lib/lyx2lyx/lyxconvert_218.py

   1 # This file is part of lyx2lyx
   2 # Copyright (C) 2002 Dekel Tsur <dekel@lyx.org>
   3 #
   4 # This program is free software; you can redistribute it and/or
   5 # modify it under the terms of the GNU General Public License
   6 # as published by the Free Software Foundation; either version 2
   7 # of the License, or (at your option) any later version.
   8 #
   9 # This program is distributed in the hope that it will be useful,
  10 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  11 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  12 # GNU General Public License for more details.
  13 #
  14 # You should have received a copy of the GNU General Public License
  15 # along with this program; if not, write to the Free Software
  16 # Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  17
  18
  19 import sys,string,re
  20 from parser_tools import *
  21
  22 floats = {
  23     "footnote": ["\\begin_inset Foot",
  24                  "collapsed true"],
  25     "margin":   ["\\begin_inset Marginal",
  26                  "collapsed true"],
  27     "fig":      ["\\begin_inset Float figure",
  28                  "wide false",
  29                  "collapsed false"],
  30     "tab":      ["\\begin_inset Float table",
  31                  "wide false",
  32                  "collapsed false"],
  33     "alg":      ["\\begin_inset Float algorithm",
  34                  "wide false",
  35                  "collapsed false"],
  36     "wide-fig": ["\\begin_inset Float figure",
  37                  "wide true",
  38                  "collapsed false"],
  39     "wide-tab": ["\\begin_inset Float table",
  40                  "wide true",
  41                  "collapsed false"]
  42 }
  43
  44 font_tokens = ["\\family", "\\series", "\\shape", "\\size", "\\emph",
  45                "\\bar", "\\noun", "\\color", "\\lang", "\\latex"]
  46
  47 #
  48 # Change \begin_float .. \end_float into \begin_inset Float .. \end_inset
  49 #
  50
  51 def remove_oldfloat(lines, language):
  52     i = 0
  53     while 1:
  54         i = find_token(lines, "\\begin_float", i)
  55         if i == -1:
  56             break
  57         # There are no nested floats, so finding the end of the float is simple
  58         j = find_token(lines, "\\end_float", i+1)
  59
  60         floattype = string.split(lines[i])[1]
  61         if not floats.has_key(floattype):
  62             sys.stderr.write("Error! Unknown float type "+floattype+"\n")
  63             floattype = "fig"
  64
  65         # skip \end_deeper tokens
  66         i2 = i+1
  67         while check_token(lines[i2], "\\end_deeper"):
  68             i2 = i2+1
  69         if i2 > i+1:
  70             j2 = get_next_paragraph(lines, j+1)
  71             lines[j2:j2] = ["\\end_deeper "]*(i2-(i+1))
  72
  73         new = floats[floattype]+[""]
  74         new = new+lines[i2:j]+["\\end_inset ", ""]
  75
  76         # After a float, all font attributes are reseted.
  77         # We need to output '\foo default' for every attribute foo
  78         # whose value is not default before the float.
  79         # The check here is not accurate, but it doesn't matter
  80         # as extra '\foo default' commands are ignored.
  81         # In fact, it might be safer to output '\foo default' for all
  82         # font attributes.
  83         k = get_paragraph(lines, i)
  84         flag = 0
  85         for token in font_tokens:
  86             if find_token(lines, token, k, i) != -1:
  87                 if not flag:
  88                     # This is not necessary, but we want the output to be
  89                     # as similar as posible to the lyx format
  90                     flag = 1
  91                     new.append("")
  92                 if token == "\\lang":
  93                     new.append(token+" "+language)
  94                 else:
  95                     new.append(token+" default ")
  96
  97         lines[i:j+1] = new
  98         i = i+1
  99
 100 def remove_oldminipage(lines):
 101     i = 0
 102     flag = 0
 103     while 1:
 104         i = find_token(lines, "\\pextra_type 2", i)
 105         if i == -1:
 106             break
 107         hfill = 0
 108         line = string.split(lines[i])
 109         if line[4] == "\\pextra_hfill":
 110             hfill = 1
 111             line[4:6] = []
 112         if line[4] == "\\pextra_start_minipage":
 113             # We just ignore this
 114             line[4:6] = []
 115
 116         position = line[3]
 117         width = line[5]
 118         if line[4] == "\\pextra_widthp":
 119             width = line[5]+"col%"
 120
 121
 122         start = ["\\begin_inset Minipage",
 123                  "position " + position,
 124                  "inner_position 0",
 125                  'height "0pt"',
 126                  'width "%s"' % width,
 127                  "collapsed false"
 128                  ]
 129         if flag:
 130             flag = 0
 131             if hfill:
 132                 start = ["","\hfill",""]+start
 133         else:
 134             start = ["\\layout Standard"] + start
 135
 136         j = find_token_backwards(lines,"\\layout", i-1)
 137         j0 = j
 138         mid = lines[j:i]
 139
 140         j = find_tokens(lines, ["\\layout", "\\end_float"], i+1)
 141         # j can be -1, but this is still ok
 142         mid = mid+lines[i+1:j]
 143
 144         count = 0
 145         while 1:
 146             # collect more paragraphs to the minipage
 147             count = count+1
 148             if j == -1 or not check_token(lines[j], "\\layout"):
 149                 break
 150             i = find_tokens(lines, ["\\layout", "\\pextra_type"], j+1)
 151             if i == -1 or not check_token(lines[i], "\\pextra_type"):
 152                 break
 153             line = string.split(lines[i])
 154             if line[4] == "\\pextra_hfill":
 155                 line[4:6] = []
 156             if line[4] == "\\pextra_start_minipage" and line[5] == "1":
 157                 flag = 1
 158                 break
 159             j = find_token_backwards(lines,"\\layout", i-1)
 160             mid = mid+lines[j:i]
 161             j = find_tokens(lines, ["\\layout", "\\end_float"], i+1)
 162             mid = mid+lines[i+1:j]
 163
 164         end = ["\\end_inset "]
 165
 166         lines[j0:j] = start+mid+end
 167         i = i+1
 168
 169 def is_empty(lines):
 170     return filter(is_nonempty_line, lines) == []
 171
 172 font_rexp =  re.compile(r"\\(family|series|shape|size|emph|numeric|bar|noun)")
 173 ert_rexp = re.compile(r"\\begin_inset|.*\\SpecialChar")
 174 spchar_rexp = re.compile(r"(.*)(\\SpecialChar.*)")
 175 ert_begin = ["\\begin_inset ERT",
 176              "status Collapsed",
 177              "",
 178              "\\layout Standard"]
 179
 180 def remove_oldert(lines):
 181     i = 0
 182     while 1:
 183         i = find_tokens(lines, ["\\latex latex", "\\layout LaTeX"], i)
 184         if i == -1:
 185             break
 186         j = i+1
 187         while 1:
 188             j = find_tokens(lines, ["\\latex default", "\\begin_inset", "\\layout", "\\end_float", "\\the_end"],
 189                             j)
 190             if check_token(lines[j], "\\begin_inset"):
 191                 j = find_end_of_inset(lines, j)
 192             else:
 193                 break
 194
 195         if check_token(lines[j], "\\layout"):
 196             while j-1 >= 0 and check_token(lines[j-1], "\\begin_deeper"):
 197                 j = j-1
 198
 199         # We need to remove insets, special chars & font commands from ERT text
 200         new = []
 201         new2 = []
 202         if check_token(lines[i], "\\layout LaTeX"):
 203             new = ["\layout Standard", "", ""]
 204             # We have a problem with classes in which Standard is not the default layout!
 205
 206         k = i+1
 207         while 1:
 208             k2 = find_re(lines, ert_rexp, k, j)
 209             inset = specialchar = 0
 210             if k2 == -1:
 211                 k2 = j
 212             elif check_token(lines[k2], "\\begin_inset"):
 213                 inset = 1
 214             else:
 215                 specialchar = 1
 216                 mo = spchar_rexp.match(lines[k2])
 217                 lines[k2] = mo.group(1)
 218                 specialchar_str = mo.group(2)
 219                 k2 = k2+1
 220
 221             tmp = []
 222             for line in lines[k:k2]:
 223                 if font_rexp.match(line):
 224                     if new2 == []:
 225                         # This is not necessary, but we want the output to be
 226                         # as similar as posible to the lyx format
 227                         new2 = [""]
 228                     new2.append(line)
 229                 elif not check_token(line, "\\latex"):
 230                     tmp.append(line)
 231
 232             if is_empty(tmp):
 233                 if filter(lambda x:x != "", tmp) != []:
 234                     if new == []:
 235                         # This is not necessary, but we want the output to be
 236                         # as similar as posible to the lyx format
 237                         lines[i-1] = lines[i-1]+" "
 238                     else:
 239                         new = new+[" "]
 240             else:
 241                 new = new+ert_begin+tmp+["\\end_inset ", ""]
 242
 243             if inset:
 244                 k3 = find_end_of_inset(lines, k2)
 245                 new = new+[""]+lines[k2:k3+1]+[""] # Put an empty line after \end_inset
 246                 k = k3+1
 247                 # Skip the empty line after \end_inset
 248                 if not is_nonempty_line(lines[k]):
 249                     k = k+1
 250                     new.append("")
 251             elif specialchar:
 252                 if new == []:
 253                     # This is not necessary, but we want the output to be
 254                     # as similar as posible to the lyx format
 255                     lines[i-1] = lines[i-1]+specialchar_str
 256                     new = [""]
 257                 else:
 258                     new = new+[specialchar_str, ""]
 259                 k = k2
 260             else:
 261                 break
 262
 263         new = new+new2
 264         if not check_token(lines[j], "\\latex "):
 265             new = new+[""]+[lines[j]]
 266         lines[i:j+1] = new
 267         i = i+1
 268
 269     i = 0
 270     while 1:
 271         i = find_token(lines, "\\latex ", i)
 272         if i == -1:
 273             break
 274         del lines[i]
 275
 276
 277 def remove_oldertinset(lines):
 278     i = 0
 279     while 1:
 280         i = find_token(lines, "\\begin_inset ERT", i)
 281         if i == -1:
 282             break
 283         j = find_end_of_inset(lines, i)
 284         k = find_token(lines, "\\layout", i+1)
 285         l = get_paragraph(lines, i)
 286         if lines[k] == lines[l]: # same layout
 287             k = k+1
 288         new = lines[k:j]
 289         lines[i:j+1] = new
 290         i = i+1
 291
 292 def is_ert_paragraph(lines, i):
 293     i = find_nonempty_line(lines, i+1)
 294     if not check_token(lines[i], "\\begin_inset ERT"):
 295         return 0
 296     j = find_end_of_inset(lines, i)
 297     k = find_nonempty_line(lines, j+1)
 298     return check_token(lines[k], "\\layout")
 299
 300 def combine_ert(lines):
 301     i = 0
 302     while 1:
 303         i = find_token(lines, "\\begin_inset ERT", i)
 304         if i == -1:
 305             break
 306         j = find_token_backwards(lines,"\\layout", i-1)
 307         count = 0
 308         text = []
 309         while is_ert_paragraph(lines, j):
 310
 311             count = count+1
 312             i2 = find_token(lines, "\\layout", j+1)
 313             k = find_token(lines, "\\end_inset", i2+1)
 314             text = text+lines[i2:k]
 315             j = find_token(lines, "\\layout", k+1)
 316             if j == -1:
 317                 break
 318
 319         if count >= 2:
 320             j = find_token(lines, "\\layout", i+1)
 321             lines[j:k] = text
 322
 323         i = i+1
 324
 325 oldunits = ["pt", "cm", "in", "text%", "col%"]
 326
 327 def get_length(lines, name, start, end):
 328     i = find_token(lines, name, start, end)
 329     if i == -1:
 330         return ""
 331     x = string.split(lines[i])
 332     return x[2]+oldunits[int(x[1])]
 333
 334 def write_attribute(x, token, value):
 335     if value != "":
 336         x.append("\t"+token+" "+value)
 337
 338 def remove_figinset(lines):
 339     i = 0
 340     while 1:
 341         i = find_token(lines, "\\begin_inset Figure", i)
 342         if i == -1:
 343             break
 344         j = find_end_of_inset(lines, i)
 345
 346         lyxwidth = string.split(lines[i])[3]+"pt"
 347         lyxheight = string.split(lines[i])[4]+"pt"
 348
 349         filename = get_value(lines, "file", i+1, j)
 350
 351         width = get_length(lines, "width", i+1, j)
 352         # what does width=5 mean ?
 353         height = get_length(lines, "height", i+1, j)
 354         rotateAngle = get_value(lines, "angle", i+1, j)
 355         if width == "" and height == "":
 356             size_type = "0"
 357         else:
 358             size_type = "1"
 359
 360         flags = get_value(lines, "flags", i+1, j)
 361         x = int(flags)%4
 362         if x == 1:
 363             display = "monochrome"
 364         elif x == 2:
 365             display = "gray"
 366         else:
 367             display = "color"
 368
 369         subcaptionText = get_value(lines, "subcaption", i+1, j)
 370         if subcaptionText != "":
 371             subcaptionText = '"'+subcaptionText+'"'
 372         k = find_token(lines, "subfigure", i+1,j)
 373         if k == -1:
 374             subcaption = 0
 375         else:
 376             subcaption = 1
 377
 378         new = ["\\begin_inset Graphics FormatVersion 1"]
 379         write_attribute(new, "filename", filename)
 380         write_attribute(new, "display", display)
 381         if subcaption:
 382             new.append("\tsubcaption")
 383         write_attribute(new, "subcaptionText", subcaptionText)
 384         write_attribute(new, "size_type", size_type)
 385         write_attribute(new, "width", width)
 386         write_attribute(new, "height", height)
 387         if rotateAngle != "":
 388             new.append("\trotate")
 389             write_attribute(new, "rotateAngle", rotateAngle)
 390         write_attribute(new, "rotateOrigin", "leftBaseline")
 391         write_attribute(new, "lyxsize_type", "1")
 392         write_attribute(new, "lyxwidth", lyxwidth)
 393         write_attribute(new, "lyxheight", lyxheight)
 394         new = new + ["\end_inset"]
 395         lines[i:j+1] = new
 396
 397 attr_re = re.compile(r' \w*="(false|0|)"')
 398 line_re = re.compile(r'<(features|column|row|cell)')
 399
 400 def update_tabular(lines):
 401     i = 0
 402     while 1:
 403         i = find_token(lines, '\\begin_inset  Tabular', i)
 404         if i == -1:
 405             break
 406
 407         for k in get_tabular_lines(lines, i):
 408             if check_token(lines[k], "<lyxtabular"):
 409                 lines[k] = string.replace(lines[k], 'version="2"', 'version="3"')
 410             elif check_token(lines[k], "<column"):
 411                 lines[k] = string.replace(lines[k], 'width=""', 'width="0pt"')
 412
 413             if line_re.match(lines[k]):
 414                 lines[k] = re.sub(attr_re, "", lines[k])
 415
 416         i = i+1
 417
 418 def change_preamble(lines):
 419     i = find_token(lines, "\\use_amsmath", 0)
 420     if i == -1:
 421         return
 422     lines[i+1:i+1] = ["\\use_natbib 0",
 423                       "\use_numerical_citations 0"]
 424
 425 def convert(header, body):
 426     language = get_value(header, "\\language", 0)
 427     if language == "":
 428         language = "english"
 429
 430     change_preamble(header)
 431     update_tabular(body)
 432     remove_oldminipage(body)
 433     remove_oldfloat(body, language)
 434     remove_figinset(body)
 435     remove_oldertinset(body)
 436     remove_oldert(body)
 437     combine_ert(body)
 438
 439 if __name__ == "__main__":
 440     pass