Show More
@@ -1,571 +1,571 b'' | |||||
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 http://mercurial.selenic.com/wiki/HelpStyleGuide |
|
17 | Remember to update http://mercurial.selenic.com/wiki/HelpStyleGuide | |
18 | when adding support for new constructs. |
|
18 | when adding support for new constructs. | |
19 | """ |
|
19 | """ | |
20 |
|
20 | |||
21 | import re |
|
21 | import re | |
22 | import util, encoding |
|
22 | import util, encoding | |
23 | from i18n import _ |
|
23 | from i18n import _ | |
24 |
|
24 | |||
25 |
|
25 | |||
26 | def replace(text, substs): |
|
26 | def replace(text, substs): | |
27 | utext = text.decode(encoding.encoding) |
|
27 | utext = text.decode(encoding.encoding) | |
28 | for f, t in substs: |
|
28 | for f, t in substs: | |
29 | utext = utext.replace(f, t) |
|
29 | utext = utext.replace(f, t) | |
30 | return utext.encode(encoding.encoding) |
|
30 | return utext.encode(encoding.encoding) | |
31 |
|
31 | |||
32 |
|
32 | |||
33 | _blockre = re.compile(r"\n(?:\s*\n)+") |
|
33 | _blockre = re.compile(r"\n(?:\s*\n)+") | |
34 |
|
34 | |||
35 | def findblocks(text): |
|
35 | def findblocks(text): | |
36 | """Find continuous blocks of lines in text. |
|
36 | """Find continuous blocks of lines in text. | |
37 |
|
37 | |||
38 | Returns a list of dictionaries representing the blocks. Each block |
|
38 | Returns a list of dictionaries representing the blocks. Each block | |
39 | has an 'indent' field and a 'lines' field. |
|
39 | has an 'indent' field and a 'lines' field. | |
40 | """ |
|
40 | """ | |
41 | blocks = [] |
|
41 | blocks = [] | |
42 | for b in _blockre.split(text.lstrip('\n').rstrip()): |
|
42 | for b in _blockre.split(text.lstrip('\n').rstrip()): | |
43 | lines = b.splitlines() |
|
43 | lines = b.splitlines() | |
44 | indent = min((len(l) - len(l.lstrip())) for l in lines) |
|
44 | indent = min((len(l) - len(l.lstrip())) for l in lines) | |
45 | lines = [l[indent:] for l in lines] |
|
45 | lines = [l[indent:] for l in lines] | |
46 | blocks.append(dict(indent=indent, lines=lines)) |
|
46 | blocks.append(dict(indent=indent, lines=lines)) | |
47 | return blocks |
|
47 | return blocks | |
48 |
|
48 | |||
49 |
|
49 | |||
50 | def findliteralblocks(blocks): |
|
50 | def findliteralblocks(blocks): | |
51 | """Finds literal blocks and adds a 'type' field to the blocks. |
|
51 | """Finds literal blocks and adds a 'type' field to the blocks. | |
52 |
|
52 | |||
53 | Literal blocks are given the type 'literal', all other blocks are |
|
53 | Literal blocks are given the type 'literal', all other blocks are | |
54 | given type the 'paragraph'. |
|
54 | given type the 'paragraph'. | |
55 | """ |
|
55 | """ | |
56 | i = 0 |
|
56 | i = 0 | |
57 | while i < len(blocks): |
|
57 | while i < len(blocks): | |
58 | # Searching for a block that looks like this: |
|
58 | # Searching for a block that looks like this: | |
59 | # |
|
59 | # | |
60 | # +------------------------------+ |
|
60 | # +------------------------------+ | |
61 | # | paragraph | |
|
61 | # | paragraph | | |
62 | # | (ends with "::") | |
|
62 | # | (ends with "::") | | |
63 | # +------------------------------+ |
|
63 | # +------------------------------+ | |
64 | # +---------------------------+ |
|
64 | # +---------------------------+ | |
65 | # | indented literal block | |
|
65 | # | indented literal block | | |
66 | # +---------------------------+ |
|
66 | # +---------------------------+ | |
67 | blocks[i]['type'] = 'paragraph' |
|
67 | blocks[i]['type'] = 'paragraph' | |
68 | if blocks[i]['lines'][-1].endswith('::') and i + 1 < len(blocks): |
|
68 | if blocks[i]['lines'][-1].endswith('::') and i + 1 < len(blocks): | |
69 | indent = blocks[i]['indent'] |
|
69 | indent = blocks[i]['indent'] | |
70 | adjustment = blocks[i + 1]['indent'] - indent |
|
70 | adjustment = blocks[i + 1]['indent'] - indent | |
71 |
|
71 | |||
72 | if blocks[i]['lines'] == ['::']: |
|
72 | if blocks[i]['lines'] == ['::']: | |
73 | # Expanded form: remove block |
|
73 | # Expanded form: remove block | |
74 | del blocks[i] |
|
74 | del blocks[i] | |
75 | i -= 1 |
|
75 | i -= 1 | |
76 | elif blocks[i]['lines'][-1].endswith(' ::'): |
|
76 | elif blocks[i]['lines'][-1].endswith(' ::'): | |
77 | # Partially minimized form: remove space and both |
|
77 | # Partially minimized form: remove space and both | |
78 | # colons. |
|
78 | # colons. | |
79 | blocks[i]['lines'][-1] = blocks[i]['lines'][-1][:-3] |
|
79 | blocks[i]['lines'][-1] = blocks[i]['lines'][-1][:-3] | |
80 | else: |
|
80 | else: | |
81 | # Fully minimized form: remove just one colon. |
|
81 | # Fully minimized form: remove just one colon. | |
82 | blocks[i]['lines'][-1] = blocks[i]['lines'][-1][:-1] |
|
82 | blocks[i]['lines'][-1] = blocks[i]['lines'][-1][:-1] | |
83 |
|
83 | |||
84 | # List items are formatted with a hanging indent. We must |
|
84 | # List items are formatted with a hanging indent. We must | |
85 | # correct for this here while we still have the original |
|
85 | # correct for this here while we still have the original | |
86 | # information on the indentation of the subsequent literal |
|
86 | # information on the indentation of the subsequent literal | |
87 | # blocks available. |
|
87 | # blocks available. | |
88 | m = _bulletre.match(blocks[i]['lines'][0]) |
|
88 | m = _bulletre.match(blocks[i]['lines'][0]) | |
89 | if m: |
|
89 | if m: | |
90 | indent += m.end() |
|
90 | indent += m.end() | |
91 | adjustment -= m.end() |
|
91 | adjustment -= m.end() | |
92 |
|
92 | |||
93 | # Mark the following indented blocks. |
|
93 | # Mark the following indented blocks. | |
94 | while i + 1 < len(blocks) and blocks[i + 1]['indent'] > indent: |
|
94 | while i + 1 < len(blocks) and blocks[i + 1]['indent'] > indent: | |
95 | blocks[i + 1]['type'] = 'literal' |
|
95 | blocks[i + 1]['type'] = 'literal' | |
96 | blocks[i + 1]['indent'] -= adjustment |
|
96 | blocks[i + 1]['indent'] -= adjustment | |
97 | i += 1 |
|
97 | i += 1 | |
98 | i += 1 |
|
98 | i += 1 | |
99 | return blocks |
|
99 | return blocks | |
100 |
|
100 | |||
101 | _bulletre = re.compile(r'(-|[0-9A-Za-z]+\.|\(?[0-9A-Za-z]+\)|\|) ') |
|
101 | _bulletre = re.compile(r'(-|[0-9A-Za-z]+\.|\(?[0-9A-Za-z]+\)|\|) ') | |
102 | _optionre = re.compile(r'^(-([a-zA-Z0-9]), )?(--[a-z0-9-]+)' |
|
102 | _optionre = re.compile(r'^(-([a-zA-Z0-9]), )?(--[a-z0-9-]+)' | |
103 | r'((.*) +)(.*)$') |
|
103 | r'((.*) +)(.*)$') | |
104 | _fieldre = re.compile(r':(?![: ])([^:]*)(?<! ):[ ]+(.*)') |
|
104 | _fieldre = re.compile(r':(?![: ])([^:]*)(?<! ):[ ]+(.*)') | |
105 | _definitionre = re.compile(r'[^ ]') |
|
105 | _definitionre = re.compile(r'[^ ]') | |
106 | _tablere = re.compile(r'(=+\s+)*=+') |
|
106 | _tablere = re.compile(r'(=+\s+)*=+') | |
107 |
|
107 | |||
108 | def splitparagraphs(blocks): |
|
108 | def splitparagraphs(blocks): | |
109 | """Split paragraphs into lists.""" |
|
109 | """Split paragraphs into lists.""" | |
110 | # Tuples with (list type, item regexp, single line items?). Order |
|
110 | # Tuples with (list type, item regexp, single line items?). Order | |
111 | # matters: definition lists has the least specific regexp and must |
|
111 | # matters: definition lists has the least specific regexp and must | |
112 | # come last. |
|
112 | # come last. | |
113 | listtypes = [('bullet', _bulletre, True), |
|
113 | listtypes = [('bullet', _bulletre, True), | |
114 | ('option', _optionre, True), |
|
114 | ('option', _optionre, True), | |
115 | ('field', _fieldre, True), |
|
115 | ('field', _fieldre, True), | |
116 | ('definition', _definitionre, False)] |
|
116 | ('definition', _definitionre, False)] | |
117 |
|
117 | |||
118 | def match(lines, i, itemre, singleline): |
|
118 | def match(lines, i, itemre, singleline): | |
119 | """Does itemre match an item at line i? |
|
119 | """Does itemre match an item at line i? | |
120 |
|
120 | |||
121 | A list item can be followed by an idented line or another list |
|
121 | A list item can be followed by an idented line or another list | |
122 | item (but only if singleline is True). |
|
122 | item (but only if singleline is True). | |
123 | """ |
|
123 | """ | |
124 | line1 = lines[i] |
|
124 | line1 = lines[i] | |
125 | line2 = i + 1 < len(lines) and lines[i + 1] or '' |
|
125 | line2 = i + 1 < len(lines) and lines[i + 1] or '' | |
126 | if not itemre.match(line1): |
|
126 | if not itemre.match(line1): | |
127 | return False |
|
127 | return False | |
128 | if singleline: |
|
128 | if singleline: | |
129 | return line2 == '' or line2[0] == ' ' or itemre.match(line2) |
|
129 | return line2 == '' or line2[0] == ' ' or itemre.match(line2) | |
130 | else: |
|
130 | else: | |
131 | return line2.startswith(' ') |
|
131 | return line2.startswith(' ') | |
132 |
|
132 | |||
133 | i = 0 |
|
133 | i = 0 | |
134 | while i < len(blocks): |
|
134 | while i < len(blocks): | |
135 | if blocks[i]['type'] == 'paragraph': |
|
135 | if blocks[i]['type'] == 'paragraph': | |
136 | lines = blocks[i]['lines'] |
|
136 | lines = blocks[i]['lines'] | |
137 | for type, itemre, singleline in listtypes: |
|
137 | for type, itemre, singleline in listtypes: | |
138 | if match(lines, 0, itemre, singleline): |
|
138 | if match(lines, 0, itemre, singleline): | |
139 | items = [] |
|
139 | items = [] | |
140 | for j, line in enumerate(lines): |
|
140 | for j, line in enumerate(lines): | |
141 | if match(lines, j, itemre, singleline): |
|
141 | if match(lines, j, itemre, singleline): | |
142 | items.append(dict(type=type, lines=[], |
|
142 | items.append(dict(type=type, lines=[], | |
143 | indent=blocks[i]['indent'])) |
|
143 | indent=blocks[i]['indent'])) | |
144 | items[-1]['lines'].append(line) |
|
144 | items[-1]['lines'].append(line) | |
145 | blocks[i:i + 1] = items |
|
145 | blocks[i:i + 1] = items | |
146 | break |
|
146 | break | |
147 | i += 1 |
|
147 | i += 1 | |
148 | return blocks |
|
148 | return blocks | |
149 |
|
149 | |||
150 |
|
150 | |||
151 | _fieldwidth = 12 |
|
151 | _fieldwidth = 12 | |
152 |
|
152 | |||
153 | def updatefieldlists(blocks): |
|
153 | def updatefieldlists(blocks): | |
154 | """Find key and maximum key width for field lists.""" |
|
154 | """Find key and maximum key width for field lists.""" | |
155 | i = 0 |
|
155 | i = 0 | |
156 | while i < len(blocks): |
|
156 | while i < len(blocks): | |
157 | if blocks[i]['type'] != 'field': |
|
157 | if blocks[i]['type'] != 'field': | |
158 | i += 1 |
|
158 | i += 1 | |
159 | continue |
|
159 | continue | |
160 |
|
160 | |||
161 | keywidth = 0 |
|
161 | keywidth = 0 | |
162 | j = i |
|
162 | j = i | |
163 | while j < len(blocks) and blocks[j]['type'] == 'field': |
|
163 | while j < len(blocks) and blocks[j]['type'] == 'field': | |
164 | m = _fieldre.match(blocks[j]['lines'][0]) |
|
164 | m = _fieldre.match(blocks[j]['lines'][0]) | |
165 | key, rest = m.groups() |
|
165 | key, rest = m.groups() | |
166 | blocks[j]['lines'][0] = rest |
|
166 | blocks[j]['lines'][0] = rest | |
167 | blocks[j]['key'] = key |
|
167 | blocks[j]['key'] = key | |
168 | keywidth = max(keywidth, len(key)) |
|
168 | keywidth = max(keywidth, len(key)) | |
169 | j += 1 |
|
169 | j += 1 | |
170 |
|
170 | |||
171 | for block in blocks[i:j]: |
|
171 | for block in blocks[i:j]: | |
172 | block['keywidth'] = keywidth |
|
172 | block['keywidth'] = keywidth | |
173 | i = j + 1 |
|
173 | i = j + 1 | |
174 |
|
174 | |||
175 | return blocks |
|
175 | return blocks | |
176 |
|
176 | |||
177 |
|
177 | |||
178 | def updateoptionlists(blocks): |
|
178 | def updateoptionlists(blocks): | |
179 | i = 0 |
|
179 | i = 0 | |
180 | while i < len(blocks): |
|
180 | while i < len(blocks): | |
181 | if blocks[i]['type'] != 'option': |
|
181 | if blocks[i]['type'] != 'option': | |
182 | i += 1 |
|
182 | i += 1 | |
183 | continue |
|
183 | continue | |
184 |
|
184 | |||
185 | optstrwidth = 0 |
|
185 | optstrwidth = 0 | |
186 | j = i |
|
186 | j = i | |
187 | while j < len(blocks) and blocks[j]['type'] == 'option': |
|
187 | while j < len(blocks) and blocks[j]['type'] == 'option': | |
188 | m = _optionre.match(blocks[j]['lines'][0]) |
|
188 | m = _optionre.match(blocks[j]['lines'][0]) | |
189 |
|
189 | |||
190 | shortoption = m.group(2) |
|
190 | shortoption = m.group(2) | |
191 | group3 = m.group(3) |
|
191 | group3 = m.group(3) | |
192 | longoption = group3[2:].strip() |
|
192 | longoption = group3[2:].strip() | |
193 | desc = m.group(6).strip() |
|
193 | desc = m.group(6).strip() | |
194 | longoptionarg = m.group(5).strip() |
|
194 | longoptionarg = m.group(5).strip() | |
195 | blocks[j]['lines'][0] = desc |
|
195 | blocks[j]['lines'][0] = desc | |
196 |
|
196 | |||
197 | noshortop = '' |
|
197 | noshortop = '' | |
198 | if not shortoption: |
|
198 | if not shortoption: | |
199 | noshortop = ' ' |
|
199 | noshortop = ' ' | |
200 |
|
200 | |||
201 | opt = "%s%s" % (shortoption and "-%s " % shortoption or '', |
|
201 | opt = "%s%s" % (shortoption and "-%s " % shortoption or '', | |
202 | ("%s--%s %s") % (noshortop, longoption, |
|
202 | ("%s--%s %s") % (noshortop, longoption, | |
203 | longoptionarg)) |
|
203 | longoptionarg)) | |
204 | opt = opt.rstrip() |
|
204 | opt = opt.rstrip() | |
205 | blocks[j]['optstr'] = opt |
|
205 | blocks[j]['optstr'] = opt | |
206 | optstrwidth = max(optstrwidth, encoding.colwidth(opt)) |
|
206 | optstrwidth = max(optstrwidth, encoding.colwidth(opt)) | |
207 | j += 1 |
|
207 | j += 1 | |
208 |
|
208 | |||
209 | for block in blocks[i:j]: |
|
209 | for block in blocks[i:j]: | |
210 | block['optstrwidth'] = optstrwidth |
|
210 | block['optstrwidth'] = optstrwidth | |
211 | i = j + 1 |
|
211 | i = j + 1 | |
212 | return blocks |
|
212 | return blocks | |
213 |
|
213 | |||
214 | def prunecontainers(blocks, keep): |
|
214 | def prunecontainers(blocks, keep): | |
215 | """Prune unwanted containers. |
|
215 | """Prune unwanted containers. | |
216 |
|
216 | |||
217 | The blocks must have a 'type' field, i.e., they should have been |
|
217 | The blocks must have a 'type' field, i.e., they should have been | |
218 | run through findliteralblocks first. |
|
218 | run through findliteralblocks first. | |
219 | """ |
|
219 | """ | |
220 | pruned = [] |
|
220 | pruned = [] | |
221 | i = 0 |
|
221 | i = 0 | |
222 | while i + 1 < len(blocks): |
|
222 | while i + 1 < len(blocks): | |
223 | # Searching for a block that looks like this: |
|
223 | # Searching for a block that looks like this: | |
224 | # |
|
224 | # | |
225 | # +-------+---------------------------+ |
|
225 | # +-------+---------------------------+ | |
226 | # | ".. container ::" type | |
|
226 | # | ".. container ::" type | | |
227 | # +---+ | |
|
227 | # +---+ | | |
228 | # | blocks | |
|
228 | # | blocks | | |
229 | # +-------------------------------+ |
|
229 | # +-------------------------------+ | |
230 | if (blocks[i]['type'] == 'paragraph' and |
|
230 | if (blocks[i]['type'] == 'paragraph' and | |
231 | blocks[i]['lines'][0].startswith('.. container::')): |
|
231 | blocks[i]['lines'][0].startswith('.. container::')): | |
232 | indent = blocks[i]['indent'] |
|
232 | indent = blocks[i]['indent'] | |
233 | adjustment = blocks[i + 1]['indent'] - indent |
|
233 | adjustment = blocks[i + 1]['indent'] - indent | |
234 | containertype = blocks[i]['lines'][0][15:] |
|
234 | containertype = blocks[i]['lines'][0][15:] | |
235 | prune = containertype not in keep |
|
235 | prune = containertype not in keep | |
236 | if prune: |
|
236 | if prune: | |
237 | pruned.append(containertype) |
|
237 | pruned.append(containertype) | |
238 |
|
238 | |||
239 | # Always delete "..container:: type" block |
|
239 | # Always delete "..container:: type" block | |
240 | del blocks[i] |
|
240 | del blocks[i] | |
241 | j = i |
|
241 | j = i | |
|
242 | i -= 1 | |||
242 | while j < len(blocks) and blocks[j]['indent'] > indent: |
|
243 | while j < len(blocks) and blocks[j]['indent'] > indent: | |
243 | if prune: |
|
244 | if prune: | |
244 | del blocks[j] |
|
245 | del blocks[j] | |
245 | i -= 1 # adjust outer index |
|
|||
246 | else: |
|
246 | else: | |
247 | blocks[j]['indent'] -= adjustment |
|
247 | blocks[j]['indent'] -= adjustment | |
248 | j += 1 |
|
248 | j += 1 | |
249 | i += 1 |
|
249 | i += 1 | |
250 | return blocks, pruned |
|
250 | return blocks, pruned | |
251 |
|
251 | |||
252 |
|
252 | |||
253 | _sectionre = re.compile(r"""^([-=`:.'"~^_*+#])\1+$""") |
|
253 | _sectionre = re.compile(r"""^([-=`:.'"~^_*+#])\1+$""") | |
254 |
|
254 | |||
255 | def findtables(blocks): |
|
255 | def findtables(blocks): | |
256 | '''Find simple tables |
|
256 | '''Find simple tables | |
257 |
|
257 | |||
258 | Only simple one-line table elements are supported |
|
258 | Only simple one-line table elements are supported | |
259 | ''' |
|
259 | ''' | |
260 |
|
260 | |||
261 | for block in blocks: |
|
261 | for block in blocks: | |
262 | # Searching for a block that looks like this: |
|
262 | # Searching for a block that looks like this: | |
263 | # |
|
263 | # | |
264 | # === ==== === |
|
264 | # === ==== === | |
265 | # A B C |
|
265 | # A B C | |
266 | # === ==== === <- optional |
|
266 | # === ==== === <- optional | |
267 | # 1 2 3 |
|
267 | # 1 2 3 | |
268 | # x y z |
|
268 | # x y z | |
269 | # === ==== === |
|
269 | # === ==== === | |
270 | if (block['type'] == 'paragraph' and |
|
270 | if (block['type'] == 'paragraph' and | |
271 | len(block['lines']) > 4 and |
|
271 | len(block['lines']) > 4 and | |
272 | _tablere.match(block['lines'][0]) and |
|
272 | _tablere.match(block['lines'][0]) and | |
273 | block['lines'][0] == block['lines'][-1]): |
|
273 | block['lines'][0] == block['lines'][-1]): | |
274 | block['type'] = 'table' |
|
274 | block['type'] = 'table' | |
275 | block['header'] = False |
|
275 | block['header'] = False | |
276 | div = block['lines'][0] |
|
276 | div = block['lines'][0] | |
277 | columns = [x for x in xrange(len(div)) |
|
277 | columns = [x for x in xrange(len(div)) | |
278 | if div[x] == '=' and (x == 0 or div[x - 1] == ' ')] |
|
278 | if div[x] == '=' and (x == 0 or div[x - 1] == ' ')] | |
279 | rows = [] |
|
279 | rows = [] | |
280 | for l in block['lines'][1:-1]: |
|
280 | for l in block['lines'][1:-1]: | |
281 | if l == div: |
|
281 | if l == div: | |
282 | block['header'] = True |
|
282 | block['header'] = True | |
283 | continue |
|
283 | continue | |
284 | row = [] |
|
284 | row = [] | |
285 | for n, start in enumerate(columns): |
|
285 | for n, start in enumerate(columns): | |
286 | if n + 1 < len(columns): |
|
286 | if n + 1 < len(columns): | |
287 | row.append(l[start:columns[n + 1]].strip()) |
|
287 | row.append(l[start:columns[n + 1]].strip()) | |
288 | else: |
|
288 | else: | |
289 | row.append(l[start:].strip()) |
|
289 | row.append(l[start:].strip()) | |
290 | rows.append(row) |
|
290 | rows.append(row) | |
291 | block['table'] = rows |
|
291 | block['table'] = rows | |
292 |
|
292 | |||
293 | return blocks |
|
293 | return blocks | |
294 |
|
294 | |||
295 | def findsections(blocks): |
|
295 | def findsections(blocks): | |
296 | """Finds sections. |
|
296 | """Finds sections. | |
297 |
|
297 | |||
298 | The blocks must have a 'type' field, i.e., they should have been |
|
298 | The blocks must have a 'type' field, i.e., they should have been | |
299 | run through findliteralblocks first. |
|
299 | run through findliteralblocks first. | |
300 | """ |
|
300 | """ | |
301 | for block in blocks: |
|
301 | for block in blocks: | |
302 | # Searching for a block that looks like this: |
|
302 | # Searching for a block that looks like this: | |
303 | # |
|
303 | # | |
304 | # +------------------------------+ |
|
304 | # +------------------------------+ | |
305 | # | Section title | |
|
305 | # | Section title | | |
306 | # | ------------- | |
|
306 | # | ------------- | | |
307 | # +------------------------------+ |
|
307 | # +------------------------------+ | |
308 | if (block['type'] == 'paragraph' and |
|
308 | if (block['type'] == 'paragraph' and | |
309 | len(block['lines']) == 2 and |
|
309 | len(block['lines']) == 2 and | |
310 | encoding.colwidth(block['lines'][0]) == len(block['lines'][1]) and |
|
310 | encoding.colwidth(block['lines'][0]) == len(block['lines'][1]) and | |
311 | _sectionre.match(block['lines'][1])): |
|
311 | _sectionre.match(block['lines'][1])): | |
312 | block['underline'] = block['lines'][1][0] |
|
312 | block['underline'] = block['lines'][1][0] | |
313 | block['type'] = 'section' |
|
313 | block['type'] = 'section' | |
314 | del block['lines'][1] |
|
314 | del block['lines'][1] | |
315 | return blocks |
|
315 | return blocks | |
316 |
|
316 | |||
317 |
|
317 | |||
318 | def inlineliterals(blocks): |
|
318 | def inlineliterals(blocks): | |
319 | substs = [('``', '"')] |
|
319 | substs = [('``', '"')] | |
320 | for b in blocks: |
|
320 | for b in blocks: | |
321 | if b['type'] in ('paragraph', 'section'): |
|
321 | if b['type'] in ('paragraph', 'section'): | |
322 | b['lines'] = [replace(l, substs) for l in b['lines']] |
|
322 | b['lines'] = [replace(l, substs) for l in b['lines']] | |
323 | return blocks |
|
323 | return blocks | |
324 |
|
324 | |||
325 |
|
325 | |||
326 | def hgrole(blocks): |
|
326 | def hgrole(blocks): | |
327 | substs = [(':hg:`', '"hg '), ('`', '"')] |
|
327 | substs = [(':hg:`', '"hg '), ('`', '"')] | |
328 | for b in blocks: |
|
328 | for b in blocks: | |
329 | if b['type'] in ('paragraph', 'section'): |
|
329 | if b['type'] in ('paragraph', 'section'): | |
330 | # Turn :hg:`command` into "hg command". This also works |
|
330 | # Turn :hg:`command` into "hg command". This also works | |
331 | # when there is a line break in the command and relies on |
|
331 | # when there is a line break in the command and relies on | |
332 | # the fact that we have no stray back-quotes in the input |
|
332 | # the fact that we have no stray back-quotes in the input | |
333 | # (run the blocks through inlineliterals first). |
|
333 | # (run the blocks through inlineliterals first). | |
334 | b['lines'] = [replace(l, substs) for l in b['lines']] |
|
334 | b['lines'] = [replace(l, substs) for l in b['lines']] | |
335 | return blocks |
|
335 | return blocks | |
336 |
|
336 | |||
337 |
|
337 | |||
338 | def addmargins(blocks): |
|
338 | def addmargins(blocks): | |
339 | """Adds empty blocks for vertical spacing. |
|
339 | """Adds empty blocks for vertical spacing. | |
340 |
|
340 | |||
341 | This groups bullets, options, and definitions together with no vertical |
|
341 | This groups bullets, options, and definitions together with no vertical | |
342 | space between them, and adds an empty block between all other blocks. |
|
342 | space between them, and adds an empty block between all other blocks. | |
343 | """ |
|
343 | """ | |
344 | i = 1 |
|
344 | i = 1 | |
345 | while i < len(blocks): |
|
345 | while i < len(blocks): | |
346 | if (blocks[i]['type'] == blocks[i - 1]['type'] and |
|
346 | if (blocks[i]['type'] == blocks[i - 1]['type'] and | |
347 | blocks[i]['type'] in ('bullet', 'option', 'field')): |
|
347 | blocks[i]['type'] in ('bullet', 'option', 'field')): | |
348 | i += 1 |
|
348 | i += 1 | |
349 | else: |
|
349 | else: | |
350 | blocks.insert(i, dict(lines=[''], indent=0, type='margin')) |
|
350 | blocks.insert(i, dict(lines=[''], indent=0, type='margin')) | |
351 | i += 2 |
|
351 | i += 2 | |
352 | return blocks |
|
352 | return blocks | |
353 |
|
353 | |||
354 | def prunecomments(blocks): |
|
354 | def prunecomments(blocks): | |
355 | """Remove comments.""" |
|
355 | """Remove comments.""" | |
356 | i = 0 |
|
356 | i = 0 | |
357 | while i < len(blocks): |
|
357 | while i < len(blocks): | |
358 | b = blocks[i] |
|
358 | b = blocks[i] | |
359 | if b['type'] == 'paragraph' and (b['lines'][0].startswith('.. ') or |
|
359 | if b['type'] == 'paragraph' and (b['lines'][0].startswith('.. ') or | |
360 | b['lines'] == ['..']): |
|
360 | b['lines'] == ['..']): | |
361 | del blocks[i] |
|
361 | del blocks[i] | |
362 | if i < len(blocks) and blocks[i]['type'] == 'margin': |
|
362 | if i < len(blocks) and blocks[i]['type'] == 'margin': | |
363 | del blocks[i] |
|
363 | del blocks[i] | |
364 | else: |
|
364 | else: | |
365 | i += 1 |
|
365 | i += 1 | |
366 | return blocks |
|
366 | return blocks | |
367 |
|
367 | |||
368 | _admonitionre = re.compile(r"\.\. (admonition|attention|caution|danger|" |
|
368 | _admonitionre = re.compile(r"\.\. (admonition|attention|caution|danger|" | |
369 | r"error|hint|important|note|tip|warning)::", |
|
369 | r"error|hint|important|note|tip|warning)::", | |
370 | flags=re.IGNORECASE) |
|
370 | flags=re.IGNORECASE) | |
371 |
|
371 | |||
372 | def findadmonitions(blocks): |
|
372 | def findadmonitions(blocks): | |
373 | """ |
|
373 | """ | |
374 | Makes the type of the block an admonition block if |
|
374 | Makes the type of the block an admonition block if | |
375 | the first line is an admonition directive |
|
375 | the first line is an admonition directive | |
376 | """ |
|
376 | """ | |
377 | i = 0 |
|
377 | i = 0 | |
378 | while i < len(blocks): |
|
378 | while i < len(blocks): | |
379 | m = _admonitionre.match(blocks[i]['lines'][0]) |
|
379 | m = _admonitionre.match(blocks[i]['lines'][0]) | |
380 | if m: |
|
380 | if m: | |
381 | blocks[i]['type'] = 'admonition' |
|
381 | blocks[i]['type'] = 'admonition' | |
382 | admonitiontitle = blocks[i]['lines'][0][3:m.end() - 2].lower() |
|
382 | admonitiontitle = blocks[i]['lines'][0][3:m.end() - 2].lower() | |
383 |
|
383 | |||
384 | firstline = blocks[i]['lines'][0][m.end() + 1:] |
|
384 | firstline = blocks[i]['lines'][0][m.end() + 1:] | |
385 | if firstline: |
|
385 | if firstline: | |
386 | blocks[i]['lines'].insert(1, ' ' + firstline) |
|
386 | blocks[i]['lines'].insert(1, ' ' + firstline) | |
387 |
|
387 | |||
388 | blocks[i]['admonitiontitle'] = admonitiontitle |
|
388 | blocks[i]['admonitiontitle'] = admonitiontitle | |
389 | del blocks[i]['lines'][0] |
|
389 | del blocks[i]['lines'][0] | |
390 | i = i + 1 |
|
390 | i = i + 1 | |
391 | return blocks |
|
391 | return blocks | |
392 |
|
392 | |||
393 | _admonitiontitles = {'attention': _('Attention:'), |
|
393 | _admonitiontitles = {'attention': _('Attention:'), | |
394 | 'caution': _('Caution:'), |
|
394 | 'caution': _('Caution:'), | |
395 | 'danger': _('!Danger!') , |
|
395 | 'danger': _('!Danger!') , | |
396 | 'error': _('Error:'), |
|
396 | 'error': _('Error:'), | |
397 | 'hint': _('Hint:'), |
|
397 | 'hint': _('Hint:'), | |
398 | 'important': _('Important:'), |
|
398 | 'important': _('Important:'), | |
399 | 'note': _('Note:'), |
|
399 | 'note': _('Note:'), | |
400 | 'tip': _('Tip:'), |
|
400 | 'tip': _('Tip:'), | |
401 | 'warning': _('Warning!')} |
|
401 | 'warning': _('Warning!')} | |
402 |
|
402 | |||
403 | def formatoption(block, width): |
|
403 | def formatoption(block, width): | |
404 | desc = ' '.join(map(str.strip, block['lines'])) |
|
404 | desc = ' '.join(map(str.strip, block['lines'])) | |
405 | colwidth = encoding.colwidth(block['optstr']) |
|
405 | colwidth = encoding.colwidth(block['optstr']) | |
406 | usablewidth = width - 1 |
|
406 | usablewidth = width - 1 | |
407 | hanging = block['optstrwidth'] |
|
407 | hanging = block['optstrwidth'] | |
408 | initindent = '%s%s ' % (block['optstr'], ' ' * ((hanging - colwidth))) |
|
408 | initindent = '%s%s ' % (block['optstr'], ' ' * ((hanging - colwidth))) | |
409 | hangindent = ' ' * (encoding.colwidth(initindent) + 1) |
|
409 | hangindent = ' ' * (encoding.colwidth(initindent) + 1) | |
410 | return ' %s' % (util.wrap(desc, usablewidth, |
|
410 | return ' %s' % (util.wrap(desc, usablewidth, | |
411 | initindent=initindent, |
|
411 | initindent=initindent, | |
412 | hangindent=hangindent)) |
|
412 | hangindent=hangindent)) | |
413 |
|
413 | |||
414 | def formatblock(block, width): |
|
414 | def formatblock(block, width): | |
415 | """Format a block according to width.""" |
|
415 | """Format a block according to width.""" | |
416 | if width <= 0: |
|
416 | if width <= 0: | |
417 | width = 78 |
|
417 | width = 78 | |
418 | indent = ' ' * block['indent'] |
|
418 | indent = ' ' * block['indent'] | |
419 | if block['type'] == 'admonition': |
|
419 | if block['type'] == 'admonition': | |
420 | admonition = _admonitiontitles[block['admonitiontitle']] |
|
420 | admonition = _admonitiontitles[block['admonitiontitle']] | |
421 | hang = len(block['lines'][-1]) - len(block['lines'][-1].lstrip()) |
|
421 | hang = len(block['lines'][-1]) - len(block['lines'][-1].lstrip()) | |
422 |
|
422 | |||
423 | defindent = indent + hang * ' ' |
|
423 | defindent = indent + hang * ' ' | |
424 | text = ' '.join(map(str.strip, block['lines'])) |
|
424 | text = ' '.join(map(str.strip, block['lines'])) | |
425 | return '%s\n%s' % (indent + admonition, util.wrap(text, width=width, |
|
425 | return '%s\n%s' % (indent + admonition, util.wrap(text, width=width, | |
426 | initindent=defindent, |
|
426 | initindent=defindent, | |
427 | hangindent=defindent)) |
|
427 | hangindent=defindent)) | |
428 | if block['type'] == 'margin': |
|
428 | if block['type'] == 'margin': | |
429 | return '' |
|
429 | return '' | |
430 | if block['type'] == 'literal': |
|
430 | if block['type'] == 'literal': | |
431 | indent += ' ' |
|
431 | indent += ' ' | |
432 | return indent + ('\n' + indent).join(block['lines']) |
|
432 | return indent + ('\n' + indent).join(block['lines']) | |
433 | if block['type'] == 'section': |
|
433 | if block['type'] == 'section': | |
434 | underline = encoding.colwidth(block['lines'][0]) * block['underline'] |
|
434 | underline = encoding.colwidth(block['lines'][0]) * block['underline'] | |
435 | return "%s%s\n%s%s" % (indent, block['lines'][0],indent, underline) |
|
435 | return "%s%s\n%s%s" % (indent, block['lines'][0],indent, underline) | |
436 | if block['type'] == 'table': |
|
436 | if block['type'] == 'table': | |
437 | table = block['table'] |
|
437 | table = block['table'] | |
438 | # compute column widths |
|
438 | # compute column widths | |
439 | widths = [max([encoding.colwidth(e) for e in c]) for c in zip(*table)] |
|
439 | widths = [max([encoding.colwidth(e) for e in c]) for c in zip(*table)] | |
440 | text = '' |
|
440 | text = '' | |
441 | span = sum(widths) + len(widths) - 1 |
|
441 | span = sum(widths) + len(widths) - 1 | |
442 | indent = ' ' * block['indent'] |
|
442 | indent = ' ' * block['indent'] | |
443 | hang = ' ' * (len(indent) + span - widths[-1]) |
|
443 | hang = ' ' * (len(indent) + span - widths[-1]) | |
444 | f = ' '.join('%%-%ds' % n for n in widths) |
|
444 | f = ' '.join('%%-%ds' % n for n in widths) | |
445 |
|
445 | |||
446 | for row in table: |
|
446 | for row in table: | |
447 | l = f % tuple(row) |
|
447 | l = f % tuple(row) | |
448 | l = util.wrap(l, width=width, initindent=indent, hangindent=hang) |
|
448 | l = util.wrap(l, width=width, initindent=indent, hangindent=hang) | |
449 | if not text and block['header']: |
|
449 | if not text and block['header']: | |
450 | text = l + '\n' + indent + '-' * (min(width, span)) + '\n' |
|
450 | text = l + '\n' + indent + '-' * (min(width, span)) + '\n' | |
451 | else: |
|
451 | else: | |
452 | text += l + "\n" |
|
452 | text += l + "\n" | |
453 | return text |
|
453 | return text | |
454 | if block['type'] == 'definition': |
|
454 | if block['type'] == 'definition': | |
455 | term = indent + block['lines'][0] |
|
455 | term = indent + block['lines'][0] | |
456 | hang = len(block['lines'][-1]) - len(block['lines'][-1].lstrip()) |
|
456 | hang = len(block['lines'][-1]) - len(block['lines'][-1].lstrip()) | |
457 | defindent = indent + hang * ' ' |
|
457 | defindent = indent + hang * ' ' | |
458 | text = ' '.join(map(str.strip, block['lines'][1:])) |
|
458 | text = ' '.join(map(str.strip, block['lines'][1:])) | |
459 | return '%s\n%s' % (term, util.wrap(text, width=width, |
|
459 | return '%s\n%s' % (term, util.wrap(text, width=width, | |
460 | initindent=defindent, |
|
460 | initindent=defindent, | |
461 | hangindent=defindent)) |
|
461 | hangindent=defindent)) | |
462 | subindent = indent |
|
462 | subindent = indent | |
463 | if block['type'] == 'bullet': |
|
463 | if block['type'] == 'bullet': | |
464 | if block['lines'][0].startswith('| '): |
|
464 | if block['lines'][0].startswith('| '): | |
465 | # Remove bullet for line blocks and add no extra |
|
465 | # Remove bullet for line blocks and add no extra | |
466 | # indention. |
|
466 | # indention. | |
467 | block['lines'][0] = block['lines'][0][2:] |
|
467 | block['lines'][0] = block['lines'][0][2:] | |
468 | else: |
|
468 | else: | |
469 | m = _bulletre.match(block['lines'][0]) |
|
469 | m = _bulletre.match(block['lines'][0]) | |
470 | subindent = indent + m.end() * ' ' |
|
470 | subindent = indent + m.end() * ' ' | |
471 | elif block['type'] == 'field': |
|
471 | elif block['type'] == 'field': | |
472 | keywidth = block['keywidth'] |
|
472 | keywidth = block['keywidth'] | |
473 | key = block['key'] |
|
473 | key = block['key'] | |
474 |
|
474 | |||
475 | subindent = indent + _fieldwidth * ' ' |
|
475 | subindent = indent + _fieldwidth * ' ' | |
476 | if len(key) + 2 > _fieldwidth: |
|
476 | if len(key) + 2 > _fieldwidth: | |
477 | # key too large, use full line width |
|
477 | # key too large, use full line width | |
478 | key = key.ljust(width) |
|
478 | key = key.ljust(width) | |
479 | elif keywidth + 2 < _fieldwidth: |
|
479 | elif keywidth + 2 < _fieldwidth: | |
480 | # all keys are small, add only two spaces |
|
480 | # all keys are small, add only two spaces | |
481 | key = key.ljust(keywidth + 2) |
|
481 | key = key.ljust(keywidth + 2) | |
482 | subindent = indent + (keywidth + 2) * ' ' |
|
482 | subindent = indent + (keywidth + 2) * ' ' | |
483 | else: |
|
483 | else: | |
484 | # mixed sizes, use fieldwidth for this one |
|
484 | # mixed sizes, use fieldwidth for this one | |
485 | key = key.ljust(_fieldwidth) |
|
485 | key = key.ljust(_fieldwidth) | |
486 | block['lines'][0] = key + block['lines'][0] |
|
486 | block['lines'][0] = key + block['lines'][0] | |
487 | elif block['type'] == 'option': |
|
487 | elif block['type'] == 'option': | |
488 | return formatoption(block, width) |
|
488 | return formatoption(block, width) | |
489 |
|
489 | |||
490 | text = ' '.join(map(str.strip, block['lines'])) |
|
490 | text = ' '.join(map(str.strip, block['lines'])) | |
491 | return util.wrap(text, width=width, |
|
491 | return util.wrap(text, width=width, | |
492 | initindent=indent, |
|
492 | initindent=indent, | |
493 | hangindent=subindent) |
|
493 | hangindent=subindent) | |
494 |
|
494 | |||
495 | def parse(text, indent=0, keep=None): |
|
495 | def parse(text, indent=0, keep=None): | |
496 | """Parse text into a list of blocks""" |
|
496 | """Parse text into a list of blocks""" | |
497 | pruned = [] |
|
497 | pruned = [] | |
498 | blocks = findblocks(text) |
|
498 | blocks = findblocks(text) | |
499 | for b in blocks: |
|
499 | for b in blocks: | |
500 | b['indent'] += indent |
|
500 | b['indent'] += indent | |
501 | blocks = findliteralblocks(blocks) |
|
501 | blocks = findliteralblocks(blocks) | |
502 | blocks = findtables(blocks) |
|
502 | blocks = findtables(blocks) | |
503 | blocks, pruned = prunecontainers(blocks, keep or []) |
|
503 | blocks, pruned = prunecontainers(blocks, keep or []) | |
504 | blocks = findsections(blocks) |
|
504 | blocks = findsections(blocks) | |
505 | blocks = inlineliterals(blocks) |
|
505 | blocks = inlineliterals(blocks) | |
506 | blocks = hgrole(blocks) |
|
506 | blocks = hgrole(blocks) | |
507 | blocks = splitparagraphs(blocks) |
|
507 | blocks = splitparagraphs(blocks) | |
508 | blocks = updatefieldlists(blocks) |
|
508 | blocks = updatefieldlists(blocks) | |
509 | blocks = updateoptionlists(blocks) |
|
509 | blocks = updateoptionlists(blocks) | |
510 | blocks = addmargins(blocks) |
|
510 | blocks = addmargins(blocks) | |
511 | blocks = prunecomments(blocks) |
|
511 | blocks = prunecomments(blocks) | |
512 | blocks = findadmonitions(blocks) |
|
512 | blocks = findadmonitions(blocks) | |
513 | return blocks, pruned |
|
513 | return blocks, pruned | |
514 |
|
514 | |||
515 | def formatblocks(blocks, width): |
|
515 | def formatblocks(blocks, width): | |
516 | text = '\n'.join(formatblock(b, width) for b in blocks) |
|
516 | text = '\n'.join(formatblock(b, width) for b in blocks) | |
517 | return text |
|
517 | return text | |
518 |
|
518 | |||
519 | def format(text, width, indent=0, keep=None): |
|
519 | def format(text, width, indent=0, keep=None): | |
520 | """Parse and format the text according to width.""" |
|
520 | """Parse and format the text according to width.""" | |
521 | blocks, pruned = parse(text, indent, keep or []) |
|
521 | blocks, pruned = parse(text, indent, keep or []) | |
522 | text = '\n'.join(formatblock(b, width) for b in blocks) |
|
522 | text = '\n'.join(formatblock(b, width) for b in blocks) | |
523 | if keep is None: |
|
523 | if keep is None: | |
524 | return text |
|
524 | return text | |
525 | else: |
|
525 | else: | |
526 | return text, pruned |
|
526 | return text, pruned | |
527 |
|
527 | |||
528 | def getsections(blocks): |
|
528 | def getsections(blocks): | |
529 | '''return a list of (section name, nesting level, blocks) tuples''' |
|
529 | '''return a list of (section name, nesting level, blocks) tuples''' | |
530 | nest = "" |
|
530 | nest = "" | |
531 | level = 0 |
|
531 | level = 0 | |
532 | secs = [] |
|
532 | secs = [] | |
533 | for b in blocks: |
|
533 | for b in blocks: | |
534 | if b['type'] == 'section': |
|
534 | if b['type'] == 'section': | |
535 | i = b['underline'] |
|
535 | i = b['underline'] | |
536 | if i not in nest: |
|
536 | if i not in nest: | |
537 | nest += i |
|
537 | nest += i | |
538 | level = nest.index(i) + 1 |
|
538 | level = nest.index(i) + 1 | |
539 | nest = nest[:level] |
|
539 | nest = nest[:level] | |
540 | secs.append((b['lines'][0], level, [b])) |
|
540 | secs.append((b['lines'][0], level, [b])) | |
541 | else: |
|
541 | else: | |
542 | if not secs: |
|
542 | if not secs: | |
543 | # add an initial empty section |
|
543 | # add an initial empty section | |
544 | secs = [('', 0, [])] |
|
544 | secs = [('', 0, [])] | |
545 | secs[-1][2].append(b) |
|
545 | secs[-1][2].append(b) | |
546 | return secs |
|
546 | return secs | |
547 |
|
547 | |||
548 | def decorateblocks(blocks, width): |
|
548 | def decorateblocks(blocks, width): | |
549 | '''generate a list of (section name, line text) pairs for search''' |
|
549 | '''generate a list of (section name, line text) pairs for search''' | |
550 | lines = [] |
|
550 | lines = [] | |
551 | for s in getsections(blocks): |
|
551 | for s in getsections(blocks): | |
552 | section = s[0] |
|
552 | section = s[0] | |
553 | text = formatblocks(s[2], width) |
|
553 | text = formatblocks(s[2], width) | |
554 | lines.append([(section, l) for l in text.splitlines(True)]) |
|
554 | lines.append([(section, l) for l in text.splitlines(True)]) | |
555 | return lines |
|
555 | return lines | |
556 |
|
556 | |||
557 | def maketable(data, indent=0, header=False): |
|
557 | def maketable(data, indent=0, header=False): | |
558 | '''Generate an RST table for the given table data''' |
|
558 | '''Generate an RST table for the given table data''' | |
559 |
|
559 | |||
560 | widths = [max(encoding.colwidth(e) for e in c) for c in zip(*data)] |
|
560 | widths = [max(encoding.colwidth(e) for e in c) for c in zip(*data)] | |
561 | indent = ' ' * indent |
|
561 | indent = ' ' * indent | |
562 | f = indent + ' '.join('%%-%ds' % w for w in widths) + '\n' |
|
562 | f = indent + ' '.join('%%-%ds' % w for w in widths) + '\n' | |
563 | div = indent + ' '.join('=' * w for w in widths) + '\n' |
|
563 | div = indent + ' '.join('=' * w for w in widths) + '\n' | |
564 |
|
564 | |||
565 | out = [div] |
|
565 | out = [div] | |
566 | for row in data: |
|
566 | for row in data: | |
567 | out.append(f % tuple(row)) |
|
567 | out.append(f % tuple(row)) | |
568 | if header and len(data) > 1: |
|
568 | if header and len(data) > 1: | |
569 | out.insert(2, div) |
|
569 | out.insert(2, div) | |
570 | out.append(div) |
|
570 | out.append(div) | |
571 | return ''.join(out) |
|
571 | return ''.join(out) |
General Comments 0
You need to be logged in to leave comments.
Login now