X-Git-Url: https://git.lyx.org/gitweb/?a=blobdiff_plain;f=lib%2Flyx2lyx%2Flyx_1_6.py;h=2d0964c3afeab7f80a4da853715b9c4feea507ea;hb=0f18bc753c9555c505b3508df4439bd8a5db2319;hp=bec6e7f8ed498a8df90d677aea8c8a300ca9e99b;hpb=0e8f00dbe1b21ffcf88e210a300aae3ffaf3f37d;p=lyx.git diff --git a/lib/lyx2lyx/lyx_1_6.py b/lib/lyx2lyx/lyx_1_6.py index bec6e7f8ed..2d0964c3af 100644 --- a/lib/lyx2lyx/lyx_1_6.py +++ b/lib/lyx2lyx/lyx_1_6.py @@ -14,7 +14,7 @@ # # You should have received a copy of the GNU General Public License # along with this program; if not, write to the Free Software -# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. +# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA """ Convert files to the file format generated by lyx 1.6""" @@ -22,11 +22,27 @@ import re import unicodedata import sys, os -from parser_tools import find_token, find_end_of, find_tokens, get_value, get_value_string +from parser_tools import find_token, find_end_of, find_tokens, get_value #################################################################### # Private helper functions + +def get_value_string(lines, token, start, end = 0, trim = False, default = ""): + """ get_value_string(lines, token, start[[, end], trim, default]) -> string + + Return tokens after token as string, in lines, where + token is the first element. When trim is used, the first and last character + of the string is trimmed.""" + + val = get_value(lines, token, start, end, "") + if not val: + return default + if trim: + return val[1:-1] + return val + + 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") @@ -35,7 +51,7 @@ def find_end_of_inset(lines, i): # 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: +# in document.body. Instead, do something like this: # subst = wrap_inset_ert(...) # subst = subst.split('\n') # document.body[i:i+1] = subst @@ -209,7 +225,7 @@ def extract_argument(line): return (line[:pos + 1], line[pos + 1:]) -def latex2ert(line): +def latex2ert(line, isindex): '''Converts LaTeX commands into ERT. line may well be a multi-line string when it is returned.''' if not line: @@ -246,6 +262,9 @@ def latex2ert(line): # put all remaining braces in ERT line = wrap_into_ert(line, '}', '}') line = wrap_into_ert(line, '{', '{') + if isindex: + # active character that is not available in all font encodings + line = wrap_into_ert(line, '|', '|') retval += line return retval @@ -257,10 +276,12 @@ unicode_reps = read_unicodesymbols() #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): +def latex2lyx(data, isindex): '''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.''' + lines, suitable for insertion into document.body. + The bool isindex specifies whether we are in an index macro (which + has some specific active characters that need to be ERTed).''' if not data: return [""] @@ -309,19 +330,84 @@ def latex2lyx(data): g = m.group(3) if s: # this is non-math! - s = latex2ert(s) + s = latex2ert(s, isindex) 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) + g = latex2ert(g, isindex) subst = g.split('\n') retval += subst return retval +def lyxline2latex(document, line, inert): + 'Convert some LyX stuff into corresponding LaTeX stuff line-wise, as best we can.' + if 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 + return "" + + # this needs to be added to the preamble because of cases like + # \textmu, \textbackslash, etc. + add_to_preamble(document, ['% added by lyx2lyx for converted entries', + '\\@ifundefined{textmu}', + ' {\\usepackage{textcomp}}{}']) + # a lossless reversion is not possible + # try at least to handle some common insets and settings + if inert: + line = line.replace(r'\backslash', '\\') + 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 unicode_reps: + line = line.replace(rep[1], rep[0].replace('\\\\', '\\') + "{}") + 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'') + return line + + def lyx2latex(document, lines): 'Convert some LyX stuff into corresponding LaTeX stuff, as best we can.' # clean up multiline stuff @@ -329,73 +415,15 @@ def lyx2latex(document, lines): ert_end = 0 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 unicode_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 + 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 + inert = ert_end >= curline + content += lyxline2latex(document, lines[curline], inert) + return content @@ -410,6 +438,7 @@ def convert_ltcaption(document): j = find_end_of_inset(document.body, i + 1) if j == -1: document.warning("Malformed LyX document: Could not find end of tabular.") + i += 1 continue nrows = int(document.body[i+1].split('"')[3]) @@ -460,6 +489,7 @@ def revert_ltcaption(document): j = find_end_of_inset(document.body, i + 1) if j == -1: document.warning("Malformed LyX document: Could not find end of tabular.") + i += 1 continue m = i + 1 @@ -500,6 +530,7 @@ def convert_tablines(document): j = find_end_of_inset(document.body, i + 1) if j == -1: document.warning("Malformed LyX document: Could not find end of tabular.") + i += 1 continue m = i + 1 @@ -559,9 +590,10 @@ def revert_tablines(document): i = find_token(document.body, "\\begin_inset Tabular", i) if i == -1: return - j = find_end_of_inset(document.body, i + 1) + j = find_end_of_inset(document.body, i) if j == -1: document.warning("Malformed LyX document: Could not find end of tabular.") + i += 1 continue m = i + 1 @@ -638,6 +670,7 @@ def fix_wrong_tables(document): j = find_end_of_inset(document.body, i + 1) if j == -1: document.warning("Malformed LyX document: Could not find end of tabular.") + i += 1 continue m = i + 1 @@ -760,7 +793,7 @@ def convert_flex(document): document.body[i] = document.body[i].replace('\\begin_inset CharStyle', '\\begin_inset Flex') def revert_flex(document): - "Convert Flex to CharStyle" + "Revert Flex to CharStyle" i = 0 while True: i = find_token(document.body, "\\begin_inset Flex", i) @@ -940,6 +973,8 @@ def remove_inzip_options(document): if j == -1: # should not happen document.warning("Malformed LyX document: Could not find end of graphics inset.") + i += 1 + continue # If there's a inzip param, just remove that k = find_token(document.body, "\tinzipName", i + 1, j) if k != -1: @@ -1089,7 +1124,7 @@ def convert_latexcommand_index(document): linelist = [""] else: fullcontent = m.group(1) - linelist = latex2lyx(fullcontent) + linelist = latex2lyx(fullcontent, True) #document.warning(fullcontent) linelist = ["\\begin_inset Index", "status collapsed", "\\begin_layout Standard", ""] + \ @@ -1228,6 +1263,8 @@ def revert_inset_info(document): if j == -1: # should not happen document.warning("Malformed LyX document: Could not find end of Info inset.") + i += 1 + continue type = 'unknown' arg = '' for k in range(i, j+1): @@ -1239,7 +1276,9 @@ def revert_inset_info(document): if arg[len(arg) - 1] == '"': arg = arg[:len(arg) - 1] # \" to straight quote - arg = arg.replace(r'\"','"') + arg = arg.replace(r'\"', '"') + # \ to \backslash + arg = arg.replace(r'\\', "\\backslash\n") if document.body[k].startswith("type"): type = document.body[k][4:].strip().strip('"') # I think there is a newline after \\end_inset, which should be removed. @@ -1312,13 +1351,13 @@ def convert_url(document): j = find_token(document.body, "target", i) if j == -1: document.warning("Malformed LyX document: Can't find target for url inset") - i = j + i += 1 continue target = document.body[j][8:-1] k = find_token(document.body, "\\end_inset", j) if k == -1: document.warning("Malformed LyX document: Can't find end of url inset") - i = k + i = j continue newstuff = ["\\begin_inset Flex URL", "status collapsed", "", @@ -1328,7 +1367,7 @@ def convert_url(document): "\\end_layout", ""] document.body[i:k] = newstuff - i = k + i = i + len(newstuff) def convert_ams_classes(document): tc = document.textclass @@ -1561,7 +1600,7 @@ def convert_macro_global(document): "Remove TeX code command \global when it is in front of a macro" # math macros are nowadays already defined \global, so that an additional # \global would make the document uncompilable, see - # http://bugzilla.lyx.org/show_bug.cgi?id=5371 + # http://www.lyx.org/trac/ticket/5371 # We're looking for something like this: # \begin_inset ERT # status collapsed @@ -1745,7 +1784,7 @@ def revert_module_names(document): return newmodlist = [] for mod in modlist: - if modulemap.has_key(mod): + if mod in modulemap: newmodlist.append(modulemap[mod]) else: document.warning("Can't find module %s in the module map!" % mod) @@ -1783,19 +1822,24 @@ def revert_framed_notes(document): if j == -1: # should not happen document.warning("Malformed LyX document: Could not find end of Box inset.") + i += 1 + continue k = find_token(document.body, "status", i + 1, j) if k == -1: document.warning("Malformed LyX document: Missing `status' tag in Box inset.") - return + i = j + continue status = document.body[k] l = find_default_layout(document, i + 1, j) if l == -1: document.warning("Malformed LyX document: Missing `\\begin_layout' in Box inset.") - return + i = j + continue m = find_token(document.body, "\\end_layout", i + 1, j) if m == -1: document.warning("Malformed LyX document: Missing `\\end_layout' in Box inset.") - return + i = j + continue ibox = find_token(document.body, "has_inner_box 1", i + 1, k) pbox = find_token(document.body, "use_parbox 1", i + 1, k) if ibox == -1 and pbox == -1: @@ -1858,7 +1902,8 @@ def revert_nobreakdash(document): j = find_token(document.header, "\\use_amsmath", 0) if j == -1: document.warning("Malformed LyX document: Missing '\\use_amsmath'.") - return + i += 1 + continue document.header[j] = "\\use_amsmath 2" else: i = i + 1 @@ -2032,7 +2077,8 @@ def revert_rotfloat(document): l = find_default_layout(document, i + 1, j) if l == -1: document.warning("Malformed LyX document: Missing `\\begin_layout' in Float inset.") - return + i = j + continue subst = ['\\begin_layout Standard', '\\begin_inset ERT', 'status collapsed', '', @@ -2094,7 +2140,8 @@ def revert_widesideways(document): l = find_default_layout(document, i + 1, j) if l == -1: document.warning("Malformed LyX document: Missing `\\begin_layout' in Float inset.") - return + i = j + continue subst = ['\\begin_layout Standard', '\\begin_inset ERT', 'status collapsed', '', '\\begin_layout Standard', '', '', '\\backslash', @@ -2167,7 +2214,7 @@ def convert_subfig(document): addedLines -= 1 subst = ['\\begin_inset Float figure', 'wide false', 'sideways false', 'status open', '', '\\begin_layout Plain Layout', '\\begin_inset Caption', - '', '\\begin_layout Plain Layout'] + latex2lyx(caption) + \ + '', '\\begin_layout Plain Layout'] + latex2lyx(caption, False) + \ [ '\\end_layout', '', '\\end_inset', '', '\\end_layout', '', '\\begin_layout Plain Layout'] document.body[i : i] = subst @@ -2219,7 +2266,7 @@ def revert_subfig(document): if l == -1: document.warning("Malformed lyx document: Missing '\\end_inset' (embedded float).") i += 1 - j == -1 + j = -1 continue # escape to the outer loop m = find_default_layout(document, k + 1, l) # caption? @@ -2253,7 +2300,7 @@ def revert_subfig(document): if opt != -1: optend = find_end_of_inset(document.body, opt) if optend == -1: - document.warning("Malformed lyx document: Missing '\\end_inset' (OptArg).") + document.warning("Malformed LyX document: Missing '\\end_inset' (OptArg).") return optc = find_default_layout(document, opt, optend) if optc == -1: @@ -2271,14 +2318,15 @@ def revert_subfig(document): continue elif line in document.body[opt:optend]: continue - elif not line.startswith('\\'): - caption += line.strip() + else: + inert = True + caption += lyxline2latex(document, line, inert) if len(label) > 0: - caption += "\\backslash\nlabel{" + label + "}" - subst = '\\begin_layout Plain Layout\n\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout Plain Layout\n\n}' + alignment_end + \ + caption += "\n\\backslash\nlabel{" + label + "}" + subst = '\\begin_layout PlainLayout\n\\begin_inset ERT\nstatus collapsed\n\n' \ + '\\begin_layout PlainLayout\n\n}' + alignment_end + \ '\n\\end_layout\n\n\\end_inset\n\n' \ - '\\end_layout\n\n\\begin_layout Plain Layout\n' + '\\end_layout\n\n\\begin_layout PlainLayout\n' subst = subst.split('\n') document.body[l : l+1] = subst addedLines = len(subst) - 1 @@ -2289,7 +2337,7 @@ def revert_subfig(document): del document.body[k+1:m-1] addedLines -= (m - 1 - (k + 1)) insertion = '\\begin_inset ERT\nstatus collapsed\n\n' \ - '\\begin_layout Plain Layout\n\n' + alignment_beg + '\\backslash\n' \ + '\\begin_layout PlainLayout\n\n' + alignment_beg + '\n\\backslash\n' \ 'subfloat' if len(shortcap) > 0: insertion = insertion + "[" + shortcap + "]" @@ -2299,6 +2347,7 @@ def revert_subfig(document): insertion = insertion.split('\n') document.body[k : k + 1] = insertion addedLines += len(insertion) - 1 + al = find_token(document.body, '\\align ', k - 1, j + addedLines) if al != -1: del document.body[al] addedLines -= 1 @@ -2361,6 +2410,7 @@ def revert_spaceinset(document): j = find_end_of_inset(document.body, i) if j == -1: document.warning("Malformed LyX document: Could not find end of space inset.") + i += 1 continue document.body[i] = document.body[i].replace('\\begin_inset Space', '\\InsetSpace') del document.body[j] @@ -2463,6 +2513,7 @@ def revert_protected_hfill(document): j = find_end_of_inset(document.body, i) if j == -1: document.warning("Malformed LyX document: Could not find end of space inset.") + i += 1 continue del document.body[j] subst = document.body[i].replace('\\begin_inset Space \\hspace*{\\fill}', \ @@ -2484,6 +2535,7 @@ def revert_leftarrowfill(document): j = find_end_of_inset(document.body, i) if j == -1: document.warning("Malformed LyX document: Could not find end of space inset.") + i += 1 continue del document.body[j] subst = document.body[i].replace('\\begin_inset Space \\leftarrowfill{}', \ @@ -2505,6 +2557,7 @@ def revert_rightarrowfill(document): j = find_end_of_inset(document.body, i) if j == -1: document.warning("Malformed LyX document: Could not find end of space inset.") + i += 1 continue del document.body[j] subst = document.body[i].replace('\\begin_inset Space \\rightarrowfill{}', \ @@ -2526,6 +2579,7 @@ def revert_upbracefill(document): j = find_end_of_inset(document.body, i) if j == -1: document.warning("Malformed LyX document: Could not find end of space inset.") + i += 1 continue del document.body[j] subst = document.body[i].replace('\\begin_inset Space \\upbracefill{}', \ @@ -2547,6 +2601,7 @@ def revert_downbracefill(document): j = find_end_of_inset(document.body, i) if j == -1: document.warning("Malformed LyX document: Could not find end of space inset.") + i += 1 continue del document.body[j] subst = document.body[i].replace('\\begin_inset Space \\downbracefill{}', \ @@ -2614,6 +2669,7 @@ def revert_pagebreaks(document): j = find_end_of_inset(document.body, i) if j == -1: document.warning("Malformed LyX document: Could not find end of Newpage inset.") + i += 1 continue del document.body[j] document.body[i] = document.body[i].replace('\\begin_inset Newpage newpage', '\\newpage') @@ -2650,6 +2706,7 @@ def revert_linebreaks(document): j = find_end_of_inset(document.body, i) if j == -1: document.warning("Malformed LyX document: Could not find end of Newline inset.") + i += 1 continue del document.body[j] document.body[i] = document.body[i].replace('\\begin_inset Newline newline', '\\newline') @@ -2822,7 +2879,7 @@ def update_apa_styles(document): 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 + # routine is needed to fix http://www.lyx.org/trac/ticket/4868 i = 0 j = 0 i = find_token(document.header, "\\papersize executivepaper", 0) @@ -2925,7 +2982,7 @@ def convert_plain_layout(document): def revert_plain_layout(document): - " Convert 'PlainLayout' to 'Plain Layout'" + " Revert 'Plain Layout' to 'PlainLayout'" i = 0 while True: i = find_token(document.body, '\\begin_layout Plain Layout', i) @@ -2937,7 +2994,7 @@ def revert_plain_layout(document): def revert_plainlayout(document): - " Convert 'PlainLayout' to 'Plain Layout'" + " Revert 'PlainLayout' to 'Standard'" i = 0 while True: i = find_token(document.body, '\\begin_layout PlainLayout', i)