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.
56 Convert lengths (in LyX form) to their LaTeX representation. Returns
57 (bool, length), where the bool tells us if it was a percentage, and
58 the length is the LaTeX representation.
63 from parser_tools import find_token, find_end_of_inset
64 from unicode_symbols import unicode_reps
67 # This will accept either a list of lines or a single line.
68 # It is bad practice to pass something with embedded newlines,
69 # though we will handle that.
70 def add_to_preamble(document, text):
71 " Add text to the preamble if it is not already there. "
73 if not type(text) is list:
74 # split on \n just in case
75 # it'll give us the one element list we want
76 # if there's no \n, too
77 text = text.split('\n')
80 prelen = len(document.preamble)
82 i = find_token(document.preamble, text[0], i)
85 # we need a perfect match
88 if i >= prelen or line != document.preamble[i]:
95 document.preamble.extend(["% Added by lyx2lyx"])
96 document.preamble.extend(text)
99 # Note that text can be either a list of lines or a single line.
100 # It should really be a list.
101 def insert_to_preamble(document, text, index = 0):
102 """ Insert text to the preamble at a given line"""
104 if not type(text) is list:
105 # split on \n just in case
106 # it'll give us the one element list we want
107 # if there's no \n, too
108 text = text.split('\n')
110 text.insert(0, "% Added by lyx2lyx")
111 document.preamble[index:index] = text
114 def put_cmd_in_ert(arg):
116 arg should be a list of lines we want to wrap in ERT.
117 Returns a list of strings, with the lines so wrapped.
120 ret = ["\\begin_inset ERT", "status collapsed", "", "\\begin_layout Plain Layout", ""]
121 # It will be faster for us to work with a single string internally.
122 # That way, we only go through the unicode_reps loop once.
123 if type(arg) is list:
127 for rep in unicode_reps:
128 s = s.replace(rep[1], rep[0].replace('\\\\', '\\'))
129 s = s.replace('\\', "\\backslash\n")
130 ret += s.splitlines()
131 ret += ["\\end_layout", "", "\\end_inset"]
135 def get_ert(lines, i):
136 'Convert an ERT inset into LaTeX.'
137 if not lines[i].startswith("\\begin_inset ERT"):
139 j = find_end_of_inset(lines, i)
142 while i < j and not lines[i].startswith("status"):
148 if lines[i] == "\\begin_layout Plain Layout":
153 while i + 1 < j and lines[i+1] == "":
155 elif lines[i] == "\\end_layout":
156 while i + 1 < j and lines[i+1] == "":
158 elif lines[i] == "\\backslash":
166 def lyx2latex(document, lines):
167 'Convert some LyX stuff into corresponding LaTeX stuff, as best we can.'
174 for curline in range(len(lines)):
175 line = lines[curline]
176 if line.startswith("\\begin_inset Note Note"):
177 # We want to skip LyX notes, so remember where the inset ends
178 note_end = find_end_of_inset(lines, curline + 1)
180 elif note_end >= curline:
183 elif line.startswith("\\begin_inset ERT"):
184 # We don't want to replace things inside ERT, so figure out
185 # where the end of the inset is.
186 ert_end = find_end_of_inset(lines, curline + 1)
188 elif line.startswith("\\begin_inset Formula"):
190 elif line.startswith("\\begin_inset Quotes"):
191 # For now, we do a very basic reversion. Someone who understands
192 # quotes is welcome to fix it up.
193 qtype = line[20:].strip()
207 elif line.startswith("\\begin_inset Newline newline"):
209 elif line.startswith("\\begin_inset space"):
210 line = line[18:].strip()
211 if line.startswith("\\hspace"):
212 # Account for both \hspace and \hspace*
215 elif line == "\\space{}":
217 elif line == "\\thinspace{}":
220 # The LyX length is in line[8:], after the \length keyword
221 length = latex_length(line[8:])[1]
222 line = hspace + "{" + length + "}"
224 elif line.isspace() or \
225 line.startswith("\\begin_layout") or \
226 line.startswith("\\end_layout") or \
227 line.startswith("\\begin_inset") or \
228 line.startswith("\\end_inset") or \
229 line.startswith("\\lang") or \
230 line.strip() == "status collapsed" or \
231 line.strip() == "status open":
235 # this needs to be added to the preamble because of cases like
236 # \textmu, \textbackslash, etc.
237 add_to_preamble(document, ['% added by lyx2lyx for converted index entries',
238 '\\@ifundefined{textmu}',
239 ' {\\usepackage{textcomp}}{}'])
240 # a lossless reversion is not possible
241 # try at least to handle some common insets and settings
242 if ert_end >= curline:
243 line = line.replace(r'\backslash', '\\')
245 # No need to add "{}" after single-nonletter macros
246 line = line.replace('&', '\\&')
247 line = line.replace('#', '\\#')
248 line = line.replace('^', '\\textasciicircum{}')
249 line = line.replace('%', '\\%')
250 line = line.replace('_', '\\_')
251 line = line.replace('$', '\\$')
253 # Do the LyX text --> LaTeX conversion
254 for rep in unicode_reps:
255 line = line.replace(rep[1], rep[0] + "{}")
256 line = line.replace(r'\backslash', r'\textbackslash{}')
257 line = line.replace(r'\series bold', r'\bfseries{}').replace(r'\series default', r'\mdseries{}')
258 line = line.replace(r'\shape italic', r'\itshape{}').replace(r'\shape smallcaps', r'\scshape{}')
259 line = line.replace(r'\shape slanted', r'\slshape{}').replace(r'\shape default', r'\upshape{}')
260 line = line.replace(r'\emph on', r'\em{}').replace(r'\emph default', r'\em{}')
261 line = line.replace(r'\noun on', r'\scshape{}').replace(r'\noun default', r'\upshape{}')
262 line = line.replace(r'\bar under', r'\underbar{').replace(r'\bar default', r'}')
263 line = line.replace(r'\family sans', r'\sffamily{}').replace(r'\family default', r'\normalfont{}')
264 line = line.replace(r'\family typewriter', r'\ttfamily{}').replace(r'\family roman', r'\rmfamily{}')
265 line = line.replace(r'\InsetSpace ', r'').replace(r'\SpecialChar ', r'')
270 def latex_length(slen):
272 Convert lengths to their LaTeX representation. Returns (bool, length),
273 where the bool tells us if it was a percentage, and the length is the
274 LaTeX representation.
278 # the slen has the form
279 # ValueUnit+ValueUnit-ValueUnit or
280 # ValueUnit+-ValueUnit
281 # the + and - (glue lengths) are optional
282 # the + always precedes the -
284 # Convert relative lengths to LaTeX units
285 units = {"text%":"\\textwidth", "col%":"\\columnwidth",
286 "page%":"\\paperwidth", "line%":"\\linewidth",
287 "theight%":"\\textheight", "pheight%":"\\paperheight"}
288 for unit in list(units.keys()):
293 minus = slen.rfind("-", 1, i)
294 plus = slen.rfind("+", 0, i)
295 latex_unit = units[unit]
296 if plus == -1 and minus == -1:
298 value = str(float(value)/100)
299 end = slen[i + len(unit):]
300 slen = value + latex_unit + end
302 value = slen[plus + 1:i]
303 value = str(float(value)/100)
304 begin = slen[:plus + 1]
305 end = slen[i+len(unit):]
306 slen = begin + value + latex_unit + end
308 value = slen[minus + 1:i]
309 value = str(float(value)/100)
310 begin = slen[:minus + 1]
311 slen = begin + value + latex_unit
313 # replace + and -, but only if the - is not the first character
314 slen = slen[0] + slen[1:].replace("+", " plus ").replace("-", " minus ")
315 # handle the case where "+-1mm" was used, because LaTeX only understands
316 # "plus 1mm minus 1mm"
317 if slen.find("plus minus"):
318 lastvaluepos = slen.rfind(" ")
319 lastvalue = slen[lastvaluepos:]
320 slen = slen.replace(" ", lastvalue + " ")
321 return (percent, slen)
324 def revert_flex_inset(lines, name, LaTeXname):
325 " Convert flex insets to TeX code "
328 i = find_token(lines, '\\begin_inset Flex ' + name, i)
331 z = find_end_of_inset(lines, i)
333 document.warning("Can't find end of Flex " + name + " inset.")
336 # remove the \end_inset
337 lines[z - 2:z + 1] = put_cmd_in_ert("}")
338 # we need to reset character layouts if necessary
339 j = find_token(lines, '\\emph on', i, z)
340 k = find_token(lines, '\\noun on', i, z)
341 l = find_token(lines, '\\series', i, z)
342 m = find_token(lines, '\\family', i, z)
343 n = find_token(lines, '\\shape', i, z)
344 o = find_token(lines, '\\color', i, z)
345 p = find_token(lines, '\\size', i, z)
346 q = find_token(lines, '\\bar under', i, z)
347 r = find_token(lines, '\\uuline on', i, z)
348 s = find_token(lines, '\\uwave on', i, z)
349 t = find_token(lines, '\\strikeout on', i, z)
351 lines.insert(z - 2, "\\emph default")
353 lines.insert(z - 2, "\\noun default")
355 lines.insert(z - 2, "\\series default")
357 lines.insert(z - 2, "\\family default")
359 lines.insert(z - 2, "\\shape default")
361 lines.insert(z - 2, "\\color inherit")
363 lines.insert(z - 2, "\\size default")
365 lines.insert(z - 2, "\\bar default")
367 lines.insert(z - 2, "\\uuline default")
369 lines.insert(z - 2, "\\uwave default")
371 lines.insert(z - 2, "\\strikeout default")
372 lines[i:i + 4] = put_cmd_in_ert(LaTeXname + "{")
376 def revert_font_attrs(lines, name, LaTeXname):
377 " Reverts font changes to TeX code "
381 i = find_token(lines, name + ' on', i)
384 j = find_token(lines, name + ' default', i)
385 k = find_token(lines, name + ' on', i + 1)
386 # if there is no default set, the style ends with the layout
387 # assure hereby that we found the correct layout end
388 if j != -1 and (j < k or k == -1):
389 lines[j:j + 1] = put_cmd_in_ert("}")
391 j = find_token(lines, '\\end_layout', i)
392 lines[j:j] = put_cmd_in_ert("}")
393 lines[i:i + 1] = put_cmd_in_ert(LaTeXname + "{")
398 def revert_layout_command(lines, name, LaTeXname):
399 " Reverts a command from a layout to TeX code "
402 i = find_token(lines, '\\begin_layout ' + name, i)
406 # find the next layout
409 j = find_token(lines, '\\begin_layout', j)
411 # if nothing was found it was the last layout of the document
413 lines[l - 4:l - 4] = put_cmd_in_ert("}")
415 # exclude plain layout because this can be TeX code or another inset
416 elif lines[j] != '\\begin_layout Plain Layout':
417 lines[j - 2:j - 2] = put_cmd_in_ert("}")
421 lines[i] = '\\begin_layout Standard'
422 lines[i + 1:i + 1] = put_cmd_in_ert(LaTeXname + "{")
427 " Converts an RRGGBB-type hexadecimal string to a float in [0.0,1.0] "
434 return str(val / 256.0)
438 "'true' goes to True, case-insensitively, and we strip whitespace."
439 s = s.strip().lower()