491 | initindent=defindent, |
|
491 | initindent=defindent, | |
492 | hangindent=defindent)) |
|
492 | hangindent=defindent)) | |
493 | if block['type'] == 'margin': |
|
493 | if block['type'] == 'margin': | |
494 | return '\n' |
|
494 | return '\n' | |
495 | if block['type'] == 'literal': |
|
495 | if block['type'] == 'literal': | |
496 | indent += ' ' |
|
496 | indent += ' ' | |
497 | return indent + ('\n' + indent).join(block['lines']) + '\n' |
|
497 | return indent + ('\n' + indent).join(block['lines']) + '\n' | |
498 | if block['type'] == 'section': |
|
498 | if block['type'] == 'section': | |
499 | underline = encoding.colwidth(block['lines'][0]) * block['underline'] |
|
499 | underline = encoding.colwidth(block['lines'][0]) * block['underline'] | |
500 | return "%s%s\n%s%s\n" % (indent, block['lines'][0],indent, underline) |
|
500 | return "%s%s\n%s%s\n" % (indent, block['lines'][0],indent, underline) | |
501 | if block['type'] == 'table': |
|
501 | if block['type'] == 'table': | |
502 | table = block['table'] |
|
502 | table = block['table'] | |
503 | # compute column widths |
|
503 | # compute column widths | |
504 | widths = [max([encoding.colwidth(e) for e in c]) for c in zip(*table)] |
|
504 | widths = [max([encoding.colwidth(e) for e in c]) for c in zip(*table)] | |
505 | text = '' |
|
505 | text = '' | |
506 | span = sum(widths) + len(widths) - 1 |
|
506 | span = sum(widths) + len(widths) - 1 | |
507 | indent = ' ' * block['indent'] |
|
507 | indent = ' ' * block['indent'] | |
508 | hang = ' ' * (len(indent) + span - widths[-1]) |
|
508 | hang = ' ' * (len(indent) + span - widths[-1]) | |
509 |
|
509 | |||
510 | for row in table: |
|
510 | for row in table: | |
511 | l = [] |
|
511 | l = [] | |
512 | for w, v in zip(widths, row): |
|
512 | for w, v in zip(widths, row): | |
513 | pad = ' ' * (w - encoding.colwidth(v)) |
|
513 | pad = ' ' * (w - encoding.colwidth(v)) | |
514 | l.append(v + pad) |
|
514 | l.append(v + pad) | |
515 | l = ' '.join(l) |
|
515 | l = ' '.join(l) | |
516 | l = util.wrap(l, width=width, initindent=indent, hangindent=hang) |
|
516 | l = util.wrap(l, width=width, initindent=indent, hangindent=hang) | |
517 | if not text and block['header']: |
|
517 | if not text and block['header']: | |
518 | text = l + '\n' + indent + '-' * (min(width, span)) + '\n' |
|
518 | text = l + '\n' + indent + '-' * (min(width, span)) + '\n' | |
519 | else: |
|
519 | else: | |
520 | text += l + "\n" |
|
520 | text += l + "\n" | |
521 | return text |
|
521 | return text | |
522 | if block['type'] == 'definition': |
|
522 | if block['type'] == 'definition': | |
523 | term = indent + block['lines'][0] |
|
523 | term = indent + block['lines'][0] | |
524 | hang = len(block['lines'][-1]) - len(block['lines'][-1].lstrip()) |
|
524 | hang = len(block['lines'][-1]) - len(block['lines'][-1].lstrip()) | |
525 | defindent = indent + hang * ' ' |
|
525 | defindent = indent + hang * ' ' | |
526 | text = ' '.join(map(str.strip, block['lines'][1:])) |
|
526 | text = ' '.join(map(str.strip, block['lines'][1:])) | |
527 | return '%s\n%s\n' % (term, util.wrap(text, width=width, |
|
527 | return '%s\n%s\n' % (term, util.wrap(text, width=width, | |
528 | initindent=defindent, |
|
528 | initindent=defindent, | |
529 | hangindent=defindent)) |
|
529 | hangindent=defindent)) | |
530 | subindent = indent |
|
530 | subindent = indent | |
531 | if block['type'] == 'bullet': |
|
531 | if block['type'] == 'bullet': | |
532 | if block['lines'][0].startswith('| '): |
|
532 | if block['lines'][0].startswith('| '): | |
533 | # Remove bullet for line blocks and add no extra |
|
533 | # Remove bullet for line blocks and add no extra | |
534 | # indentation. |
|
534 | # indentation. | |
535 | block['lines'][0] = block['lines'][0][2:] |
|
535 | block['lines'][0] = block['lines'][0][2:] | |
536 | else: |
|
536 | else: | |
537 | m = _bulletre.match(block['lines'][0]) |
|
537 | m = _bulletre.match(block['lines'][0]) | |
538 | subindent = indent + m.end() * ' ' |
|
538 | subindent = indent + m.end() * ' ' | |
539 | elif block['type'] == 'field': |
|
539 | elif block['type'] == 'field': | |
540 | key = block['key'] |
|
540 | key = block['key'] | |
541 | subindent = indent + _fieldwidth * ' ' |
|
541 | subindent = indent + _fieldwidth * ' ' | |
542 | if len(key) + 2 > _fieldwidth: |
|
542 | if len(key) + 2 > _fieldwidth: | |
543 | # key too large, use full line width |
|
543 | # key too large, use full line width | |
544 | key = key.ljust(width) |
|
544 | key = key.ljust(width) | |
545 | else: |
|
545 | else: | |
546 | # key fits within field width |
|
546 | # key fits within field width | |
547 | key = key.ljust(_fieldwidth) |
|
547 | key = key.ljust(_fieldwidth) | |
548 | block['lines'][0] = key + block['lines'][0] |
|
548 | block['lines'][0] = key + block['lines'][0] | |
549 | elif block['type'] == 'option': |
|
549 | elif block['type'] == 'option': | |
550 | return formatoption(block, width) |
|
550 | return formatoption(block, width) | |
551 |
|
551 | |||
552 |
text = ' '.join(map( |
|
552 | text = ' '.join(map(bytes.strip, block['lines'])) | |
553 | return util.wrap(text, width=width, |
|
553 | return util.wrap(text, width=width, | |
554 | initindent=indent, |
|
554 | initindent=indent, | |
555 | hangindent=subindent) + '\n' |
|
555 | hangindent=subindent) + '\n' | |
556 |
|
556 | |||
557 | def formathtml(blocks): |
|
557 | def formathtml(blocks): | |
558 | """Format RST blocks as HTML""" |
|
558 | """Format RST blocks as HTML""" | |
559 |
|
559 | |||
560 | out = [] |
|
560 | out = [] | |
561 | headernest = '' |
|
561 | headernest = '' | |
562 | listnest = [] |
|
562 | listnest = [] | |
563 |
|
563 | |||
564 | def escape(s): |
|
564 | def escape(s): | |
565 | return cgi.escape(s, True) |
|
565 | return cgi.escape(s, True) | |
566 |
|
566 | |||
567 | def openlist(start, level): |
|
567 | def openlist(start, level): | |
568 | if not listnest or listnest[-1][0] != start: |
|
568 | if not listnest or listnest[-1][0] != start: | |
569 | listnest.append((start, level)) |
|
569 | listnest.append((start, level)) | |
570 | out.append('<%s>\n' % start) |
|
570 | out.append('<%s>\n' % start) | |
571 |
|
571 | |||
572 | blocks = [b for b in blocks if b['type'] != 'margin'] |
|
572 | blocks = [b for b in blocks if b['type'] != 'margin'] | |
573 |
|
573 | |||
574 | for pos, b in enumerate(blocks): |
|
574 | for pos, b in enumerate(blocks): | |
575 | btype = b['type'] |
|
575 | btype = b['type'] | |
576 | level = b['indent'] |
|
576 | level = b['indent'] | |
577 | lines = b['lines'] |
|
577 | lines = b['lines'] | |
578 |
|
578 | |||
579 | if btype == 'admonition': |
|
579 | if btype == 'admonition': | |
580 | admonition = escape(_admonitiontitles[b['admonitiontitle']]) |
|
580 | admonition = escape(_admonitiontitles[b['admonitiontitle']]) | |
581 | text = escape(' '.join(map(str.strip, lines))) |
|
581 | text = escape(' '.join(map(str.strip, lines))) | |
582 | out.append('<p>\n<b>%s</b> %s\n</p>\n' % (admonition, text)) |
|
582 | out.append('<p>\n<b>%s</b> %s\n</p>\n' % (admonition, text)) | |
583 | elif btype == 'paragraph': |
|
583 | elif btype == 'paragraph': | |
584 | out.append('<p>\n%s\n</p>\n' % escape('\n'.join(lines))) |
|
584 | out.append('<p>\n%s\n</p>\n' % escape('\n'.join(lines))) | |
585 | elif btype == 'margin': |
|
585 | elif btype == 'margin': | |
586 | pass |
|
586 | pass | |
587 | elif btype == 'literal': |
|
587 | elif btype == 'literal': | |
588 | out.append('<pre>\n%s\n</pre>\n' % escape('\n'.join(lines))) |
|
588 | out.append('<pre>\n%s\n</pre>\n' % escape('\n'.join(lines))) | |
589 | elif btype == 'section': |
|
589 | elif btype == 'section': | |
590 | i = b['underline'] |
|
590 | i = b['underline'] | |
591 | if i not in headernest: |
|
591 | if i not in headernest: | |
592 | headernest += i |
|
592 | headernest += i | |
593 | level = headernest.index(i) + 1 |
|
593 | level = headernest.index(i) + 1 | |
594 | out.append('<h%d>%s</h%d>\n' % (level, escape(lines[0]), level)) |
|
594 | out.append('<h%d>%s</h%d>\n' % (level, escape(lines[0]), level)) | |
595 | elif btype == 'table': |
|
595 | elif btype == 'table': | |
596 | table = b['table'] |
|
596 | table = b['table'] | |
597 | out.append('<table>\n') |
|
597 | out.append('<table>\n') | |
598 | for row in table: |
|
598 | for row in table: | |
599 | out.append('<tr>') |
|
599 | out.append('<tr>') | |
600 | for v in row: |
|
600 | for v in row: | |
601 | out.append('<td>') |
|
601 | out.append('<td>') | |
602 | out.append(escape(v)) |
|
602 | out.append(escape(v)) | |
603 | out.append('</td>') |
|
603 | out.append('</td>') | |
604 | out.append('\n') |
|
604 | out.append('\n') | |
605 | out.pop() |
|
605 | out.pop() | |
606 | out.append('</tr>\n') |
|
606 | out.append('</tr>\n') | |
607 | out.append('</table>\n') |
|
607 | out.append('</table>\n') | |
608 | elif btype == 'definition': |
|
608 | elif btype == 'definition': | |
609 | openlist('dl', level) |
|
609 | openlist('dl', level) | |
610 | term = escape(lines[0]) |
|
610 | term = escape(lines[0]) | |
611 | text = escape(' '.join(map(str.strip, lines[1:]))) |
|
611 | text = escape(' '.join(map(str.strip, lines[1:]))) | |
612 | out.append(' <dt>%s\n <dd>%s\n' % (term, text)) |
|
612 | out.append(' <dt>%s\n <dd>%s\n' % (term, text)) | |
613 | elif btype == 'bullet': |
|
613 | elif btype == 'bullet': | |
614 | bullet, head = lines[0].split(' ', 1) |
|
614 | bullet, head = lines[0].split(' ', 1) | |
615 | if bullet in ('*', '-'): |
|
615 | if bullet in ('*', '-'): | |
616 | openlist('ul', level) |
|
616 | openlist('ul', level) | |
617 | else: |
|
617 | else: | |
618 | openlist('ol', level) |
|
618 | openlist('ol', level) | |
619 | out.append(' <li> %s\n' % escape(' '.join([head] + lines[1:]))) |
|
619 | out.append(' <li> %s\n' % escape(' '.join([head] + lines[1:]))) | |
620 | elif btype == 'field': |
|
620 | elif btype == 'field': | |
621 | openlist('dl', level) |
|
621 | openlist('dl', level) | |
622 | key = escape(b['key']) |
|
622 | key = escape(b['key']) | |
623 | text = escape(' '.join(map(str.strip, lines))) |
|
623 | text = escape(' '.join(map(str.strip, lines))) | |
624 | out.append(' <dt>%s\n <dd>%s\n' % (key, text)) |
|
624 | out.append(' <dt>%s\n <dd>%s\n' % (key, text)) | |
625 | elif btype == 'option': |
|
625 | elif btype == 'option': | |
626 | openlist('dl', level) |
|
626 | openlist('dl', level) | |
627 | opt = escape(b['optstr']) |
|
627 | opt = escape(b['optstr']) | |
628 | desc = escape(' '.join(map(str.strip, lines))) |
|
628 | desc = escape(' '.join(map(str.strip, lines))) | |
629 | out.append(' <dt>%s\n <dd>%s\n' % (opt, desc)) |
|
629 | out.append(' <dt>%s\n <dd>%s\n' % (opt, desc)) | |
630 |
|
630 | |||
631 | # close lists if indent level of next block is lower |
|
631 | # close lists if indent level of next block is lower | |
632 | if listnest: |
|
632 | if listnest: | |
633 | start, level = listnest[-1] |
|
633 | start, level = listnest[-1] | |
634 | if pos == len(blocks) - 1: |
|
634 | if pos == len(blocks) - 1: | |
635 | out.append('</%s>\n' % start) |
|
635 | out.append('</%s>\n' % start) | |
636 | listnest.pop() |
|
636 | listnest.pop() | |
637 | else: |
|
637 | else: | |
638 | nb = blocks[pos + 1] |
|
638 | nb = blocks[pos + 1] | |
639 | ni = nb['indent'] |
|
639 | ni = nb['indent'] | |
640 | if (ni < level or |
|
640 | if (ni < level or | |
641 | (ni == level and |
|
641 | (ni == level and | |
642 | nb['type'] not in 'definition bullet field option')): |
|
642 | nb['type'] not in 'definition bullet field option')): | |
643 | out.append('</%s>\n' % start) |
|
643 | out.append('</%s>\n' % start) | |
644 | listnest.pop() |
|
644 | listnest.pop() | |
645 |
|
645 | |||
646 | return ''.join(out) |
|
646 | return ''.join(out) | |
647 |
|
647 | |||
648 | def parse(text, indent=0, keep=None, admonitions=None): |
|
648 | def parse(text, indent=0, keep=None, admonitions=None): | |
649 | """Parse text into a list of blocks""" |
|
649 | """Parse text into a list of blocks""" | |
650 | pruned = [] |
|
650 | pruned = [] | |
651 | blocks = findblocks(text) |
|
651 | blocks = findblocks(text) | |
652 | for b in blocks: |
|
652 | for b in blocks: | |
653 | b['indent'] += indent |
|
653 | b['indent'] += indent | |
654 | blocks = findliteralblocks(blocks) |
|
654 | blocks = findliteralblocks(blocks) | |
655 | blocks = findtables(blocks) |
|
655 | blocks = findtables(blocks) | |
656 | blocks, pruned = prunecontainers(blocks, keep or []) |
|
656 | blocks, pruned = prunecontainers(blocks, keep or []) | |
657 | blocks = findsections(blocks) |
|
657 | blocks = findsections(blocks) | |
658 | blocks = inlineliterals(blocks) |
|
658 | blocks = inlineliterals(blocks) | |
659 | blocks = hgrole(blocks) |
|
659 | blocks = hgrole(blocks) | |
660 | blocks = splitparagraphs(blocks) |
|
660 | blocks = splitparagraphs(blocks) | |
661 | blocks = updatefieldlists(blocks) |
|
661 | blocks = updatefieldlists(blocks) | |
662 | blocks = updateoptionlists(blocks) |
|
662 | blocks = updateoptionlists(blocks) | |
663 | blocks = findadmonitions(blocks, admonitions=admonitions) |
|
663 | blocks = findadmonitions(blocks, admonitions=admonitions) | |
664 | blocks = addmargins(blocks) |
|
664 | blocks = addmargins(blocks) | |
665 | blocks = prunecomments(blocks) |
|
665 | blocks = prunecomments(blocks) | |
666 | return blocks, pruned |
|
666 | return blocks, pruned | |
667 |
|
667 | |||
668 | def formatblocks(blocks, width): |
|
668 | def formatblocks(blocks, width): | |
669 | text = ''.join(formatblock(b, width) for b in blocks) |
|
669 | text = ''.join(formatblock(b, width) for b in blocks) | |
670 | return text |
|
670 | return text | |
671 |
|
671 | |||
672 | def format(text, width=80, indent=0, keep=None, style='plain', section=None): |
|
672 | def format(text, width=80, indent=0, keep=None, style='plain', section=None): | |
673 | """Parse and format the text according to width.""" |
|
673 | """Parse and format the text according to width.""" | |
674 | blocks, pruned = parse(text, indent, keep or []) |
|
674 | blocks, pruned = parse(text, indent, keep or []) | |
675 | parents = [] |
|
675 | parents = [] | |
676 | if section: |
|
676 | if section: | |
677 | sections = getsections(blocks) |
|
677 | sections = getsections(blocks) | |
678 | blocks = [] |
|
678 | blocks = [] | |
679 | i = 0 |
|
679 | i = 0 | |
680 | lastparents = [] |
|
680 | lastparents = [] | |
681 | synthetic = [] |
|
681 | synthetic = [] | |
682 | collapse = True |
|
682 | collapse = True | |
683 | while i < len(sections): |
|
683 | while i < len(sections): | |
684 | name, nest, b = sections[i] |
|
684 | name, nest, b = sections[i] | |
685 | del parents[nest:] |
|
685 | del parents[nest:] | |
686 | parents.append(i) |
|
686 | parents.append(i) | |
687 | if name == section: |
|
687 | if name == section: | |
688 | if lastparents != parents: |
|
688 | if lastparents != parents: | |
689 | llen = len(lastparents) |
|
689 | llen = len(lastparents) | |
690 | plen = len(parents) |
|
690 | plen = len(parents) | |
691 | if llen and llen != plen: |
|
691 | if llen and llen != plen: | |
692 | collapse = False |
|
692 | collapse = False | |
693 | s = [] |
|
693 | s = [] | |
694 | for j in xrange(3, plen - 1): |
|
694 | for j in xrange(3, plen - 1): | |
695 | parent = parents[j] |
|
695 | parent = parents[j] | |
696 | if (j >= llen or |
|
696 | if (j >= llen or | |
697 | lastparents[j] != parent): |
|
697 | lastparents[j] != parent): | |
698 | s.append(len(blocks)) |
|
698 | s.append(len(blocks)) | |
699 | sec = sections[parent][2] |
|
699 | sec = sections[parent][2] | |
700 | blocks.append(sec[0]) |
|
700 | blocks.append(sec[0]) | |
701 | blocks.append(sec[-1]) |
|
701 | blocks.append(sec[-1]) | |
702 | if s: |
|
702 | if s: | |
703 | synthetic.append(s) |
|
703 | synthetic.append(s) | |
704 |
|
704 | |||
705 | lastparents = parents[:] |
|
705 | lastparents = parents[:] | |
706 | blocks.extend(b) |
|
706 | blocks.extend(b) | |
707 |
|
707 | |||
708 | ## Also show all subnested sections |
|
708 | ## Also show all subnested sections | |
709 | while i + 1 < len(sections) and sections[i + 1][1] > nest: |
|
709 | while i + 1 < len(sections) and sections[i + 1][1] > nest: | |
710 | i += 1 |
|
710 | i += 1 | |
711 | blocks.extend(sections[i][2]) |
|
711 | blocks.extend(sections[i][2]) | |
712 | i += 1 |
|
712 | i += 1 | |
713 | if collapse: |
|
713 | if collapse: | |
714 | synthetic.reverse() |
|
714 | synthetic.reverse() | |
715 | for s in synthetic: |
|
715 | for s in synthetic: | |
716 | path = [blocks[syn]['lines'][0] for syn in s] |
|
716 | path = [blocks[syn]['lines'][0] for syn in s] | |
717 | real = s[-1] + 2 |
|
717 | real = s[-1] + 2 | |
718 | realline = blocks[real]['lines'] |
|
718 | realline = blocks[real]['lines'] | |
719 | realline[0] = ('"%s"' % |
|
719 | realline[0] = ('"%s"' % | |
720 | '.'.join(path + [realline[0]]).replace('"', '')) |
|
720 | '.'.join(path + [realline[0]]).replace('"', '')) | |
721 | del blocks[s[0]:real] |
|
721 | del blocks[s[0]:real] | |
722 |
|
722 | |||
723 | if style == 'html': |
|
723 | if style == 'html': | |
724 | text = formathtml(blocks) |
|
724 | text = formathtml(blocks) | |
725 | else: |
|
725 | else: | |
726 | text = ''.join(formatblock(b, width) for b in blocks) |
|
726 | text = ''.join(formatblock(b, width) for b in blocks) | |
727 | if keep is None: |
|
727 | if keep is None: | |
728 | return text |
|
728 | return text | |
729 | else: |
|
729 | else: | |
730 | return text, pruned |
|
730 | return text, pruned | |
731 |
|
731 | |||
732 | def getsections(blocks): |
|
732 | def getsections(blocks): | |
733 | '''return a list of (section name, nesting level, blocks) tuples''' |
|
733 | '''return a list of (section name, nesting level, blocks) tuples''' | |
734 | nest = "" |
|
734 | nest = "" | |
735 | level = 0 |
|
735 | level = 0 | |
736 | secs = [] |
|
736 | secs = [] | |
737 |
|
737 | |||
738 | def getname(b): |
|
738 | def getname(b): | |
739 | if b['type'] == 'field': |
|
739 | if b['type'] == 'field': | |
740 | x = b['key'] |
|
740 | x = b['key'] | |
741 | else: |
|
741 | else: | |
742 | x = b['lines'][0] |
|
742 | x = b['lines'][0] | |
743 | x = encoding.lower(x).strip('"') |
|
743 | x = encoding.lower(x).strip('"') | |
744 | if '(' in x: |
|
744 | if '(' in x: | |
745 | x = x.split('(')[0] |
|
745 | x = x.split('(')[0] | |
746 | return x |
|
746 | return x | |
747 |
|
747 | |||
748 | for b in blocks: |
|
748 | for b in blocks: | |
749 | if b['type'] == 'section': |
|
749 | if b['type'] == 'section': | |
750 | i = b['underline'] |
|
750 | i = b['underline'] | |
751 | if i not in nest: |
|
751 | if i not in nest: | |
752 | nest += i |
|
752 | nest += i | |
753 | level = nest.index(i) + 1 |
|
753 | level = nest.index(i) + 1 | |
754 | nest = nest[:level] |
|
754 | nest = nest[:level] | |
755 | secs.append((getname(b), level, [b])) |
|
755 | secs.append((getname(b), level, [b])) | |
756 | elif b['type'] in ('definition', 'field'): |
|
756 | elif b['type'] in ('definition', 'field'): | |
757 | i = ' ' |
|
757 | i = ' ' | |
758 | if i not in nest: |
|
758 | if i not in nest: | |
759 | nest += i |
|
759 | nest += i | |
760 | level = nest.index(i) + 1 |
|
760 | level = nest.index(i) + 1 | |
761 | nest = nest[:level] |
|
761 | nest = nest[:level] | |
762 | for i in range(1, len(secs) + 1): |
|
762 | for i in range(1, len(secs) + 1): | |
763 | sec = secs[-i] |
|
763 | sec = secs[-i] | |
764 | if sec[1] < level: |
|
764 | if sec[1] < level: | |
765 | break |
|
765 | break | |
766 | siblings = [a for a in sec[2] if a['type'] == 'definition'] |
|
766 | siblings = [a for a in sec[2] if a['type'] == 'definition'] | |
767 | if siblings: |
|
767 | if siblings: | |
768 | siblingindent = siblings[-1]['indent'] |
|
768 | siblingindent = siblings[-1]['indent'] | |
769 | indent = b['indent'] |
|
769 | indent = b['indent'] | |
770 | if siblingindent < indent: |
|
770 | if siblingindent < indent: | |
771 | level += 1 |
|
771 | level += 1 | |
772 | break |
|
772 | break | |
773 | elif siblingindent == indent: |
|
773 | elif siblingindent == indent: | |
774 | level = sec[1] |
|
774 | level = sec[1] | |
775 | break |
|
775 | break | |
776 | secs.append((getname(b), level, [b])) |
|
776 | secs.append((getname(b), level, [b])) | |
777 | else: |
|
777 | else: | |
778 | if not secs: |
|
778 | if not secs: | |
779 | # add an initial empty section |
|
779 | # add an initial empty section | |
780 | secs = [('', 0, [])] |
|
780 | secs = [('', 0, [])] | |
781 | if b['type'] != 'margin': |
|
781 | if b['type'] != 'margin': | |
782 | pointer = 1 |
|
782 | pointer = 1 | |
783 | bindent = b['indent'] |
|
783 | bindent = b['indent'] | |
784 | while pointer < len(secs): |
|
784 | while pointer < len(secs): | |
785 | section = secs[-pointer][2][0] |
|
785 | section = secs[-pointer][2][0] | |
786 | if section['type'] != 'margin': |
|
786 | if section['type'] != 'margin': | |
787 | sindent = section['indent'] |
|
787 | sindent = section['indent'] | |
788 | if len(section['lines']) > 1: |
|
788 | if len(section['lines']) > 1: | |
789 | sindent += len(section['lines'][1]) - \ |
|
789 | sindent += len(section['lines'][1]) - \ | |
790 | len(section['lines'][1].lstrip(' ')) |
|
790 | len(section['lines'][1].lstrip(' ')) | |
791 | if bindent >= sindent: |
|
791 | if bindent >= sindent: | |
792 | break |
|
792 | break | |
793 | pointer += 1 |
|
793 | pointer += 1 | |
794 | if pointer > 1: |
|
794 | if pointer > 1: | |
795 | blevel = secs[-pointer][1] |
|
795 | blevel = secs[-pointer][1] | |
796 | if section['type'] != b['type']: |
|
796 | if section['type'] != b['type']: | |
797 | blevel += 1 |
|
797 | blevel += 1 | |
798 | secs.append(('', blevel, [])) |
|
798 | secs.append(('', blevel, [])) | |
799 | secs[-1][2].append(b) |
|
799 | secs[-1][2].append(b) | |
800 | return secs |
|
800 | return secs | |
801 |
|
801 | |||
802 | def decorateblocks(blocks, width): |
|
802 | def decorateblocks(blocks, width): | |
803 | '''generate a list of (section name, line text) pairs for search''' |
|
803 | '''generate a list of (section name, line text) pairs for search''' | |
804 | lines = [] |
|
804 | lines = [] | |
805 | for s in getsections(blocks): |
|
805 | for s in getsections(blocks): | |
806 | section = s[0] |
|
806 | section = s[0] | |
807 | text = formatblocks(s[2], width) |
|
807 | text = formatblocks(s[2], width) | |
808 | lines.append([(section, l) for l in text.splitlines(True)]) |
|
808 | lines.append([(section, l) for l in text.splitlines(True)]) | |
809 | return lines |
|
809 | return lines | |
810 |
|
810 | |||
811 | def maketable(data, indent=0, header=False): |
|
811 | def maketable(data, indent=0, header=False): | |
812 | '''Generate an RST table for the given table data as a list of lines''' |
|
812 | '''Generate an RST table for the given table data as a list of lines''' | |
813 |
|
813 | |||
814 | widths = [max(encoding.colwidth(e) for e in c) for c in zip(*data)] |
|
814 | widths = [max(encoding.colwidth(e) for e in c) for c in zip(*data)] | |
815 | indent = ' ' * indent |
|
815 | indent = ' ' * indent | |
816 | div = indent + ' '.join('=' * w for w in widths) + '\n' |
|
816 | div = indent + ' '.join('=' * w for w in widths) + '\n' | |
817 |
|
817 | |||
818 | out = [div] |
|
818 | out = [div] | |
819 | for row in data: |
|
819 | for row in data: | |
820 | l = [] |
|
820 | l = [] | |
821 | for w, v in zip(widths, row): |
|
821 | for w, v in zip(widths, row): | |
822 | if '\n' in v: |
|
822 | if '\n' in v: | |
823 | # only remove line breaks and indentation, long lines are |
|
823 | # only remove line breaks and indentation, long lines are | |
824 | # handled by the next tool |
|
824 | # handled by the next tool | |
825 | v = ' '.join(e.lstrip() for e in v.split('\n')) |
|
825 | v = ' '.join(e.lstrip() for e in v.split('\n')) | |
826 | pad = ' ' * (w - encoding.colwidth(v)) |
|
826 | pad = ' ' * (w - encoding.colwidth(v)) | |
827 | l.append(v + pad) |
|
827 | l.append(v + pad) | |
828 | out.append(indent + ' '.join(l) + "\n") |
|
828 | out.append(indent + ' '.join(l) + "\n") | |
829 | if header and len(data) > 1: |
|
829 | if header and len(data) > 1: | |
830 | out.insert(2, div) |
|
830 | out.insert(2, div) | |
831 | out.append(div) |
|
831 | out.append(div) | |
832 | return out |
|
832 | return out |
General Comments 0
You need to be logged in to leave comments.
Login now