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