X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=lib%2Flyx2lyx%2Flyx_1_6.py;h=3b98896ac628c12d7b6c545b392a95a0400810da;hb=d45ca67d1d3e2d4fdc344cd2a1517725bc91723a;hp=0ee1e42ba281069add549e5478f2492fc9527a95;hpb=17c3888d25a1a38da21fff955600ca208cff8867;p=lyx.git diff --git a/lib/lyx2lyx/lyx_1_6.py b/lib/lyx2lyx/lyx_1_6.py index 0ee1e42ba2..3b98896ac6 100644 --- a/lib/lyx2lyx/lyx_1_6.py +++ b/lib/lyx2lyx/lyx_1_6.py @@ -31,11 +31,30 @@ def find_end_of_inset(lines, i): " Find end of inset, where lines[i] is included." return find_end_of(lines, i, "\\begin_inset", "\\end_inset") +# WARNING! +# DO NOT do this: +# document.body[i] = wrap_insert_ert(...) +# wrap_into_ert may returns a multiline string, which should NOT appear +# in document.body. Insetad, do something like this: +# subst = wrap_inset_ert(...) +# subst = subst.split('\n') +# document.body[i:i+1] = subst +# i+= len(subst) - 1 +# where the last statement resets the counter to accord with the added +# lines. def wrap_into_ert(string, src, dst): - " Wrap a something into an ERT" + '''Within string, replace occurrences of src with dst, wrapped into ERT + E.g.: wrap_into_ert('sch\"on', "\\", "\\backslash") is: + sch\\backslash"on''' return string.replace(src, '\n\\begin_inset ERT\nstatus collapsed\n\\begin_layout Standard\n' + dst + '\n\\end_layout\n\\end_inset\n') +def put_cmd_in_ert(string): + string = string.replace('\\', "\\backslash\n") + string = "\\begin_inset ERT\nstatus collapsed\n\\begin_layout Standard\n" \ + + string + "\n\\end_layout\n\\end_inset" + return string + def add_to_preamble(document, text): """ Add text to the preamble if it is not already there. Only the first line is checked!""" @@ -45,6 +64,11 @@ def add_to_preamble(document, text): document.preamble.extend(text) +def insert_to_preamble(index, document, text): + """ Insert text to the preamble at a given line""" + + document.preamble.insert(index, text) + # Convert a LyX length into a LaTeX length def convert_len(len): units = {"text%":"\\backslash\ntextwidth", "col%":"\\backslash\ncolumnwidth", @@ -68,7 +92,15 @@ def len2value(len): # No number means 1.0 return 1.0 -#################################################################### +# Unfortunately, this doesn't really work, since Standard isn't always default. +# But it's as good as we can do right now. +def find_default_layout(document, start, end): + l = find_token(document.body, "\\begin_layout Standard", start, end) + if l == -1: + l = find_token(document.body, "\\begin_layout PlainLayout", start, end) + if l == -1: + l = find_token(document.body, "\\begin_layout Plain Layout", start, end) + return l def get_option(document, m, option, default): l = document.body[m].find(option) @@ -94,6 +126,352 @@ def set_option(document, m, option, value): document.body[m] = document.body[m][:-1] + ' ' + option + '="' + value + '">' return l + +def read_unicodesymbols(): + " Read the unicodesymbols list of unicode characters and corresponding commands." + pathname = os.path.abspath(os.path.dirname(sys.argv[0])) + fp = open(os.path.join(pathname.strip('lyx2lyx'), 'unicodesymbols')) + spec_chars = [] + # Two backslashes, followed by some non-word character, and then a character + # in brackets. The idea is to check for constructs like: \"{u}, which is how + # they are written in the unicodesymbols file; but they can also be written + # as: \"u. + r = re.compile(r'\\\\(\W)\{(\w)\}') + for line in fp.readlines(): + if line[0] != '#' and line.strip() != "": + line=line.replace(' "',' ') # remove all quotation marks with spaces before + line=line.replace('" ',' ') # remove all quotation marks with spaces after + line=line.replace(r'\"','"') # replace \" by " (for characters with diaeresis) + try: + [ucs4,command,dead] = line.split(None,2) + if command[0:1] != "\\": + continue + spec_chars.append([command, unichr(eval(ucs4))]) + except: + continue + m = r.match(command) + if m != None: + command = "\\\\" + # If the character is a double-quote, then we need to escape it, too, + # since it is done that way in the LyX file. + if m.group(1) == "\"": + command += "\\" + command += m.group(1) + m.group(2) + spec_chars.append([command, unichr(eval(ucs4))]) + fp.close() + return spec_chars + + +def extract_argument(line): + 'Extracts a LaTeX argument from the start of line. Returns (arg, rest).' + + if not line: + return (None, "") + + bracere = re.compile("(\s*)(.*)") + n = bracere.match(line) + whitespace = n.group(1) + stuff = n.group(2) + brace = stuff[:1] + if brace != "[" and brace != "{": + return (None, line) + + # find closing brace + remain = stuff[1:] + pos = 0 + num = 1 + term = "}" + if brace == "[": + term = "]" + skip = False + for c in remain: + if skip: + skip = False + elif c == "\\": + skip = True + elif c == brace: + num += 1 + elif c == term: + num -= 1 + if c == 0: + break + pos += 1 + if num != 0: + # We never found the matching brace + # So, to be on the safe side, let's just return everything + # which will then get wrapped as ERT + return (line, "") + return (line[:pos + 1], line[pos + 1:]) + + +def latex2ert(line): + '''Converts LaTeX commands into ERT. line may well be a multi-line + string when it is returned.''' + if not line: + return line + + retval = "" + ## FIXME Escaped \ ?? + # This regex looks for a LaTeX command---i.e., something of the form + # "\alPhaStuFF", or "\X", where X is any character---where the command + # may also be preceded by an additional backslash, which is how it would + # appear (e.g.) in an InsetIndex. + labelre = re.compile(r'(.*?)\\?(\\(?:[a-zA-Z]+|.))(.*)') + + m = labelre.match(line) + while m != None: + retval += m.group(1) + cmd = m.group(2) + end = m.group(3) + + while True: + (arg, rest) = extract_argument(end) + if arg == None: + break + cmd += arg + end = rest + # If we wanted to put labels into an InsetLabel, for example, then we + # would just need to test here for cmd == "label" and then take some + # appropriate action, i.e., to use arg to get the content and then + # wrap it appropriately. + cmd = put_cmd_in_ert(cmd) + retval += "\n" + cmd + "\n" + line = end + m = labelre.match(line) + retval += line + return retval + + +#Bug 5022.... +#Might should do latex2ert first, then deal with stuff that DOESN'T +#end up inside ERT. That routine could be modified so that it returned +#a list of lines, and we could then skip ERT bits and only deal with +#the other bits. +def latex2lyx(data): + '''Takes a string, possibly multi-line, and returns the result of + converting LaTeX constructs into LyX constructs. Returns a list of + lines, suitable for insertion into document.body.''' + + retval = [] + + # Convert LaTeX to Unicode + reps = read_unicodesymbols() + # Commands of this sort need to be checked to make sure they are + # followed by a non-alpha character, lest we replace too much. + hardone = re.compile(r'^\\\\[a-zA-Z]+$') + + for rep in reps: + if hardone.match(rep[0]): + pos = 0 + while True: + pos = data.find(rep[0], pos) + if pos == -1: + break + nextpos = pos + len(rep[0]) + if nextpos < len(data) and data[nextpos].isalpha(): + # not the end of that command + pos = nextpos + continue + data = data[:pos] + rep[1] + data[nextpos:] + pos = nextpos + else: + data = data.replace(rep[0], rep[1]) + + # Generic, \" -> ": + data = wrap_into_ert(data, r'\"', '"') + + # Math: + mathre = re.compile('^(.*?)(\$.*?\$)(.*)') + lines = data.split('\n') + for line in lines: + #document.warning("LINE: " + line) + #document.warning(str(i) + ":" + document.body[i]) + #document.warning("LAST: " + document.body[-1]) + g = line + m = mathre.match(g) + while m != None: + s = m.group(1) + f = m.group(2).replace('\\\\', '\\') + g = m.group(3) + if s: + # this is non-math! + s = latex2ert(s) + subst = s.split('\n') + retval += subst + retval.append("\\begin_inset Formula " + f) + retval.append("\\end_inset") + m = mathre.match(g) + # Handle whatever is left, which is just text + g = latex2ert(g) + subst = g.split('\n') + retval += subst + return retval + + +def lyx2latex(document, lines): + 'Convert some LyX stuff into corresponding LaTeX stuff, as best we can.' + # clean up multiline stuff + content = "" + ert_end = 0 + reps = read_unicodesymbols() + + for curline in range(len(lines)): + line = lines[curline] + if line.startswith("\\begin_inset ERT"): + # We don't want to replace things inside ERT, so figure out + # where the end of the inset is. + ert_end = find_end_of_inset(lines, curline + 1) + continue + elif line.startswith("\\begin_inset Formula"): + line = line[20:] + elif line.startswith("\\begin_inset Quotes"): + # For now, we do a very basic reversion. Someone who understands + # quotes is welcome to fix it up. + qtype = line[20:].strip() + # lang = qtype[0] + side = qtype[1] + dbls = qtype[2] + if side == "l": + if dbls == "d": + line = "``" + else: + line = "`" + else: + if dbls == "d": + line = "''" + else: + line = "'" + elif line.isspace() or \ + line.startswith("\\begin_layout") or \ + line.startswith("\\end_layout") or \ + line.startswith("\\begin_inset") or \ + line.startswith("\\end_inset") or \ + line.startswith("\\lang") or \ + line.strip() == "status collapsed" or \ + line.strip() == "status open": + #skip all that stuff + continue + + # this needs to be added to the preamble because of cases like + # \textmu, \textbackslash, etc. + add_to_preamble(document, ['% added by lyx2lyx for converted index entries', + '\\@ifundefined{textmu}', + ' {\\usepackage{textcomp}}{}']) + # a lossless reversion is not possible + # try at least to handle some common insets and settings + if ert_end >= curline: + line = line.replace(r'\backslash', r'\\') + else: + line = line.replace('&', '\\&{}') + line = line.replace('#', '\\#{}') + line = line.replace('^', '\\^{}') + line = line.replace('%', '\\%{}') + line = line.replace('_', '\\_{}') + line = line.replace('$', '\\${}') + + # Do the LyX text --> LaTeX conversion + for rep in reps: + line = line.replace(rep[1], rep[0] + "{}") + line = line.replace(r'\backslash', r'\textbackslash{}') + line = line.replace(r'\series bold', r'\bfseries{}').replace(r'\series default', r'\mdseries{}') + line = line.replace(r'\shape italic', r'\itshape{}').replace(r'\shape smallcaps', r'\scshape{}') + line = line.replace(r'\shape slanted', r'\slshape{}').replace(r'\shape default', r'\upshape{}') + line = line.replace(r'\emph on', r'\em{}').replace(r'\emph default', r'\em{}') + line = line.replace(r'\noun on', r'\scshape{}').replace(r'\noun default', r'\upshape{}') + line = line.replace(r'\bar under', r'\underbar{').replace(r'\bar default', r'}') + line = line.replace(r'\family sans', r'\sffamily{}').replace(r'\family default', r'\normalfont{}') + line = line.replace(r'\family typewriter', r'\ttfamily{}').replace(r'\family roman', r'\rmfamily{}') + line = line.replace(r'\InsetSpace ', r'').replace(r'\SpecialChar ', r'') + content += line + return content + + +#################################################################### + +def convert_ltcaption(document): + i = 0 + while True: + i = find_token(document.body, "\\begin_inset Tabular", i) + if i == -1: + return + j = find_end_of_inset(document.body, i + 1) + if j == -1: + document.warning("Malformed LyX document: Could not find end of tabular.") + continue + + nrows = int(document.body[i+1].split('"')[3]) + ncols = int(document.body[i+1].split('"')[5]) + + m = i + 1 + for k in range(nrows): + m = find_token(document.body, "", m + 1) + # first look for caption insets + mcap = find_token(document.body, "\\begin_inset Caption", m + 1, mend) + # then look for ERT captions + if mcap == -1: + mcap = find_token(document.body, "caption", m + 1, mend) + if mcap > -1: + mcap = find_token(document.body, "\\backslash", mcap - 1, mcap) + if mcap > -1: + caption = 'true' + if caption == 'true': + if (k == 0): + set_option(document, r, 'caption', 'true') + set_option(document, m, 'multicolumn', '1') + set_option(document, m, 'bottomline', 'false') + set_option(document, m, 'topline', 'false') + set_option(document, m, 'rightline', 'false') + set_option(document, m, 'leftline', 'false') + #j = find_end_of_inset(document.body, j + 1) + else: + set_option(document, m, 'multicolumn', '2') + m = m + 1 + m = m + 1 + + i = j + 1 + + +#FIXME Use of wrap_into_ert can confuse lyx2lyx +def revert_ltcaption(document): + i = 0 + while True: + i = find_token(document.body, "\\begin_inset Tabular", i) + if i == -1: + return + j = find_end_of_inset(document.body, i + 1) + if j == -1: + document.warning("Malformed LyX document: Could not find end of tabular.") + continue + + m = i + 1 + nrows = int(document.body[i+1].split('"')[3]) + ncols = int(document.body[i+1].split('"')[5]) + + for k in range(nrows): + m = find_token(document.body, " ": - fullcontent = wrap_into_ert(fullcontent, r'\"', '"') - #fullcontent = fullcontent.replace(r'\"', '\n\\begin_inset ERT\nstatus collapsed\n\\begin_layout standard\n"\n\\end_layout\n\\end_inset\n') - # Math: - r = re.compile('^(.*?)(\$.*?\$)(.*)') - g = fullcontent - while r.match(g): - m = r.match(g) - s = m.group(1) - f = m.group(2).replace('\\\\', '\\') - g = m.group(3) - if s: - # this is non-math! - s = wrap_into_ert(s, r'\\', '\\backslash') - s = wrap_into_ert(s, '{', '{') - s = wrap_into_ert(s, '}', '}') - document.body.insert(i + 3, s) - i += 1 - document.body.insert(i + 3, "\\begin_inset Formula " + f) - document.body.insert(i + 4, "\\end_inset") - i += 2 - # Generic, \\ -> \backslash: - g = wrap_into_ert(g, r'\\', '\\backslash{}') - g = wrap_into_ert(g, '{', '{') - g = wrap_into_ert(g, '}', '}') - document.body.insert(i + 3, g) - document.body[i + 4] = "\\end_layout" - i = i + 5 + j = find_end_of_inset(document.body, i + 2) + if j == -1: + document.warning("Unable to find end of index inset at line " + i + "!") + i += 2 + continue + m = r1.match(document.body[i + 2]) + if m == None: + document.warning("Unable to match: " + document.body[i+2]) + # this can happen with empty index insets! + linelist = [""] + else: + fullcontent = m.group(1) + linelist = latex2lyx(fullcontent) + #document.warning(fullcontent) + + linelist = ["\\begin_inset Index", "status collapsed", "\\begin_layout Standard", ""] + \ + linelist + ["\\end_layout"] + document.body[i : j] = linelist + i += len(linelist) - (j - i) def revert_latexcommand_index(document): @@ -700,34 +1094,13 @@ def revert_latexcommand_index(document): j = find_end_of_inset(document.body, i + 1) if j == -1: return - del document.body[j - 1] - del document.body[j - 2] # \end_layout - document.body[i] = "\\begin_inset CommandInset index" - document.body[i + 1] = "LatexCommand index" - # clean up multiline stuff - content = "" - for k in range(i + 3, j - 2): - line = document.body[k] - if line.startswith("\\begin_inset ERT"): - line = line[16:] - if line.startswith("\\begin_inset Formula"): - line = line[20:] - if line.startswith("\\begin_layout Standard"): - line = line[22:] - if line.startswith("\\end_layout"): - line = line[11:] - if line.startswith("\\end_inset"): - line = line[10:] - if line.startswith("status collapsed"): - line = line[16:] - line = line.replace(u'ä', r'\\\"a').replace(u'ö', r'\\\"o').replace(u'ü', r'\\\"u') - content = content + line; - document.body[i + 3] = "name " + '"' + content + '"' - for k in range(i + 4, j - 2): - del document.body[i + 4] - document.body.insert(i + 4, "") - del document.body[i + 2] # \begin_layout standard - i = i + 5 + + content = lyx2latex(document, document.body[i:j]) + # escape quotes + content = content.replace('"', r'\"') + document.body[i:j] = ["\\begin_inset CommandInset index", "LatexCommand index", + "name " + '"' + content + '"', ""] + i += 5 def revert_wraptable(document): @@ -759,6 +1132,24 @@ def revert_vietnamese(document): j = j + 1 +def convert_japanese_cjk(document): + "Set language japanese to japanese-cjk" + # Set document language from japanese-plain to japanese + i = 0 + if document.language == "japanese": + document.language = "japanese-cjk" + i = find_token(document.header, "\\language", 0) + if i != -1: + document.header[i] = "\\language japanese-cjk" + j = 0 + while True: + j = find_token(document.body, "\\lang japanese", j) + if j == -1: + return + document.body[j] = document.body[j].replace("\\lang japanese", "\\lang japanese-cjk") + j = j + 1 + + def revert_japanese(document): "Set language japanese-plain to japanese" # Set document language from japanese-plain to japanese @@ -777,6 +1168,24 @@ def revert_japanese(document): j = j + 1 +def revert_japanese_cjk(document): + "Set language japanese-cjk to japanese" + # Set document language from japanese-plain to japanese + i = 0 + if document.language == "japanese-cjk": + document.language = "japanese" + i = find_token(document.header, "\\language", 0) + if i != -1: + document.header[i] = "\\language japanese" + j = 0 + while True: + j = find_token(document.body, "\\lang japanese-cjk", j) + if j == -1: + return + document.body[j] = document.body[j].replace("\\lang japanese-cjk", "\\lang japanese") + j = j + 1 + + def revert_japanese_encoding(document): "Set input encoding form EUC-JP-plain to EUC-JP etc." # Set input encoding form EUC-JP-plain to EUC-JP etc. @@ -929,7 +1338,8 @@ def convert_ams_classes(document): return m = r.match(document.body[i]) if m == None: - document.warning("Weirdly formed \\begin_layout at line %d of body!" % i) + # This is an empty layout + # document.warning("Weirdly formed \\begin_layout at line %d of body!" % i) i += 1 continue m = m.group(1) @@ -949,6 +1359,35 @@ def revert_href(document): ["\\begin_inset CommandInset url", "LatexCommand url"] i = i + 2 +def revert_url(document): + 'Reverts Flex URL insets to old-style URL insets' + i = 0 + while True: + i = find_token(document.body, "\\begin_inset Flex URL", i) + if i == -1: + return + j = find_end_of_inset(document.body, i) + if j == -1: + document.warning("Can't find end of inset in revert_url!") + return + k = find_default_layout(document, i, j) + if k == -1: + document.warning("Can't find default layout in revert_url!") + i = j + continue + l = find_end_of(document.body, k, "\\begin_layout", "\\end_layout") + if l == -1 or l >= j: + document.warning("Can't find end of default layout in revert_url!") + i = j + continue + # OK, so the inset's data is between lines k and l. + data = " ".join(document.body[k+1:l]) + data = data.strip() + newinset = ["\\begin_inset LatexCommand url", "target \"" + data + "\"",\ + "", "\\end_inset"] + document.body[i:j+1] = newinset + i = i + len(newinset) + def convert_include(document): 'Converts include insets to new format.' @@ -982,40 +1421,50 @@ def convert_include(document): def revert_include(document): 'Reverts include insets to old format.' i = 0 + r0 = re.compile('preview.*') r1 = re.compile('LatexCommand (.+)') - r2 = re.compile('filename (.+)') - r3 = re.compile('options (.*)') + r2 = re.compile('filename "(.+)"') + r3 = re.compile('lstparams "(.*)"') while True: i = find_token(document.body, "\\begin_inset CommandInset include", i) if i == -1: return - previewline = document.body[i + 1] - m = r1.match(document.body[i + 2]) + nextline = i + 1 + m = r1.match(document.body[nextline]) if m == None: document.warning("Malformed LyX document: No LatexCommand line for `" + document.body[i] + "' on line " + str(i) + ".") i += 1 continue cmd = m.group(1) - m = r2.match(document.body[i + 3]) + nextline += 1 + if r0.match(document.body[nextline]): + previewline = document.body[nextline] + nextline += 1 + else: + previewline = "" + m = r2.match(document.body[nextline]) if m == None: document.warning("Malformed LyX document: No filename line for `" + \ document.body[i] + "' on line " + str(i) + ".") i += 2 continue fn = m.group(1) + nextline += 1 options = "" - numlines = 4 if (cmd == "lstinputlisting"): - m = r3.match(document.body[i + 4]) + m = r3.match(document.body[nextline]) if m != None: options = m.group(1) numlines = 5 + nextline += 1 newline = "\\begin_inset Include \\" + cmd + "{" + fn + "}" if options: newline += ("[" + options + "]") - insertion = [newline, previewline] - document.body[i : i + numlines] = insertion + insertion = [newline] + if previewline != "": + insertion.append(previewline) + document.body[i : nextline] = insertion i += 2 @@ -1191,14 +1640,20 @@ def convert_framed_notes(document): i = 0 while 1: i = find_tokens(document.body, ["\\begin_inset Note Framed", "\\begin_inset Note Shaded"], i) - if i == -1: return - document.body[i] = document.body[i].replace("\\begin_inset Note", "\\begin_inset Box") - document.body.insert(i + 1, 'position "t"\nhor_pos "c"\nhas_inner_box 0\ninner_pos "t"\n' \ - 'use_parbox 0\nwidth "100col%"\nspecial "none"\nheight "1in"\n' \ - 'height_special "totalheight"') - i = i + 1 + subst = [document.body[i].replace("\\begin_inset Note", "\\begin_inset Box"), + 'position "t"', + 'hor_pos "c"', + 'has_inner_box 0', + 'inner_pos "t"', + 'use_parbox 0', + 'width "100col%"', + 'special "none"', + 'height "1in"', + 'height_special "totalheight"'] + document.body[i:i+1] = subst + i = i + 9 def convert_module_names(document): @@ -1274,9 +1729,9 @@ def revert_framed_notes(document): document.warning("Malformed LyX document: Missing `status' tag in Box inset.") return status = document.body[k] - l = find_token(document.body, "\\begin_layout Standard", i + 1, j) + l = find_default_layout(document, i + 1, j) if l == -1: - document.warning("Malformed LyX document: Missing `\\begin_layout Standard' in Box inset.") + document.warning("Malformed LyX document: Missing `\\begin_layout' in Box inset.") return m = find_token(document.body, "\\end_layout", i + 1, j) if m == -1: @@ -1289,50 +1744,79 @@ def revert_framed_notes(document): del document.body[i+1:k] else: document.body[i] = document.body[i].replace("\\begin_inset Box Shaded", "\\begin_inset Box Frameless") - document.body.insert(l + 1, "\\begin_inset Note Shaded\n" + status + "\n\\begin_layout Standard\n") - document.body.insert(m + 1, "\\end_layout\n\\end_inset") + subst1 = [document.body[l], + "\\begin_inset Note Shaded", + status, + '\\begin_layout Standard'] + document.body[l:l + 1] = subst1 + subst2 = [document.body[m], "\\end_layout", "\\end_inset"] + document.body[m:m + 1] = subst2 i = i + 1 def revert_slash(document): 'Revert \\SpecialChar \\slash{} to ERT' - for i in range(len(document.body)): - document.body[i] = document.body[i].replace('\\SpecialChar \\slash{}', \ - '\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout Standard\n\n\n\\backslash\n' \ - 'slash{}\n\\end_layout\n\n\\end_inset\n\n') + r = re.compile(r'\\SpecialChar \\slash{}') + i = 0 + while i < len(document.body): + m = r.match(document.body[i]) + if m: + subst = ['\\begin_inset ERT', + 'status collapsed', '', + '\\begin_layout Standard', + '', '', '\\backslash', + 'slash{}', + '\\end_layout', '', + '\\end_inset', ''] + document.body[i: i+1] = subst + i = i + len(subst) + else: + i = i + 1 def revert_nobreakdash(document): 'Revert \\SpecialChar \\nobreakdash- to ERT' - found = 0 - for i in range(len(document.body)): + i = 0 + while i < len(document.body): line = document.body[i] r = re.compile(r'\\SpecialChar \\nobreakdash-') m = r.match(line) if m: - found = 1 - document.body[i] = document.body[i].replace('\\SpecialChar \\nobreakdash-', \ - '\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout Standard\n\n\n\\backslash\n' \ - 'nobreakdash-\n\\end_layout\n\n\\end_inset\n\n') - if not found: - return - j = find_token(document.header, "\\use_amsmath", 0) - if j == -1: - document.warning("Malformed LyX document: Missing '\\use_amsmath'.") - return - document.header[j] = "\\use_amsmath 2" + subst = ['\\begin_inset ERT', + 'status collapsed', '', + '\\begin_layout Standard', '', '', + '\\backslash', + 'nobreakdash-', + '\\end_layout', '', + '\\end_inset', ''] + document.body[i:i+1] = subst + i = i + len(subst) + j = find_token(document.header, "\\use_amsmath", 0) + if j == -1: + document.warning("Malformed LyX document: Missing '\\use_amsmath'.") + return + document.header[j] = "\\use_amsmath 2" + else: + i = i + 1 +#Returns number of lines added/removed def revert_nocite_key(body, start, end): 'key "..." -> \nocite{...}' - for i in range(start, end): - if (body[i][0:5] == 'key "'): - body[i] = body[i].replace('key "', "\\backslash\nnocite{") - body[i] = body[i].replace('"', "}") + r = re.compile(r'^key "(.*)"') + i = start + j = end + while i < j: + m = r.match(body[i]) + if m: + body[i:i+1] = ["\\backslash", "nocite{" + m.group(1) + "}"] + j += 1 # because we added a line + i += 2 # skip that line else: - body[i] = "" + del body[i] + j -= 1 # because we deleted a line + # no need to change i, since it now points to the next line + return j - end def revert_nocite(document): @@ -1342,20 +1826,29 @@ def revert_nocite(document): i = find_token(document.body, "\\begin_inset CommandInset citation", i) if i == -1: return - i = i + 1 - if (document.body[i] == "LatexCommand nocite"): - j = find_end_of_inset(document.body, i + 1) - if j == -1: - #this should not happen - document.warning("End of CommandInset citation not found in revert_nocite!") - revert_nocite_key(document.body, i + 1, len(document.body)) - return - revert_nocite_key(document.body, i + 1, j) - document.body[i-1] = "\\begin_inset ERT" - document.body[i] = "status collapsed\n\n" \ - "\\begin_layout Standard" - document.body.insert(j, "\\end_layout\n"); - i = j + if (document.body[i+1] != "LatexCommand nocite"): + # note that we already incremented i + i = i + 1 + continue + insetEnd = find_end_of_inset(document.body, i) + if insetEnd == -1: + #this should not happen + document.warning("End of CommandInset citation not found in revert_nocite!") + return + + paramLocation = i + 2 #start of the inset's parameters + addedLines = 0 + document.body[i:i+2] = \ + ["\\begin_inset ERT", "status collapsed", "", "\\begin_layout Standard"] + # that added two lines + paramLocation += 2 + insetEnd += 2 + #print insetEnd, document.body[i: insetEnd + 1] + insetEnd += revert_nocite_key(document.body, paramLocation, insetEnd) + #print insetEnd, document.body[i: insetEnd + 1] + document.body.insert(insetEnd, "\\end_layout") + document.body.insert(insetEnd + 1, "") + i = insetEnd + 1 def revert_btprintall(document): @@ -1375,14 +1868,22 @@ def revert_btprintall(document): #this should not happen document.warning("End of CommandInset bibtex not found in revert_btprintall!") j = len(document.body) + # this range isn't really right, but it should be OK, since we shouldn't + # see more than one matching line in each inset + addedlines = 0 for k in range(i, j): if (document.body[k] == 'btprint "btPrintAll"'): del document.body[k] - document.body.insert(i, "\\begin_inset ERT\n" \ - "status collapsed\n\n\\begin_layout Standard\n\n" \ - "\\backslash\nnocite{*}\n" \ - "\\end_layout\n\\end_inset\n") - i = j + subst = ["\\begin_inset ERT", + "status collapsed", "", + "\\begin_layout Standard", "", + "\\backslash", + "nocite{*}", + "\\end_layout", + "\\end_inset"] + document.body[i:i] = subst + addlines = addedlines + len(subst) - 1 + i = j + addedlines def revert_bahasam(document): @@ -1440,104 +1941,140 @@ def revert_rotfloat(document): " Revert sideways custom floats. " i = 0 while 1: - i = find_token(document.body, "\\begin_inset Float", i) + # whitespace intended (exclude \\begin_inset FloatList) + i = find_token(document.body, "\\begin_inset Float ", i) if i == -1: return line = document.body[i] r = re.compile(r'\\begin_inset Float (.*)$') m = r.match(line) + if m == None: + document.warning("Unable to match line " + str(i) + " of body!") + i += 1 + continue floattype = m.group(1) if floattype == "figure" or floattype == "table": - i = i + 1 + i += 1 continue j = find_end_of_inset(document.body, i) if j == -1: - document.warning("Malformed lyx document: Missing '\\end_inset'.") - i = i + 1 + document.warning("Malformed lyx document: Missing '\\end_inset' in revert_rotfloat.") + i += 1 continue - if get_value(document.body, 'sideways', i, j) != "false": - l = find_token(document.body, "\\begin_layout Standard", i + 1, j) - if l == -1: - document.warning("Malformed LyX document: Missing `\\begin_layout Standard' in Float inset.") - return - document.body[j] = '\\begin_layout Standard\n\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout Standard\n\n\n\\backslash\n' \ - 'end{sideways' + floattype + '}\n\\end_layout\n\n\\end_inset\n' - del document.body[i+1:l-1] - document.body[i] = '\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout Standard\n\n\n\\backslash\n' \ - 'begin{sideways' + floattype + '}\n\\end_layout\n\n\\end_inset\n\n\\end_layout\n\n' - if floattype == "algorithm": - add_to_preamble(document, - ['% Commands inserted by lyx2lyx for sideways algorithm float', - '\\usepackage{rotfloat}\n' - '\\floatstyle{ruled}\n' - '\\newfloat{algorithm}{tbp}{loa}\n' - '\\floatname{algorithm}{Algorithm}\n']) - else: - document.warning("Cannot create preamble definition for custom float" + floattype + ".") - i = i + 1 + addedLines = 0 + if get_value(document.body, 'sideways', i, j) == "false": + i += 1 continue - i = i + 1 + l = find_default_layout(document, i + 1, j) + if l == -1: + document.warning("Malformed LyX document: Missing `\\begin_layout' in Float inset.") + return + subst = ['\\begin_layout Standard', + '\\begin_inset ERT', + 'status collapsed', '', + '\\begin_layout Standard', '', '', + '\\backslash', '', + 'end{sideways' + floattype + '}', + '\\end_layout', '', '\\end_inset'] + document.body[j : j+1] = subst + addedLines = len(subst) - 1 + del document.body[i+1 : l] + addedLines -= (l-1) - (i+1) + subst = ['\\begin_inset ERT', 'status collapsed', '', + '\\begin_layout Standard', '', '', '\\backslash', + 'begin{sideways' + floattype + '}', + '\\end_layout', '', '\\end_inset', '', + '\\end_layout', ''] + document.body[i : i+1] = subst + addedLines += len(subst) - 1 + if floattype == "algorithm": + add_to_preamble(document, + ['% Commands inserted by lyx2lyx for sideways algorithm float', + '\\usepackage{rotfloat}', + '\\floatstyle{ruled}', + '\\newfloat{algorithm}{tbp}{loa}', + '\\floatname{algorithm}{Algorithm}']) + else: + document.warning("Cannot create preamble definition for custom float" + floattype + ".") + i += addedLines + 1 def revert_widesideways(document): " Revert wide sideways floats. " i = 0 while 1: - i = find_token(document.body, '\\begin_inset Float', i) + # whitespace intended (exclude \\begin_inset FloatList) + i = find_token(document.body, '\\begin_inset Float ', i) if i == -1: return line = document.body[i] r = re.compile(r'\\begin_inset Float (.*)$') m = r.match(line) + if m == None: + document.warning("Unable to match line " + str(i) + " of body!") + i += 1 + continue floattype = m.group(1) if floattype != "figure" and floattype != "table": - i = i + 1 + i += 1 continue j = find_end_of_inset(document.body, i) if j == -1: - document.warning("Malformed lyx document: Missing '\\end_inset'.") - i = i + 1 + document.warning("Malformed lyx document: Missing '\\end_inset' in revert_widesideways.") + i += 1 continue - if get_value(document.body, 'sideways', i, j) != "false": - if get_value(document.body, 'wide', i, j) != "false": - l = find_token(document.body, "\\begin_layout Standard", i + 1, j) - if l == -1: - document.warning("Malformed LyX document: Missing `\\begin_layout Standard' in Float inset.") - return - document.body[j] = '\\begin_layout Standard\n\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout Standard\n\n\n\\backslash\n' \ - 'end{sideways' + floattype + '*}\n\\end_layout\n\n\\end_inset\n' - del document.body[i+1:l-1] - document.body[i] = '\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout Standard\n\n\n\\backslash\n' \ - 'begin{sideways' + floattype + '*}\n\\end_layout\n\n\\end_inset\n\n\\end_layout\n\n' - add_to_preamble(document, - ['\\usepackage{rotfloat}\n']) - i = i + 1 - continue - i = i + 1 - - -def revert_external_embedding(document): - ' Remove embed tag from external inset ' + if get_value(document.body, 'sideways', i, j) == "false" or \ + get_value(document.body, 'wide', i, j) == "false": + i += 1 + continue + l = find_default_layout(document, i + 1, j) + if l == -1: + document.warning("Malformed LyX document: Missing `\\begin_layout' in Float inset.") + return + subst = ['\\begin_layout Standard', '\\begin_inset ERT', + 'status collapsed', '', + '\\begin_layout Standard', '', '', '\\backslash', + 'end{sideways' + floattype + '*}', + '\\end_layout', '', '\\end_inset'] + document.body[j : j+1] = subst + addedLines = len(subst) - 1 + del document.body[i+1:l-1] + addedLines -= (l-1) - (i+1) + subst = ['\\begin_inset ERT', 'status collapsed', '', + '\\begin_layout Standard', '', '', '\\backslash', + 'begin{sideways' + floattype + '*}', '\\end_layout', '', + '\\end_inset', '', '\\end_layout', ''] + document.body[i : i+1] = subst + addedLines += len(subst) - 1 + add_to_preamble(document, ['\\usepackage{rotfloat}\n']) + i += addedLines + 1 + + +def revert_inset_embedding(document, type): + ' Remove embed tag from certain type of insets' i = 0 while 1: - i = find_token(document.body, "\\begin_inset External", i) + i = find_token(document.body, "\\begin_inset %s" % type, i) if i == -1: return j = find_end_of_inset(document.body, i) if j == -1: - document.warning("Malformed lyx document: Missing '\\end_inset'.") + document.warning("Malformed lyx document: Missing '\\end_inset' in revert_inset_embedding.") i = i + 1 continue k = find_token(document.body, "\tembed", i, j) + if k == -1: + k = find_token(document.body, "embed", i, j) if k != -1: del document.body[k] i = i + 1 +def revert_external_embedding(document): + ' Remove embed tag from external inset ' + revert_inset_embedding(document, 'External') + + def convert_subfig(document): " Convert subfigures to subfloats. " i = 0 @@ -1545,54 +2082,72 @@ def convert_subfig(document): i = find_token(document.body, '\\begin_inset Graphics', i) if i == -1: return - j = find_end_of_inset(document.body, i) - if j == -1: - document.warning("Malformed lyx document: Missing '\\end_inset'.") - i = i + 1 + endInset = find_end_of_inset(document.body, i) + if endInset == -1: + document.warning("Malformed lyx document: Missing '\\end_inset' in convert_subfig.") + i += 1 continue - k = find_token(document.body, '\tsubcaption', i, j) + k = find_token(document.body, '\tsubcaption', i, endInset) if k == -1: - i = i + 1 + i = endInset + continue + l = find_token(document.body, '\tsubcaptionText', i, endInset) + if l == -1: + document.warning("Malformed lyx document: Can't find subcaptionText!") + i = endInset continue - l = find_token(document.body, '\tsubcaptionText', i, j) caption = document.body[l][16:].strip('"') - savestr = document.body[i] del document.body[l] del document.body[k] - document.body[i] = '\\begin_inset Float figure\nwide false\nsideways false\n' \ - 'status open\n\n\\begin_layout PlainLayout\n\\begin_inset Caption\n\n\\begin_layout PlainLayout\n' \ - + caption + '\n\\end_layout\n\n\\end_inset\n\n\\end_layout\n\n\\begin_layout PlainLayout\n' + savestr - savestr = document.body[j] - document.body[j] = '\n\\end_layout\n\n\\end_inset\n' + savestr + addedLines = -2 + subst = ['\\begin_inset Float figure', 'wide false', 'sideways false', + 'status open', '', '\\begin_layout Plain Layout', '\\begin_inset Caption', + '', '\\begin_layout Plain Layout'] + latex2lyx(caption) + \ + [ '\\end_layout', '', '\\end_inset', '', + '\\end_layout', '', '\\begin_layout Plain Layout'] + document.body[i : i] = subst + addedLines += len(subst) + endInset += addedLines + subst = ['', '\\end_inset', '', '\\end_layout'] + document.body[endInset : endInset] = subst + addedLines += len(subst) + i += addedLines + 1 def revert_subfig(document): " Revert subfloats. " i = 0 while 1: - i = find_token(document.body, '\\begin_inset Float', i) + # whitespace intended (exclude \\begin_inset FloatList) + i = find_tokens(document.body, ['\\begin_inset Float ', '\\begin_inset Wrap'], i) if i == -1: return - while 1: + j = 0 + addedLines = 0 + while j != -1: j = find_end_of_inset(document.body, i) if j == -1: - document.warning("Malformed lyx document: Missing '\\end_inset' (float).") - i = i + 1 - continue + document.warning("Malformed lyx document: Missing '\\end_inset' (float) at line " + str(i + len(document.header)) + ".\n\t" + document.body[i]) + # document.warning(document.body[i-1] + "\n" + document.body[i+1]) + i += 1 + continue # this will get us back to the outer loop, since j == -1 # look for embedded float (= subfloat) - k = find_token(document.body, '\\begin_inset Float', i + 1, j) + # whitespace intended (exclude \\begin_inset FloatList) + k = find_token(document.body, '\\begin_inset Float ', i + 1, j) if k == -1: break l = find_end_of_inset(document.body, k) if l == -1: document.warning("Malformed lyx document: Missing '\\end_inset' (embedded float).") - i = i + 1 - continue - m = find_token(document.body, "\\begin_layout PlainLayout", k + 1, l) + i += 1 + j == -1 + continue # escape to the outer loop + m = find_default_layout(document, k + 1, l) # caption? cap = find_token(document.body, '\\begin_inset Caption', k + 1, l) caption = '' shortcap = '' + capend = cap if cap != -1: capend = find_end_of_inset(document.body, cap) if capend == -1: @@ -1621,9 +2176,9 @@ def revert_subfig(document): if optend == -1: document.warning("Malformed lyx document: Missing '\\end_inset' (OptArg).") return - optc = find_token(document.body, "\\begin_layout PlainLayout", opt, optend) + optc = find_default_layout(document, opt, optend) if optc == -1: - document.warning("Malformed LyX document: Missing `\\begin_layout PlainLayout' in Float inset.") + document.warning("Malformed LyX document: Missing `\\begin_layout' in Float inset.") return optcend = find_end_of(document.body, optc, "\\begin_layout", "\\end_layout") for line in document.body[optc:optcend]: @@ -1641,57 +2196,75 @@ def revert_subfig(document): caption += line.strip() if len(label) > 0: caption += "\\backslash\nlabel{" + label + "}" - document.body[l] = '\\begin_layout PlainLayout\n\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout PlainLayout\n\n}\n\\end_layout\n\n\\end_inset\n\n\\end_layout\n\n\\begin_layout PlainLayout\n' - del document.body[cap:capend+1] + subst = '\\begin_layout Plain Layout\n\\begin_inset ERT\nstatus collapsed\n\n' \ + '\\begin_layout Plain Layout\n\n}\n\\end_layout\n\n\\end_inset\n\n' \ + '\\end_layout\n\n\\begin_layout Plain Layout\n' + subst = subst.split('\n') + document.body[l : l+1] = subst + addedLines = len(subst) - 1 + # this is before l and so is unchanged by the multiline insertion + if cap != capend: + del document.body[cap:capend+1] + addedLines -= (capend + 1 - cap) del document.body[k+1:m-1] + addedLines -= (m - 1 - (k + 1)) insertion = '\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout PlainLayout\n\n\\backslash\n' \ - 'subfloat' + '\\begin_layout Plain Layout\n\n\\backslash\n' \ + 'subfloat' if len(shortcap) > 0: insertion = insertion + "[" + shortcap + "]" if len(caption) > 0: insertion = insertion + "[" + caption + "]" insertion = insertion + '{%\n\\end_layout\n\n\\end_inset\n\n\\end_layout\n' - document.body[k] = insertion - add_to_preamble(document, - ['\\usepackage{subfig}\n']) - i = i + 1 + insertion = insertion.split('\n') + document.body[k : k + 1] = insertion + addedLines += len(insertion) - 1 + add_to_preamble(document, ['\\usepackage{subfig}\n']) + i += addedLines + 1 def revert_wrapplacement(document): " Revert placement options wrap floats (wrapfig). " i = 0 while True: - i = find_token(document.body, "lines", i) + i = find_token(document.body, "\\begin_inset Wrap figure", i) if i == -1: return - j = find_token(document.body, "placement", i+1) - if j != i + 1: + e = find_end_of_inset(document.body, i) + j = find_token(document.body, "placement", i + 1, e) + if j == -1: document.warning("Malformed LyX document: Couldn't find placement parameter of wrap float.") - return - document.body[j] = document.body[j].replace("placement O", "placement o") - document.body[j] = document.body[j].replace("placement I", "placement i") - document.body[j] = document.body[j].replace("placement L", "placement l") - document.body[j] = document.body[j].replace("placement R", "placement r") - i = i + 1 + i += 1 + continue + r = re.compile("placement (o|i|l|r)") + m = r.match(document.body[j]) + if m == None: + document.warning("Malformed LyX document: Placement option isn't O|I|R|L!") + document.body[j] = "placement " + m.group(1).lower() + i = j def remove_extra_embedded_files(document): " Remove \extra_embedded_files from buffer params " i = find_token(document.header, '\\extra_embedded_files', 0) if i == -1: - document.warning("Malformed lyx document: Missing '\\extra_embedded_files'.") return document.header.pop(i) def convert_spaceinset(document): " Convert '\\InsetSpace foo' to '\\begin_inset Space foo\n\\end_inset' " - for i in range(len(document.body)): - if re.search(r'\InsetSpace', document.body[i]): - document.body[i] = document.body[i].replace('\\InsetSpace', '\n\\begin_inset Space') - document.body[i] = document.body[i] + "\n\\end_inset" + i = 0 + while i < len(document.body): + m = re.match(r'(.*)\\InsetSpace (.*)', document.body[i]) + if m: + before = m.group(1) + after = m.group(2) + subst = [before, "\\begin_inset Space " + after, "\\end_inset"] + document.body[i: i+1] = subst + i = i + len(subst) + else: + i = i + 1 def revert_spaceinset(document): @@ -1716,26 +2289,53 @@ def convert_hfill(document): i = find_token(document.body, "\\hfill", i) if i == -1: return - document.body[i] = document.body[i].replace('\\hfill', '\n\\begin_inset Space \\hfill{}\n\\end_inset') + subst = document.body[i].replace('\\hfill', \ + '\n\\begin_inset Space \\hfill{}\n\\end_inset') + subst = subst.split('\n') + document.body[i : i+1] = subst + i += len(subst) def revert_hfills(document): ' Revert \\hfill commands ' - for i in range(len(document.body)): - document.body[i] = document.body[i].replace('\\InsetSpace \\hfill{}', '\\hfill') - document.body[i] = document.body[i].replace('\\InsetSpace \\dotfill{}', \ - '\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout Standard\n\n\n\\backslash\n' \ - 'dotfill{}\n\\end_layout\n\n\\end_inset\n\n') - document.body[i] = document.body[i].replace('\\InsetSpace \\hrulefill{}', \ - '\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout Standard\n\n\n\\backslash\n' \ - 'hrulefill{}\n\\end_layout\n\n\\end_inset\n\n') - + hfill = re.compile(r'\\hfill') + dotfill = re.compile(r'\\dotfill') + hrulefill = re.compile(r'\\hrulefill') + i = 0 + while True: + i = find_token(document.body, "\\InsetSpace", i) + if i == -1: + return + if hfill.search(document.body[i]): + document.body[i] = \ + document.body[i].replace('\\InsetSpace \\hfill{}', '\\hfill') + i += 1 + continue + if dotfill.search(document.body[i]): + subst = document.body[i].replace('\\InsetSpace \\dotfill{}', \ + '\\begin_inset ERT\nstatus collapsed\n\n' \ + '\\begin_layout Standard\n\n\n\\backslash\n' \ + 'dotfill{}\n\\end_layout\n\n\\end_inset\n\n') + subst = subst.split('\n') + document.body[i : i+1] = subst + i += len(subst) + continue + if hrulefill.search(document.body[i]): + subst = document.body[i].replace('\\InsetSpace \\hrulefill{}', \ + '\\begin_inset ERT\nstatus collapsed\n\n' \ + '\\begin_layout Standard\n\n\n\\backslash\n' \ + 'hrulefill{}\n\\end_layout\n\n\\end_inset\n\n') + subst = subst.split('\n') + document.body[i : i+1] = subst + i += len(subst) + continue + i += 1 def revert_hspace(document): ' Revert \\InsetSpace \\hspace{} to ERT ' i = 0 + hspace = re.compile(r'\\hspace{}') + hstar = re.compile(r'\\hspace\*{}') while True: i = find_token(document.body, "\\InsetSpace \\hspace", i) if i == -1: @@ -1745,14 +2345,28 @@ def revert_hspace(document): document.warning("Malformed lyx document: Missing '\\length' in Space inset.") return del document.body[i+1] - document.body[i] = document.body[i].replace('\\InsetSpace \\hspace*{}', \ - '\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout Standard\n\n\n\\backslash\n' \ - 'hspace*{' + length + '}\n\\end_layout\n\n\\end_inset\n\n') - document.body[i] = document.body[i].replace('\\InsetSpace \\hspace{}', \ - '\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout Standard\n\n\n\\backslash\n' \ - 'hspace{' + length + '}\n\\end_layout\n\n\\end_inset\n\n') + addedLines = -1 + if hstar.search(document.body[i]): + subst = document.body[i].replace('\\InsetSpace \\hspace*{}', \ + '\\begin_inset ERT\nstatus collapsed\n\n' \ + '\\begin_layout Standard\n\n\n\\backslash\n' \ + 'hspace*{' + length + '}\n\\end_layout\n\n\\end_inset\n\n') + subst = subst.split('\n') + document.body[i : i+1] = subst + addedLines += len(subst) - 1 + i += addedLines + 1 + continue + if hspace.search(document.body[i]): + subst = document.body[i].replace('\\InsetSpace \\hspace{}', \ + '\\begin_inset ERT\nstatus collapsed\n\n' \ + '\\begin_layout Standard\n\n\n\\backslash\n' \ + 'hspace{' + length + '}\n\\end_layout\n\n\\end_inset\n\n') + subst = subst.split('\n') + document.body[i : i+1] = subst + addedLines += len(subst) - 1 + i += addedLines + 1 + continue + i += 1 def revert_protected_hfill(document): @@ -1767,10 +2381,97 @@ def revert_protected_hfill(document): document.warning("Malformed LyX document: Could not find end of space inset.") continue del document.body[j] - document.body[i] = document.body[i].replace('\\begin_inset Space \\hspace*{\\fill}', \ - '\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout Standard\n\n\n\\backslash\n' \ - 'hspace*{\n\\backslash\nfill}\n\\end_layout\n\n\\end_inset\n\n') + subst = document.body[i].replace('\\begin_inset Space \\hspace*{\\fill}', \ + '\\begin_inset ERT\nstatus collapsed\n\n' \ + '\\begin_layout Standard\n\n\n\\backslash\n' \ + 'hspace*{\n\\backslash\nfill}\n\\end_layout\n\n\\end_inset\n\n') + subst = subst.split('\n') + document.body[i : i+1] = subst + i += len(subst) + + +def revert_leftarrowfill(document): + ' Revert \\begin_inset Space \\leftarrowfill{} to ERT ' + i = 0 + while True: + i = find_token(document.body, '\\begin_inset Space \\leftarrowfill{}', i) + if i == -1: + return + j = find_end_of_inset(document.body, i) + if j == -1: + document.warning("Malformed LyX document: Could not find end of space inset.") + continue + del document.body[j] + subst = document.body[i].replace('\\begin_inset Space \\leftarrowfill{}', \ + '\\begin_inset ERT\nstatus collapsed\n\n' \ + '\\begin_layout Standard\n\n\n\\backslash\n' \ + 'leftarrowfill{}\n\\end_layout\n\n\\end_inset\n\n') + subst = subst.split('\n') + document.body[i : i+1] = subst + i += len(subst) + + +def revert_rightarrowfill(document): + ' Revert \\begin_inset Space \\rightarrowfill{} to ERT ' + i = 0 + while True: + i = find_token(document.body, '\\begin_inset Space \\rightarrowfill{}', i) + if i == -1: + return + j = find_end_of_inset(document.body, i) + if j == -1: + document.warning("Malformed LyX document: Could not find end of space inset.") + continue + del document.body[j] + subst = document.body[i].replace('\\begin_inset Space \\rightarrowfill{}', \ + '\\begin_inset ERT\nstatus collapsed\n\n' \ + '\\begin_layout Standard\n\n\n\\backslash\n' \ + 'rightarrowfill{}\n\\end_layout\n\n\\end_inset\n\n') + subst = subst.split('\n') + document.body[i : i+1] = subst + i += len(subst) + + +def revert_upbracefill(document): + ' Revert \\begin_inset Space \\upbracefill{} to ERT ' + i = 0 + while True: + i = find_token(document.body, '\\begin_inset Space \\upbracefill{}', i) + if i == -1: + return + j = find_end_of_inset(document.body, i) + if j == -1: + document.warning("Malformed LyX document: Could not find end of space inset.") + continue + del document.body[j] + subst = document.body[i].replace('\\begin_inset Space \\upbracefill{}', \ + '\\begin_inset ERT\nstatus collapsed\n\n' \ + '\\begin_layout Standard\n\n\n\\backslash\n' \ + 'upbracefill{}\n\\end_layout\n\n\\end_inset\n\n') + subst = subst.split('\n') + document.body[i : i+1] = subst + i += len(subst) + + +def revert_downbracefill(document): + ' Revert \\begin_inset Space \\downbracefill{} to ERT ' + i = 0 + while True: + i = find_token(document.body, '\\begin_inset Space \\downbracefill{}', i) + if i == -1: + return + j = find_end_of_inset(document.body, i) + if j == -1: + document.warning("Malformed LyX document: Could not find end of space inset.") + continue + del document.body[j] + subst = document.body[i].replace('\\begin_inset Space \\downbracefill{}', \ + '\\begin_inset ERT\nstatus collapsed\n\n' \ + '\\begin_layout Standard\n\n\n\\backslash\n' \ + 'downbracefill{}\n\\end_layout\n\n\\end_inset\n\n') + subst = subst.split('\n') + document.body[i : i+1] = subst + i += len(subst) def revert_local_layout(document): @@ -1795,28 +2496,28 @@ def convert_pagebreaks(document): if i == -1: break document.body[i:i+1] = ['\\begin_inset Newpage newpage', - '\\end_inset'] + '\\end_inset'] i = 0 while True: i = find_token(document.body, '\\pagebreak', i) if i == -1: break document.body[i:i+1] = ['\\begin_inset Newpage pagebreak', - '\\end_inset'] + '\\end_inset'] i = 0 while True: i = find_token(document.body, '\\clearpage', i) if i == -1: break document.body[i:i+1] = ['\\begin_inset Newpage clearpage', - '\\end_inset'] + '\\end_inset'] i = 0 while True: i = find_token(document.body, '\\cleardoublepage', i) if i == -1: break document.body[i:i+1] = ['\\begin_inset Newpage cleardoublepage', - '\\end_inset'] + '\\end_inset'] def revert_pagebreaks(document): @@ -1845,14 +2546,14 @@ def convert_linebreaks(document): if i == -1: break document.body[i:i+1] = ['\\begin_inset Newline newline', - '\\end_inset'] + '\\end_inset'] i = 0 while True: i = find_token(document.body, '\\linebreak', i) if i == -1: break document.body[i:i+1] = ['\\begin_inset Newline linebreak', - '\\end_inset'] + '\\end_inset'] def revert_linebreaks(document): @@ -1897,7 +2598,7 @@ def revert_pdfpages(document): return j = find_end_of_inset(document.body, i) if j == -1: - document.warning("Malformed lyx document: Missing '\\end_inset'.") + document.warning("Malformed lyx document: Missing '\\end_inset' in revert_pdfpages.") i = i + 1 continue if get_value(document.body, 'template', i, j) == "PDFPages": @@ -1958,6 +2659,318 @@ def revert_pdfpages(document): i = i + 1 +def revert_mexican(document): + ' Set language Spanish(Mexico) to Spanish ' + i = 0 + if document.language == "spanish-mexico": + document.language = "spanish" + i = find_token(document.header, "\\language", 0) + if i != -1: + document.header[i] = "\\language spanish" + j = 0 + while True: + j = find_token(document.body, "\\lang spanish-mexico", j) + if j == -1: + return + document.body[j] = document.body[j].replace("\\lang spanish-mexico", "\\lang spanish") + j = j + 1 + + +def remove_embedding(document): + ' Remove embed tag from all insets ' + revert_inset_embedding(document, 'Graphics') + revert_inset_embedding(document, 'External') + revert_inset_embedding(document, 'CommandInset include') + revert_inset_embedding(document, 'CommandInset bibtex') + + +def revert_master(document): + ' Remove master param ' + i = find_token(document.header, "\\master", 0) + if i != -1: + del document.header[i] + + +def revert_graphics_group(document): + ' Revert group information from graphics insets ' + i = 0 + while 1: + i = find_token(document.body, "\\begin_inset Graphics", i) + if i == -1: + return + j = find_end_of_inset(document.body, i) + if j == -1: + document.warning("Malformed lyx document: Missing '\\end_inset' in revert_graphics_group.") + i = i + 1 + continue + k = find_token(document.body, " groupId", i, j) + if k == -1: + i = i + 1 + continue + del document.body[k] + i = i + 1 + + +def update_apa_styles(document): + ' Replace obsolete styles ' + + if document.textclass != "apa": + return + + obsoletedby = { "Acknowledgments": "Acknowledgements", + "Section*": "Section", + "Subsection*": "Subsection", + "Subsubsection*": "Subsubsection", + "Paragraph*": "Paragraph", + "Subparagraph*": "Subparagraph"} + i = 0 + while 1: + i = find_token(document.body, "\\begin_layout", i) + if i == -1: + return + + layout = document.body[i][14:] + if layout in obsoletedby: + document.body[i] = "\\begin_layout " + obsoletedby[layout] + + i += 1 + + +def convert_paper_sizes(document): + ' exchange size options legalpaper and executivepaper to correct order ' + # routine is needed to fix http://bugzilla.lyx.org/show_bug.cgi?id=4868 + i = 0 + j = 0 + i = find_token(document.header, "\\papersize executivepaper", 0) + if i != -1: + document.header[i] = "\\papersize legalpaper" + return + j = find_token(document.header, "\\papersize legalpaper", 0) + if j != -1: + document.header[j] = "\\papersize executivepaper" + + +def revert_paper_sizes(document): + ' exchange size options legalpaper and executivepaper to correct order ' + i = 0 + j = 0 + i = find_token(document.header, "\\papersize executivepaper", 0) + if i != -1: + document.header[i] = "\\papersize legalpaper" + return + j = find_token(document.header, "\\papersize legalpaper", 0) + if j != -1: + document.header[j] = "\\papersize executivepaper" + + +def convert_InsetSpace(document): + " Convert '\\begin_inset Space foo' to '\\begin_inset space foo'" + i = 0 + while True: + i = find_token(document.body, "\\begin_inset Space", i) + if i == -1: + return + document.body[i] = document.body[i].replace('\\begin_inset Space', '\\begin_inset space') + + +def revert_InsetSpace(document): + " Revert '\\begin_inset space foo' to '\\begin_inset Space foo'" + i = 0 + while True: + i = find_token(document.body, "\\begin_inset space", i) + if i == -1: + return + document.body[i] = document.body[i].replace('\\begin_inset space', '\\begin_inset Space') + + +def convert_display_enum(document): + " Convert 'display foo' to 'display false/true'" + i = 0 + while True: + i = find_token(document.body, "\tdisplay", i) + if i == -1: + return + val = get_value(document.body, 'display', i) + if val == "none": + document.body[i] = document.body[i].replace('none', 'false') + if val == "default": + document.body[i] = document.body[i].replace('default', 'true') + if val == "monochrome": + document.body[i] = document.body[i].replace('monochrome', 'true') + if val == "grayscale": + document.body[i] = document.body[i].replace('grayscale', 'true') + if val == "color": + document.body[i] = document.body[i].replace('color', 'true') + if val == "preview": + document.body[i] = document.body[i].replace('preview', 'true') + i += 1 + + +def revert_display_enum(document): + " Revert 'display false/true' to 'display none/color'" + i = 0 + while True: + i = find_token(document.body, "\tdisplay", i) + if i == -1: + return + val = get_value(document.body, 'display', i) + if val == "false": + document.body[i] = document.body[i].replace('false', 'none') + if val == "true": + document.body[i] = document.body[i].replace('true', 'default') + i += 1 + + +def remove_fontsCJK(document): + ' Remove font_cjk param ' + i = find_token(document.header, "\\font_cjk", 0) + if i != -1: + del document.header[i] + + +def convert_plain_layout(document): + " Convert 'PlainLayout' to 'Plain Layout'" + i = 0 + while True: + i = find_token(document.body, '\\begin_layout PlainLayout', i) + if i == -1: + return + document.body[i] = document.body[i].replace('\\begin_layout PlainLayout', \ + '\\begin_layout Plain Layout') + i += 1 + + +def revert_plain_layout(document): + " Convert 'PlainLayout' to 'Plain Layout'" + i = 0 + while True: + i = find_token(document.body, '\\begin_layout Plain Layout', i) + if i == -1: + return + document.body[i] = document.body[i].replace('\\begin_layout Plain Layout', \ + '\\begin_layout PlainLayout') + i += 1 + + +def revert_plainlayout(document): + " Convert 'PlainLayout' to 'Plain Layout'" + i = 0 + while True: + i = find_token(document.body, '\\begin_layout PlainLayout', i) + if i == -1: + return + # This will be incorrect for some document classes, since Standard is not always + # the default. But (a) it is probably the best we can do and (b) it will actually + # work, in fact, since an unknown layout will be converted to default. + document.body[i] = document.body[i].replace('\\begin_layout PlainLayout', \ + '\\begin_layout Standard') + i += 1 + + +def revert_polytonicgreek(document): + "Set language polytonic Greek to Greek" + i = 0 + if document.language == "polutonikogreek": + document.language = "greek" + i = find_token(document.header, "\\language", 0) + if i != -1: + document.header[i] = "\\language greek" + j = 0 + while True: + j = find_token(document.body, "\\lang polutonikogreek", j) + if j == -1: + return + document.body[j] = document.body[j].replace("\\lang polutonikogreek", "\\lang greek") + j = j + 1 + + +def revert_removed_modules(document): + i = 0 + while True: + i = find_token(document.header, "\\begin_remove_modules", i) + if i == -1: + return + j = find_end_of(document.header, i, "\\begin_remove_modules", "\\end_remove_modules") + if j == -1: + # this should not happen + break + document.header[i : j + 1] = [] + + +def add_plain_layout(document): + i = 0 + while True: + i = find_token(document.body, "\\begin_layout", i) + if i == -1: + return + if len(document.body[i].split()) == 1: + document.body[i] = "\\begin_layout Plain Layout" + i += 1 + + +def revert_tabulators(document): + "Revert tabulators to 4 spaces" + i = 0 + while True: + i = find_token(document.body, "\t", i) + if i == -1: + return + document.body[i] = document.body[i].replace("\t", " ") + i += 1 + + +def revert_tabsize(document): + "Revert the tabsize parameter of listings" + i = 0 + j = 0 + while True: + # either it is the only parameter + i = find_token(document.body, 'lstparams "tabsize=4"', i) + if i != -1: + del document.body[i] + # or the last one + j = find_token(document.body, "lstparams", j) + if j == -1: + return + pos = document.body[j].find(",tabsize=") + document.body[j] = document.body[j][:pos] + '"' + i += 1 + j += 1 + + +def revert_mongolian(document): + "Set language Mongolian to English" + i = 0 + if document.language == "mongolian": + document.language = "english" + i = find_token(document.header, "\\language", 0) + if i != -1: + document.header[i] = "\\language english" + j = 0 + while True: + j = find_token(document.body, "\\lang mongolian", j) + if j == -1: + return + document.body[j] = document.body[j].replace("\\lang mongolian", "\\lang english") + j = j + 1 + + +def revert_default_options(document): + ' Remove param use_default_options ' + i = find_token(document.header, "\\use_default_options", 0) + if i != -1: + del document.header[i] + + +def convert_default_options(document): + ' Add param use_default_options and set it to false ' + i = find_token(document.header, "\\textclass", 0) + if i == -1: + document.warning("Malformed LyX document: Missing `\\textclass'.") + return + document.header.insert(i, '\\use_default_options false') + + ## # Conversion hub # @@ -1978,7 +2991,7 @@ convert = [[277, [fix_wrong_tables]], [289, [convert_latexcommand_index]], [290, []], [291, []], - [292, []], + [292, [convert_japanese_cjk]], [293, []], [294, [convert_pdf_options]], [295, [convert_htmlurl, convert_url]], @@ -2008,25 +3021,59 @@ convert = [[277, [fix_wrong_tables]], [319, [convert_spaceinset, convert_hfill]], [320, []], [321, [convert_tablines]], - [322, []], + [322, [convert_plain_layout]], [323, [convert_pagebreaks]], [324, [convert_linebreaks]], [325, [convert_japanese_plain]], - [326, []] + [326, []], + [327, []], + [328, [remove_embedding, remove_extra_embedded_files, remove_inzip_options]], + [329, []], + [330, []], + [331, [convert_ltcaption]], + [332, []], + [333, [update_apa_styles]], + [334, [convert_paper_sizes]], + [335, [convert_InsetSpace]], + [336, []], + [337, [convert_display_enum]], + [338, []], + [339, []], + [340, [add_plain_layout]], + [341, []], + [342, []], + [343, [convert_default_options]] ] -revert = [[325, [revert_pdfpages]], +revert = [[342, [revert_default_options]], + [341, [revert_mongolian]], + [340, [revert_tabulators, revert_tabsize]], + [339, []], + [338, [revert_removed_modules]], + [337, [revert_polytonicgreek]], + [336, [revert_display_enum]], + [335, [remove_fontsCJK]], + [334, [revert_InsetSpace]], + [333, [revert_paper_sizes]], + [332, []], + [331, [revert_graphics_group]], + [330, [revert_ltcaption]], + [329, [revert_leftarrowfill, revert_rightarrowfill, revert_upbracefill, revert_downbracefill]], + [328, [revert_master]], + [327, []], + [326, [revert_mexican]], + [325, [revert_pdfpages]], [324, []], [323, [revert_linebreaks]], [322, [revert_pagebreaks]], - [321, [revert_local_layout]], + [321, [revert_local_layout, revert_plain_layout]], [320, [revert_tablines]], [319, [revert_protected_hfill]], [318, [revert_spaceinset, revert_hfills, revert_hspace]], [317, [remove_extra_embedded_files]], [316, [revert_wrapplacement]], [315, [revert_subfig]], - [314, [revert_colsep]], + [314, [revert_colsep, revert_plainlayout]], [313, []], [312, [revert_module_names]], [311, [revert_rotfloat, revert_widesideways]], @@ -2046,10 +3093,10 @@ revert = [[325, [revert_pdfpages]], [297, [revert_macro_optional_params]], [296, [revert_albanian, revert_lowersorbian, revert_uppersorbian]], [295, [revert_include]], - [294, [revert_href]], + [294, [revert_href, revert_url]], [293, [revert_pdf_options_2]], [292, [revert_inset_info]], - [291, [revert_japanese, revert_japanese_encoding]], + [291, [revert_japanese, revert_japanese_encoding, revert_japanese_cjk]], [290, [revert_vietnamese]], [289, [revert_wraptable]], [288, [revert_latexcommand_index]],