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])
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("\\noindent"):
210 line = "\\noindent " # we need the space behind the command
211 elif line.startswith("\\begin_inset space"):
212 line = line[18:].strip()
213 if line.startswith("\\hspace"):
214 # Account for both \hspace and \hspace*
217 elif line == "\\space{}":
219 elif line == "\\thinspace{}":
222 # The LyX length is in line[8:], after the \length keyword
223 length = latex_length(line[8:])[1]
224 line = hspace + "{" + length + "}"
226 elif line.isspace() or \
227 line.startswith("\\begin_layout") or \
228 line.startswith("\\end_layout") or \
229 line.startswith("\\begin_inset") or \
230 line.startswith("\\end_inset") or \
231 line.startswith("\\lang") or \
232 line.strip() == "status collapsed" or \
233 line.strip() == "status open":
237 # this needs to be added to the preamble because of cases like
238 # \textmu, \textbackslash, etc.
239 add_to_preamble(document, ['% added by lyx2lyx for converted index entries',
240 '\\@ifundefined{textmu}',
241 ' {\\usepackage{textcomp}}{}'])
242 # a lossless reversion is not possible
243 # try at least to handle some common insets and settings
244 if ert_end >= curline:
245 line = line.replace(r'\backslash', '\\')
247 # No need to add "{}" after single-nonletter macros
248 line = line.replace('&', '\\&')
249 line = line.replace('#', '\\#')
250 line = line.replace('^', '\\textasciicircum{}')
251 line = line.replace('%', '\\%')
252 line = line.replace('_', '\\_')
253 line = line.replace('$', '\\$')
255 # Do the LyX text --> LaTeX conversion
256 for rep in unicode_reps:
257 line = line.replace(rep[1], rep[0])
258 line = line.replace(r'\backslash', r'\textbackslash{}')
259 line = line.replace(r'\series bold', r'\bfseries{}').replace(r'\series default', r'\mdseries{}')
260 line = line.replace(r'\shape italic', r'\itshape{}').replace(r'\shape smallcaps', r'\scshape{}')
261 line = line.replace(r'\shape slanted', r'\slshape{}').replace(r'\shape default', r'\upshape{}')
262 line = line.replace(r'\emph on', r'\em{}').replace(r'\emph default', r'\em{}')
263 line = line.replace(r'\noun on', r'\scshape{}').replace(r'\noun default', r'\upshape{}')
264 line = line.replace(r'\bar under', r'\underbar{').replace(r'\bar default', r'}')
265 line = line.replace(r'\family sans', r'\sffamily{}').replace(r'\family default', r'\normalfont{}')
266 line = line.replace(r'\family typewriter', r'\ttfamily{}').replace(r'\family roman', r'\rmfamily{}')
267 line = line.replace(r'\InsetSpace ', r'').replace(r'\SpecialChar ', r'')
272 def latex_length(slen):
274 Convert lengths to their LaTeX representation. Returns (bool, length),
275 where the bool tells us if it was a percentage, and the length is the
276 LaTeX representation.
280 # the slen has the form
281 # ValueUnit+ValueUnit-ValueUnit or
282 # ValueUnit+-ValueUnit
283 # the + and - (glue lengths) are optional
284 # the + always precedes the -
286 # Convert relative lengths to LaTeX units
287 units = {"text%":"\\textwidth", "col%":"\\columnwidth",
288 "page%":"\\paperwidth", "line%":"\\linewidth",
289 "theight%":"\\textheight", "pheight%":"\\paperheight"}
290 for unit in list(units.keys()):
295 minus = slen.rfind("-", 1, i)
296 plus = slen.rfind("+", 0, i)
297 latex_unit = units[unit]
298 if plus == -1 and minus == -1:
300 value = str(float(value)/100)
301 end = slen[i + len(unit):]
302 slen = value + latex_unit + end
304 value = slen[plus + 1:i]
305 value = str(float(value)/100)
306 begin = slen[:plus + 1]
307 end = slen[i+len(unit):]
308 slen = begin + value + latex_unit + end
310 value = slen[minus + 1:i]
311 value = str(float(value)/100)
312 begin = slen[:minus + 1]
313 slen = begin + value + latex_unit
315 # replace + and -, but only if the - is not the first character
316 slen = slen[0] + slen[1:].replace("+", " plus ").replace("-", " minus ")
317 # handle the case where "+-1mm" was used, because LaTeX only understands
318 # "plus 1mm minus 1mm"
319 if slen.find("plus minus"):
320 lastvaluepos = slen.rfind(" ")
321 lastvalue = slen[lastvaluepos:]
322 slen = slen.replace(" ", lastvalue + " ")
323 return (percent, slen)
326 def revert_flex_inset(lines, name, LaTeXname):
327 " Convert flex insets to TeX code "
330 i = find_token(lines, '\\begin_inset Flex ' + name, i)
333 z = find_end_of_inset(lines, i)
335 document.warning("Can't find end of Flex " + name + " inset.")
338 # remove the \end_inset
339 lines[z - 2:z + 1] = put_cmd_in_ert("}")
340 # we need to reset character layouts if necessary
341 j = find_token(lines, '\\emph on', i, z)
342 k = find_token(lines, '\\noun on', i, z)
343 l = find_token(lines, '\\series', i, z)
344 m = find_token(lines, '\\family', i, z)
345 n = find_token(lines, '\\shape', i, z)
346 o = find_token(lines, '\\color', i, z)
347 p = find_token(lines, '\\size', i, z)
348 q = find_token(lines, '\\bar under', i, z)
349 r = find_token(lines, '\\uuline on', i, z)
350 s = find_token(lines, '\\uwave on', i, z)
351 t = find_token(lines, '\\strikeout on', i, z)
353 lines.insert(z - 2, "\\emph default")
355 lines.insert(z - 2, "\\noun default")
357 lines.insert(z - 2, "\\series default")
359 lines.insert(z - 2, "\\family default")
361 lines.insert(z - 2, "\\shape default")
363 lines.insert(z - 2, "\\color inherit")
365 lines.insert(z - 2, "\\size default")
367 lines.insert(z - 2, "\\bar default")
369 lines.insert(z - 2, "\\uuline default")
371 lines.insert(z - 2, "\\uwave default")
373 lines.insert(z - 2, "\\strikeout default")
374 lines[i:i + 4] = put_cmd_in_ert(LaTeXname + "{")
378 def revert_font_attrs(lines, name, LaTeXname):
379 " Reverts font changes to TeX code "
383 i = find_token(lines, name + ' on', i)
386 j = find_token(lines, name + ' default', i)
387 k = find_token(lines, name + ' on', i + 1)
388 # if there is no default set, the style ends with the layout
389 # assure hereby that we found the correct layout end
390 if j != -1 and (j < k or k == -1):
391 lines[j:j + 1] = put_cmd_in_ert("}")
393 j = find_token(lines, '\\end_layout', i)
394 lines[j:j] = put_cmd_in_ert("}")
395 lines[i:i + 1] = put_cmd_in_ert(LaTeXname + "{")
400 def revert_layout_command(lines, name, LaTeXname):
401 " Reverts a command from a layout to TeX code "
404 i = find_token(lines, '\\begin_layout ' + name, i)
408 # find the next layout
411 j = find_token(lines, '\\begin_layout', j)
413 # if nothing was found it was the last layout of the document
415 lines[l - 4:l - 4] = put_cmd_in_ert("}")
417 # exclude plain layout because this can be TeX code or another inset
418 elif lines[j] != '\\begin_layout Plain Layout':
419 lines[j - 2:j - 2] = put_cmd_in_ert("}")
423 lines[i] = '\\begin_layout Standard'
424 lines[i + 1:i + 1] = put_cmd_in_ert(LaTeXname + "{")
429 " Converts an RRGGBB-type hexadecimal string to a float in [0.0,1.0] "
436 return str(val / 256.0)
440 "'true' goes to True, case-insensitively, and we strip whitespace."
441 s = s.strip().lower()