1 # This file is part of lyx2lyx
2 # -*- coding: utf-8 -*-
3 # Copyright (C) 2011 The LyX team
5 # This program is free software; you can redistribute it and/or
6 # modify it under the terms of the GNU General Public License
7 # as published by the Free Software Foundation; either version 2
8 # of the License, or (at your option) any later version.
10 # This program is distributed in the hope that it will be useful,
11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 # GNU General Public License for more details.
15 # You should have received a copy of the GNU General Public License
16 # along with this program; if not, write to the Free Software
17 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20 This module offers several free functions to help with lyx2lyx'ing.
21 More documentaton is below, but here is a quick guide to what
22 they do. Optional arguments are marked by brackets.
24 add_to_preamble(document, text):
25 Here, text can be either a single line or a list of lines. It
26 is bad practice to pass something with embedded newlines, but
27 we will handle that properly.
28 The routine checks to see whether the provided material is
29 already in the preamble. If not, it adds it.
30 Prepends a comment "% Added by lyx2lyx" to text.
32 insert_to_preamble(document, text[, index]):
33 Here, text can be either a single line or a list of lines. It
34 is bad practice to pass something with embedded newlines, but
35 we will handle that properly.
36 The routine inserts text at document.preamble[index], where by
37 default index is 0, so the material is inserted at the beginning.
38 Prepends a comment "% Added by lyx2lyx" to text.
41 Here arg should be a list of strings (lines), which we want to
42 wrap in ERT. Returns a list of strings so wrapped.
43 A call to this routine will often go something like this:
44 i = find_token('\\begin_inset FunkyInset', ...)
45 j = find_end_of_inset(document.body, i)
46 content = lyx2latex(document[i:j + 1])
47 ert = put_cmd_in_ert(content)
48 document.body[i:j+1] = ert
50 lyx2latex(document, lines):
51 Here, lines is a list of lines of LyX material we want to convert
52 to LaTeX. We do the best we can and return a string containing
53 the translated material.
55 lyx2verbatim(document, lines):
56 Here, lines is a list of lines of LyX material we want to convert
57 to verbatim material (used in ERT an the like). We do the best we
58 can and return a string containing the translated material.
61 Convert lengths (in LyX form) to their LaTeX representation. Returns
62 (bool, length), where the bool tells us if it was a percentage, and
63 the length is the LaTeX representation.
69 from parser_tools import find_token, find_end_of_inset
70 from unicode_symbols import unicode_reps
73 # This will accept either a list of lines or a single line.
74 # It is bad practice to pass something with embedded newlines,
75 # though we will handle that.
76 def add_to_preamble(document, text):
77 " Add text to the preamble if it is not already there. "
79 if not type(text) is list:
80 # split on \n just in case
81 # it'll give us the one element list we want
82 # if there's no \n, too
83 text = text.split('\n')
86 prelen = len(document.preamble)
88 i = find_token(document.preamble, text[0], i)
91 # we need a perfect match
94 if i >= prelen or line != document.preamble[i]:
101 document.preamble.extend(["% Added by lyx2lyx"])
102 document.preamble.extend(text)
105 # Note that text can be either a list of lines or a single line.
106 # It should really be a list.
107 def insert_to_preamble(document, text, index = 0):
108 """ Insert text to the preamble at a given line"""
110 if not type(text) is list:
111 # split on \n just in case
112 # it'll give us the one element list we want
113 # if there's no \n, too
114 text = text.split('\n')
116 text.insert(0, "% Added by lyx2lyx")
117 document.preamble[index:index] = text
120 def put_cmd_in_ert(arg):
122 arg should be a list of lines we want to wrap in ERT.
123 Returns a list of strings, with the lines so wrapped.
126 ret = ["\\begin_inset ERT", "status collapsed", "", "\\begin_layout Plain Layout", ""]
127 # It will be faster for us to work with a single string internally.
128 # That way, we only go through the unicode_reps loop once.
129 if type(arg) is list:
133 for rep in unicode_reps:
134 s = s.replace(rep[1], rep[0])
135 s = s.replace('\\', "\\backslash\n")
136 ret += s.splitlines()
137 ret += ["\\end_layout", "", "\\end_inset"]
141 def get_ert(lines, i, verbatim = False):
142 'Convert an ERT inset into LaTeX.'
143 if not lines[i].startswith("\\begin_inset ERT"):
145 j = find_end_of_inset(lines, i)
148 while i < j and not lines[i].startswith("status"):
154 if lines[i] == "\\begin_layout Plain Layout":
159 while i + 1 < j and lines[i+1] == "":
161 elif lines[i] == "\\end_layout":
162 while i + 1 < j and lines[i+1] == "":
164 elif lines[i] == "\\backslash" and not verbatim:
172 def lyx2latex(document, lines):
173 'Convert some LyX stuff into corresponding LaTeX stuff, as best we can.'
180 for curline in range(len(lines)):
181 line = lines[curline]
182 if line.startswith("\\begin_inset Note Note"):
183 # We want to skip LyX notes, so remember where the inset ends
184 note_end = find_end_of_inset(lines, curline + 1)
186 elif note_end >= curline:
189 elif line.startswith("\\begin_inset ERT"):
190 # We don't want to replace things inside ERT, so figure out
191 # where the end of the inset is.
192 ert_end = find_end_of_inset(lines, curline + 1)
194 elif line.startswith("\\begin_inset Formula"):
196 elif line.startswith("\\begin_inset Quotes"):
197 # For now, we do a very basic reversion. Someone who understands
198 # quotes is welcome to fix it up.
199 qtype = line[20:].strip()
213 elif line.startswith("\\begin_inset Newline newline"):
215 elif line.startswith("\\noindent"):
216 line = "\\noindent " # we need the space behind the command
217 elif line.startswith("\\begin_inset space"):
218 line = line[18:].strip()
219 if line.startswith("\\hspace"):
220 # Account for both \hspace and \hspace*
223 elif line == "\\space{}":
225 elif line == "\\thinspace{}":
228 # The LyX length is in line[8:], after the \length keyword
229 length = latex_length(line[8:])[1]
230 line = hspace + "{" + length + "}"
232 elif line.isspace() or \
233 line.startswith("\\begin_layout") or \
234 line.startswith("\\end_layout") or \
235 line.startswith("\\begin_inset") or \
236 line.startswith("\\end_inset") or \
237 line.startswith("\\lang") or \
238 line.strip() == "status collapsed" or \
239 line.strip() == "status open":
243 # this needs to be added to the preamble because of cases like
244 # \textmu, \textbackslash, etc.
245 add_to_preamble(document, ['% added by lyx2lyx for converted index entries',
246 '\\@ifundefined{textmu}',
247 ' {\\usepackage{textcomp}}{}'])
248 # a lossless reversion is not possible
249 # try at least to handle some common insets and settings
250 if ert_end >= curline:
251 line = line.replace(r'\backslash', '\\')
253 # No need to add "{}" after single-nonletter macros
254 line = line.replace('&', '\\&')
255 line = line.replace('#', '\\#')
256 line = line.replace('^', '\\textasciicircum{}')
257 line = line.replace('%', '\\%')
258 line = line.replace('_', '\\_')
259 line = line.replace('$', '\\$')
261 # Do the LyX text --> LaTeX conversion
262 for rep in unicode_reps:
263 line = line.replace(rep[1], rep[0])
264 line = line.replace(r'\backslash', r'\textbackslash{}')
265 line = line.replace(r'\series bold', r'\bfseries{}').replace(r'\series default', r'\mdseries{}')
266 line = line.replace(r'\shape italic', r'\itshape{}').replace(r'\shape smallcaps', r'\scshape{}')
267 line = line.replace(r'\shape slanted', r'\slshape{}').replace(r'\shape default', r'\upshape{}')
268 line = line.replace(r'\emph on', r'\em{}').replace(r'\emph default', r'\em{}')
269 line = line.replace(r'\noun on', r'\scshape{}').replace(r'\noun default', r'\upshape{}')
270 line = line.replace(r'\bar under', r'\underbar{').replace(r'\bar default', r'}')
271 line = line.replace(r'\family sans', r'\sffamily{}').replace(r'\family default', r'\normalfont{}')
272 line = line.replace(r'\family typewriter', r'\ttfamily{}').replace(r'\family roman', r'\rmfamily{}')
273 line = line.replace(r'\InsetSpace ', r'').replace(r'\SpecialChar ', r'')
278 def lyx2verbatim(document, lines):
279 'Convert some LyX stuff into corresponding verbatim stuff, as best we can.'
281 content = lyx2latex(document, lines)
282 content = re.sub(r'\\(?!backslash)', r'\n\\backslash\n', content)
287 def latex_length(slen):
289 Convert lengths to their LaTeX representation. Returns (bool, length),
290 where the bool tells us if it was a percentage, and the length is the
291 LaTeX representation.
295 # the slen has the form
296 # ValueUnit+ValueUnit-ValueUnit or
297 # ValueUnit+-ValueUnit
298 # the + and - (glue lengths) are optional
299 # the + always precedes the -
301 # Convert relative lengths to LaTeX units
302 units = {"text%":"\\textwidth", "col%":"\\columnwidth",
303 "page%":"\\paperwidth", "line%":"\\linewidth",
304 "theight%":"\\textheight", "pheight%":"\\paperheight"}
305 for unit in list(units.keys()):
310 minus = slen.rfind("-", 1, i)
311 plus = slen.rfind("+", 0, i)
312 latex_unit = units[unit]
313 if plus == -1 and minus == -1:
315 value = str(float(value)/100)
316 end = slen[i + len(unit):]
317 slen = value + latex_unit + end
319 value = slen[plus + 1:i]
320 value = str(float(value)/100)
321 begin = slen[:plus + 1]
322 end = slen[i+len(unit):]
323 slen = begin + value + latex_unit + end
325 value = slen[minus + 1:i]
326 value = str(float(value)/100)
327 begin = slen[:minus + 1]
328 slen = begin + value + latex_unit
330 # replace + and -, but only if the - is not the first character
331 slen = slen[0] + slen[1:].replace("+", " plus ").replace("-", " minus ")
332 # handle the case where "+-1mm" was used, because LaTeX only understands
333 # "plus 1mm minus 1mm"
334 if slen.find("plus minus"):
335 lastvaluepos = slen.rfind(" ")
336 lastvalue = slen[lastvaluepos:]
337 slen = slen.replace(" ", lastvalue + " ")
338 return (percent, slen)
341 def length_in_bp(length):
342 " Convert a length in LyX format to its value in bp units "
344 em_width = 10.0 / 72.27 # assume 10pt font size
345 text_width = 8.27 / 1.7 # assume A4 with default margins
346 # scale factors are taken from Length::inInch()
347 scales = {"bp" : 1.0,
348 "cc" : (72.0 / (72.27 / (12.0 * 0.376 * 2.845))),
349 "cm" : (72.0 / 2.54),
350 "dd" : (72.0 / (72.27 / (0.376 * 2.845))),
351 "em" : (72.0 * em_width),
352 "ex" : (72.0 * em_width * 0.4305),
354 "mm" : (72.0 / 25.4),
355 "mu" : (72.0 * em_width / 18.0),
356 "pc" : (72.0 / (72.27 / 12.0)),
357 "pt" : (72.0 / (72.27)),
358 "sp" : (72.0 / (72.27 * 65536.0)),
359 "text%" : (72.0 * text_width / 100.0),
360 "col%" : (72.0 * text_width / 100.0), # assume 1 column
361 "page%" : (72.0 * text_width * 1.7 / 100.0),
362 "line%" : (72.0 * text_width / 100.0),
363 "theight%" : (72.0 * text_width * 1.787 / 100.0),
364 "pheight%" : (72.0 * text_width * 2.2 / 100.0)}
366 rx = re.compile(r'^\s*([^a-zA-Z%]+)([a-zA-Z%]+)\s*$')
369 document.warning("Invalid length value: " + length + ".")
373 if not unit in scales.keys():
374 document.warning("Unknown length unit: " + unit + ".")
376 return "%g" % (float(value) * scales[unit])
379 def revert_flex_inset(lines, name, LaTeXname):
380 " Convert flex insets to TeX code "
383 i = find_token(lines, '\\begin_inset Flex ' + name, i)
386 z = find_end_of_inset(lines, i)
388 document.warning("Can't find end of Flex " + name + " inset.")
391 # remove the \end_inset
392 lines[z - 2:z + 1] = put_cmd_in_ert("}")
393 # we need to reset character layouts if necessary
394 j = find_token(lines, '\\emph on', i, z)
395 k = find_token(lines, '\\noun on', i, z)
396 l = find_token(lines, '\\series', i, z)
397 m = find_token(lines, '\\family', i, z)
398 n = find_token(lines, '\\shape', i, z)
399 o = find_token(lines, '\\color', i, z)
400 p = find_token(lines, '\\size', i, z)
401 q = find_token(lines, '\\bar under', i, z)
402 r = find_token(lines, '\\uuline on', i, z)
403 s = find_token(lines, '\\uwave on', i, z)
404 t = find_token(lines, '\\strikeout on', i, z)
406 lines.insert(z - 2, "\\emph default")
408 lines.insert(z - 2, "\\noun default")
410 lines.insert(z - 2, "\\series default")
412 lines.insert(z - 2, "\\family default")
414 lines.insert(z - 2, "\\shape default")
416 lines.insert(z - 2, "\\color inherit")
418 lines.insert(z - 2, "\\size default")
420 lines.insert(z - 2, "\\bar default")
422 lines.insert(z - 2, "\\uuline default")
424 lines.insert(z - 2, "\\uwave default")
426 lines.insert(z - 2, "\\strikeout default")
427 lines[i:i + 4] = put_cmd_in_ert(LaTeXname + "{")
431 def revert_font_attrs(lines, name, LaTeXname):
432 " Reverts font changes to TeX code "
436 i = find_token(lines, name + ' on', i)
439 j = find_token(lines, name + ' default', i)
440 k = find_token(lines, name + ' on', i + 1)
441 # if there is no default set, the style ends with the layout
442 # assure hereby that we found the correct layout end
443 if j != -1 and (j < k or k == -1):
444 lines[j:j + 1] = put_cmd_in_ert("}")
446 j = find_token(lines, '\\end_layout', i)
447 lines[j:j] = put_cmd_in_ert("}")
448 lines[i:i + 1] = put_cmd_in_ert(LaTeXname + "{")
453 def revert_layout_command(lines, name, LaTeXname):
454 " Reverts a command from a layout to TeX code "
457 i = find_token(lines, '\\begin_layout ' + name, i)
461 # find the next layout
464 j = find_token(lines, '\\begin_layout', j)
466 # if nothing was found it was the last layout of the document
468 lines[l - 4:l - 4] = put_cmd_in_ert("}")
470 # exclude plain layout because this can be TeX code or another inset
471 elif lines[j] != '\\begin_layout Plain Layout':
472 lines[j - 2:j - 2] = put_cmd_in_ert("}")
476 lines[i] = '\\begin_layout Standard'
477 lines[i + 1:i + 1] = put_cmd_in_ert(LaTeXname + "{")
482 " Converts an RRGGBB-type hexadecimal string to a float in [0.0,1.0] "
489 return str(val / 256.0)
493 "'true' goes to True, case-insensitively, and we strip whitespace."
494 s = s.strip().lower()