1 # This file is part of lyx2lyx
2 # -*- coding: utf-8 -*-
3 # Copyright (C) 2002-2011 Dekel Tsur <dekel@lyx.org>,
4 # José Matos <jamatos@lyx.org>, Richard Heck <rgheck@comcast.net>
6 # This program is free software; you can redistribute it and/or
7 # modify it under the terms of the GNU General Public License
8 # as published by the Free Software Foundation; either version 2
9 # of the License, or (at your option) any later version.
11 # This program is distributed in the hope that it will be useful,
12 # but WITHOUT ANY WARRANTY; without even the implied warranty of
13 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 # GNU General Public License for more details.
16 # You should have received a copy of the GNU General Public License
17 # along with this program; if not, write to the Free Software
18 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
22 This module offers several free functions to help parse lines.
23 More documentaton is below, but here is a quick guide to what
24 they do. Optional arguments are marked by brackets.
26 find_token(lines, token, start[, end[, ignorews]]):
27 Returns the first line i, start <= i < end, on which
28 token is found at the beginning. Returns -1 if not
30 If ignorews is (given and) True, then differences
31 in whitespace do not count, except that there must be no
32 extra whitespace following token itself.
34 find_token_exact(lines, token, start[, end]):
35 As find_token, but with ignorews set to True.
37 find_tokens(lines, tokens, start[, end[, ignorews]]):
38 Returns the first line i, start <= i < end, on which
39 one of the tokens in tokens is found at the beginning.
40 Returns -1 if not found.
41 If ignorews is (given and) True, then differences
42 in whitespace do not count, except that there must be no
43 extra whitespace following token itself.
45 find_tokens_exact(lines, token, start[, end]):
46 As find_tokens, but with ignorews True.
48 find_token_backwards(lines, token, start):
49 find_tokens_backwards(lines, tokens, start):
50 As before, but look backwards.
52 find_re(lines, rexp, start[, end]):
53 As find_token, but rexp is a regular expression object,
54 so it has to be passed as e.g.: re.compile(r'...').
56 get_value(lines, token, start[, end[, default]):
57 Similar to find_token, but it returns what follows the
58 token on the found line. Example:
59 get_value(document.header, "\\use_xetex", 0)
60 will find a line like:
62 and, in that case, return "true". (Note that whitespace
63 is stripped.) The final argument, default, defaults to "",
64 and is what is returned if we do not find anything. So you
65 can use that to set a default.
67 get_quoted_value(lines, token, start[, end[, default]]):
68 Similar to get_value, but it will strip quotes off the
69 value, if they are present. So use this one for cases
70 where the value is normally quoted.
72 get_option_value(line, option):
73 This assumes we have a line with something like:
75 and returns value. Returns "" if not found.
77 get_bool_value(lines, token, start[, end[, default]]):
78 Like get_value, but returns a boolean.
80 del_token(lines, token, start[, end]):
81 Like find_token, but deletes the line if it finds one.
82 Returns True if a line got deleted, otherwise False.
84 find_beginning_of(lines, i, start_token, end_token):
85 Here, start_token and end_token are meant to be a matching
86 pair, like "\\begin_layout" and "\\end_layout". We look for
87 the start_token that pairs with the end_token that occurs
88 on or after line i. Returns -1 if not found.
89 So, in the layout case, this would find the \\begin_layout
90 for the layout line i is in.
92 ec = find_token(document.body, "</cell", i)
93 bc = find_beginning_of(document.body, ec, \
95 Now, assuming no -1s, bc-ec wraps the cell for line i.
97 find_end_of(lines, i, start_token, end_token):
98 Like find_beginning_of, but looking for the matching
99 end_token. This might look like:
100 bc = find_token_(document.body, "<cell", i)
101 ec = find_end_of(document.body, bc, "<cell", "</cell")
102 Now, assuming no -1s, bc-ec wrap the next cell.
104 find_end_of_inset(lines, i):
105 Specialization of find_end_of for insets.
107 find_end_of_layout(lines, i):
108 Specialization of find_end_of for layouts.
110 find_end_of_sequence(lines, i):
111 Find the end of the sequence of layouts of the same kind.
112 Considers nesting. If the last paragraph in sequence is nested,
113 the position of the last \end_deeper is returned, else
114 the position of the last \end_layout.
116 is_in_inset(lines, i, inset):
117 Checks if line i is in an inset of the given type.
118 If so, returns starting and ending lines. Otherwise,
121 is_in_inset(document.body, i, "\\begin_inset Tabular")
122 returns False unless i is within a table. If it is, then
123 it returns the line on which the table begins and the one
124 on which it ends. Note that this pair will evaulate to
127 will do what you expect.
129 get_containing_inset(lines, i):
130 Finds out what kind of inset line i is within. Returns a
131 list containing what follows \begin_inset on the line
132 on which the inset begins, plus the starting and ending line.
133 Returns False on any kind of error or if it isn't in an inset.
134 So get_containing_inset(document.body, i) might return:
135 ("CommandInset ref", 300, 306)
136 if i is within an InsetRef beginning on line 300 and ending
139 get_containing_layout(lines, i):
140 As get_containing_inset, but for layout. Additionally returns the
141 position of real paragraph start (after par params) as 4th value.
143 find_nonempty_line(lines, start[, end):
144 Finds the next non-empty line.
146 check_token(line, token):
147 Does line begin with token?
149 is_nonempty_line(line):
150 Does line contain something besides whitespace?
152 count_pars_in_inset(lines, i):
153 Counts the paragraphs inside an inset.
159 # Fast search in lists
160 def find_slice(l, sl, start = 0, stop = None):
161 """Return position of first occurence of sequence `sl` in list `l`
164 >>> find_slice([1, 2, 3, 1, 1, 2], (1, 2))
167 The return value can be used to delete or substitute the sub-list:
169 >>> l = [1, 0, 1, 1, 1, 2]
170 >>> s = find_slice(l, [0, 1, 1])
173 >>> s = find_slice(l, (1, 2))
177 The start argument works similar to list.index()
179 >>> find_slice([1, 2, 3, 1, 1 ,2], (1, 2), start = 1)
182 Use the `stop` attribute of the returned `slice` to test for success:
184 >>> s1 = find_slice([2, 3, 1], (3, 1))
185 >>> s2 = find_slice([2, 3, 1], (2, 1))
186 >>> if s1.stop and not s2.stop:
190 stop = stop or len(l)
191 N = len(sl) # lenght of sub-list
194 for j, value in enumerate(sl):
195 i = l.index(value, start, stop)
201 return slice(i+1-N, i+1)
202 except ValueError: # sub list `sl` not found
206 # Utilities for one line
207 def check_token(line, token):
208 """ check_token(line, token) -> bool
210 Return True if token is present in line and is the first element
213 Deprecated. Use line.startswith(token).
216 return line.startswith(token)
219 def is_nonempty_line(line):
220 """ is_nonempty_line(line) -> bool
222 Return False if line is either empty or it has only whitespaces,
224 return bool(line.strip())
227 # Utilities for a list of lines
228 def find_token(lines, token, start, end = 0, ignorews = False):
229 """ find_token(lines, token, start[[, end], ignorews]) -> int
231 Return the lowest line where token is found, and is the first
232 element, in lines[start, end].
234 If ignorews is True (default is False), then differences in
235 whitespace are ignored, except that there must be no extra
236 whitespace following token itself.
238 Return -1 on failure."""
240 if end == 0 or end > len(lines):
243 for i in range(start, end):
252 if lines[i][:m] == token:
257 def find_token_exact(lines, token, start, end = 0):
258 return find_token(lines, token, start, end, True)
261 def find_tokens(lines, tokens, start, end = 0, ignorews = False):
262 """ find_tokens(lines, tokens, start[[, end], ignorews]) -> int
264 Return the lowest line where one token in tokens is found, and is
265 the first element, in lines[start, end].
267 Return -1 on failure."""
268 if end == 0 or end > len(lines):
271 for i in range(start, end):
281 if lines[i][:len(token)] == token:
286 def find_tokens_exact(lines, tokens, start, end = 0):
287 return find_tokens(lines, tokens, start, end, True)
290 def find_re(lines, rexp, start, end = 0):
291 """ find_token_re(lines, rexp, start[, end]) -> int
293 Return the lowest line where rexp, a regular expression, is found
294 in lines[start, end].
296 Return -1 on failure."""
298 if end == 0 or end > len(lines):
300 for i in range(start, end):
301 if rexp.match(lines[i]):
306 def find_token_backwards(lines, token, start):
307 """ find_token_backwards(lines, token, start) -> int
309 Return the highest line where token is found, and is the first
310 element, in lines[start, end].
312 Return -1 on failure."""
314 for i in range(start, -1, -1):
316 if line[:m] == token:
321 def find_tokens_backwards(lines, tokens, start):
322 """ find_tokens_backwards(lines, token, start) -> int
324 Return the highest line where token is found, and is the first
325 element, in lines[end, start].
327 Return -1 on failure."""
328 for i in range(start, -1, -1):
331 if line[:len(token)] == token:
336 def get_value(lines, token, start, end = 0, default = ""):
337 """ get_value(lines, token, start[[, end], default]) -> string
339 Find the next line that looks like:
340 token followed by other stuff
341 Returns "followed by other stuff" with leading and trailing
345 i = find_token_exact(lines, token, start, end)
348 l = lines[i].split(None, 1)
354 def get_quoted_value(lines, token, start, end = 0, default = ""):
355 """ get_quoted_value(lines, token, start[[, end], default]) -> string
357 Find the next line that looks like:
358 token "followed by other stuff"
359 Returns "followed by other stuff" with leading and trailing
360 whitespace and quotes removed. If there are no quotes, that is OK too.
361 So use get_value to preserve possible quotes, this one to remove them,
363 Note that we will NOT strip quotes from default!
365 val = get_value(lines, token, start, end, "")
368 return val.strip('"')
371 def get_bool_value(lines, token, start, end = 0, default = None):
372 """ get_value(lines, token, start[[, end], default]) -> string
374 Find the next line that looks like:
377 Returns True if bool_value is 1 or true and
378 False if bool_value is 0 or false
381 val = get_quoted_value(lines, token, start, end, "")
383 if val == "1" or val == "true":
385 if val == "0" or val == "false":
390 def get_option_value(line, option):
391 rx = option + '\s*=\s*"([^"]+)"'
399 def set_option_value(line, option, value):
400 rx = '(' + option + '\s*=\s*")[^"]+"'
405 return re.sub(rx, '\g<1>' + value + '"', line)
408 def del_token(lines, token, start, end = 0):
409 """ del_token(lines, token, start, end) -> int
411 Find the first line in lines where token is the first element
412 and delete that line. Returns True if we deleted a line, False
415 k = find_token_exact(lines, token, start, end)
422 def find_beginning_of(lines, i, start_token, end_token):
425 i = find_tokens_backwards(lines, [start_token, end_token], i-1)
428 if check_token(lines[i], end_token):
437 def find_end_of(lines, i, start_token, end_token):
441 i = find_tokens(lines, [end_token, start_token], i+1)
444 if check_token(lines[i], start_token):
453 def find_nonempty_line(lines, start, end = 0):
456 for i in range(start, end):
457 if is_nonempty_line(lines[i]):
462 def find_end_of_inset(lines, i):
463 " Find end of inset, where lines[i] is included."
464 return find_end_of(lines, i, "\\begin_inset", "\\end_inset")
467 def find_end_of_layout(lines, i):
468 " Find end of layout, where lines[i] is included."
469 return find_end_of(lines, i, "\\begin_layout", "\\end_layout")
472 def is_in_inset(lines, i, inset):
474 Checks if line i is in an inset of the given type.
475 If so, returns starting and ending lines.
476 Otherwise, returns False.
478 is_in_inset(document.body, i, "\\begin_inset Tabular")
479 returns False unless i is within a table. If it is, then
480 it returns the line on which the table begins and the one
481 on which it ends. Note that this pair will evaulate to
484 will do what you expect.
487 stins = find_token_backwards(lines, inset, i)
490 endins = find_end_of_inset(lines, stins)
491 # note that this includes the notfound case.
494 return (stins, endins)
497 def get_containing_inset(lines, i):
499 Finds out what kind of inset line i is within. Returns a
500 list containing (i) what follows \begin_inset on the line
501 on which the inset begins, plus the starting and ending line.
502 Returns False on any kind of error or if it isn't in an inset.
506 stins = find_token_backwards(lines, "\\begin_inset", j)
509 endins = find_end_of_inset(lines, stins)
517 inset = get_value(lines, "\\begin_inset", stins)
521 return (inset, stins, endins)
524 def get_containing_layout(lines, i):
526 Finds out what kind of layout line i is within. Returns a
527 list containing what follows \begin_layout on the line
528 on which the layout begins, plus the starting and ending line
529 and the start of the paragraph (after all params). I.e, returns:
530 (layoutname, layoutstart, layoutend, startofcontent)
531 Returns False on any kind of error.
535 stlay = find_token_backwards(lines, "\\begin_layout", j)
538 endlay = find_end_of_layout(lines, stlay)
546 lay = get_value(lines, "\\begin_layout", stlay)
550 par_params = ["\\noindent", "\\indent", "\\indent-toggle", "\\leftindent",
551 "\\start_of_appendix", "\\paragraph_spacing", "\\align",
552 "\\labelwidthstring"]
556 if lines[stpar].split(' ', 1)[0] not in par_params:
558 return (lay, stlay, endlay, stpar)
561 def count_pars_in_inset(lines, i):
563 Counts the paragraphs within this inset
565 ins = get_containing_inset(lines, i)
569 for j in range(ins[1], ins[2]):
570 m = re.match(r'\\begin_layout (.*)', lines[j])
571 if m and get_containing_inset(lines, j)[0] == ins[0]:
577 def find_end_of_sequence(lines, i):
579 Returns the end of a sequence of identical layouts.
581 lay = get_containing_layout(lines, i)
588 m = re.match(r'\\begin_layout (.*)', lines[i])
589 if m and m.group(1) != layout:
591 elif lines[i] == "\\begin_deeper":
592 j = find_end_of(lines, i, "\\begin_deeper", "\\end_deeper")
597 if m and m.group(1) == layout:
598 endlay = find_end_of_layout(lines, i)
601 if i == len(lines) - 1: