# -*- coding: utf-8 -*-
# This file is part of lyx2lyx
-# -*- coding: utf-8 -*-
-# Copyright (C) 2010 The LyX team
+# Copyright (C) 2011 The LyX team
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
#
# 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 2.0"""
from parser_tools import find_token, find_end_of, find_tokens, \
find_token_exact, find_end_of_inset, find_end_of_layout, \
find_token_backwards, is_in_inset, get_value, get_quoted_value, \
- del_token, check_token
-
+ del_token, check_token, get_option_value
+
from lyx2lyx_tools import add_to_preamble, insert_to_preamble, \
put_cmd_in_ert, lyx2latex, latex_length, revert_flex_inset, \
revert_font_attrs, hex2ratio, str2bool
return True
-# DO NOT USE THIS ROUTINE ANY MORE. Better yet, replace the uses that
-# have been made of it with uses of put_cmd_in_ert.
-def old_put_cmd_in_ert(string):
- for rep in unicode_reps:
- string = string.replace(rep[1], rep[0].replace('\\\\', '\\'))
- string = string.replace('\\', "\\backslash\n")
- string = "\\begin_inset ERT\nstatus collapsed\n\\begin_layout Plain Layout\n" \
- + string + "\n\\end_layout\n\\end_inset"
- return string
-
-
###############################################################################
###
### Conversion and reversion routines
if p != -1:
q = document.body[fline].find("tabularvalignment")
if q != -1:
- # FIXME
- # This seems wrong: It removes everything after
- # tabularvalignment, too.
- document.body[fline] = document.body[fline][:q - 1] + '>'
+ document.body[fline] = re.sub(r' tabularvalignment=\"[a-z]+\"', "", document.body[fline])
i += 1
continue
# delete tabularvalignment
q = document.body[fline].find("tabularvalignment")
if q != -1:
- # FIXME
- # This seems wrong: It removes everything after
- # tabularvalignment, too.
- document.body[fline] = document.body[fline][:q - 1] + '>'
+ document.body[fline] = re.sub(r' tabularvalignment=\"[a-z]+\"', "", document.body[fline])
# don't add a box when centered
if tabularvalignment == 'c':
i = end
continue
- subst = ['\\end_layout', '\\end_inset']
+ subst = ['\\end_inset', '\\end_layout']
document.body[end:end] = subst # just inserts those lines
subst = ['\\begin_inset Box Frameless',
'position "' + tabularvalignment +'"',
i = end
continue
substi = ["\\begin_inset ERT", "status collapsed", "",
- "\\begin_layout Plain Layout", "", "", "\\backslash",
+ "\\begin_layout Plain Layout", "", "", "\\backslash",
cmd + "{", "\\end_layout", "", "\\end_inset"]
substj = ["\\size default", "", "\\begin_inset ERT", "status collapsed", "",
"\\begin_layout Plain Layout", "", "}", "\\end_layout", "", "\\end_inset"]
def revert_phantom(document):
revert_phantom_types(document, "Phantom", "phantom")
-
+
def revert_hphantom(document):
revert_phantom_types(document, "HPhantom", "hphantom")
# 2.) check font settings
# defaults
- roman = sans = typew = default
+ roman = sans = typew = "default"
osf = False
sf_scale = tt_scale = 100.0
-
+
i = find_token(document.header, "\\font_roman", 0)
if i == -1:
document.warning("Malformed LyX document: Missing \\font_roman.")
else:
sans = get_value(document.header, "\\font_sans", i)
document.header[i] = "\\font_sans default"
-
+
i = find_token(document.header, "\\font_typewriter", 0)
if i == -1:
document.warning("Malformed LyX document: Missing \\font_typewriter.")
else:
# we do not need this value.
document.header[i] = "\\font_sc false"
-
+
i = find_token(document.header, "\\font_sf_scale", 0)
if i == -1:
document.warning("Malformed LyX document: Missing \\font_sf_scale.")
sf += 'Scale=' + str(sf_scale / 100.0) + ','
sf += 'Mapping=tex-text]{' + sans + '}'
pretext.append(sf)
- if typewriter != "default":
+ if typew != "default":
tw = '\\setmonofont'
if tt_scale != 100.0:
tw += '[Scale=' + str(tt_scale / 100.0) + ']'
- tw += '{' + typewriter + '}'
+ tw += '{' + typew + '}'
pretext.append(tw)
if osf:
pretext.append('\\defaultfontfeatures{Numbers=OldStyle}')
- pretext.append('\usepackage{xunicode}')
- pretext.append('\usepackage{xltxtra}')
+ pretext.append('\\usepackage{xunicode}')
+ pretext.append('\\usepackage{xltxtra}')
insert_to_preamble(document, pretext)
def revert_outputformat(document):
" Remove default output format param "
-
+
if not del_token(document.header, '\\default_output_format', 0):
document.warning("Malformed LyX document: Missing \\default_output_format.")
'\\pagecolor{page_backgroundcolor}'])
-def revert_splitindex(document):
- " Reverts splitindex-aware documents "
+def add_use_indices(document):
+ " Add \\use_indices if it is missing "
i = find_token(document.header, '\\use_indices', 0)
+ if i != -1:
+ return i
+ i = find_token(document.header, '\\use_bibtopic', 0)
+ if i == -1:
+ i = find_token(document.header, '\\cite_engine', 0)
+ if i == -1:
+ i = find_token(document.header, '\\use_mathdots', 0)
+ if i == -1:
+ i = find_token(document.header, '\\use_mhchem', 0)
+ if i == -1:
+ i = find_token(document.header, '\\use_esint', 0)
+ if i == -1:
+ i = find_token(document.header, '\\use_amsmath', 0)
if i == -1:
document.warning("Malformed LyX document: Missing \\use_indices.")
+ return -1
+ document.header.insert(i + 1, '\\use_indices 0')
+ return i + 1
+
+
+def revert_splitindex(document):
+ " Reverts splitindex-aware documents "
+ i = add_use_indices(document)
+ if i == -1:
return
useindices = str2bool(get_value(document.header, "\\use_indices", i))
del document.header[i]
preamble = []
if useindices:
preamble.append("\\usepackage{splitidx})")
-
+
# deal with index declarations in the preamble
i = 0
while True:
if k == -1:
document.warning("Malformed LyX document: Missing \\end_index.")
return
- if useindices:
+ if useindices:
line = document.header[i]
l = re.compile(r'\\index (.*)$')
m = l.match(line)
del document.header[i:k + 1]
if preamble:
insert_to_preamble(document, preamble)
-
+
# deal with index insets
# these need to have the argument removed
i = 0
subst = put_cmd_in_ert("\\sindex[" + itype + "]{" + content + "}")
document.body[i:k + 1] = subst
i = i + 1
-
+
# deal with index_print insets
i = 0
while True:
def convert_splitindex(document):
" Converts index and printindex insets to splitindex-aware format "
+ add_use_indices(document)
i = 0
while True:
i = find_token(document.body, "\\begin_inset Index", i)
if document.body[i + 1].find('LatexCommand printindex') == -1:
document.warning("Malformed LyX document: Incomplete printindex inset.")
return
- subst = ["LatexCommand printindex",
+ subst = ["LatexCommand printindex",
"type \"idx\""]
document.body[i + 1:i + 2] = subst
i = i + 1
def revert_subindex(document):
" Reverts \\printsubindex CommandInset types "
- i = find_token(document.header, '\\use_indices', 0)
+ i = add_use_indices(document)
if i == -1:
- document.warning("Malformed LyX document: Missing \\use_indices.")
return
useindices = str2bool(get_value(document.header, "\\use_indices", i))
i = 0
def revert_printindexall(document):
" Reverts \\print[sub]index* CommandInset types "
- i = find_token(document.header, '\\use_indices', 0)
+ i = add_use_indices(document)
if i == -1:
- document.warning("Malformed LyX document: Missing \\use_indices.")
return
useindices = str2bool(get_value(document.header, "\\use_indices", i))
i = 0
def revert_custom_processors(document):
" Remove bibtex_command and index_command params "
-
+
if not del_token(document.header, '\\bibtex_command', 0):
document.warning("Malformed LyX document: Missing \\bibtex_command.")
-
+
if not del_token(document.header, '\\index_command', 0):
document.warning("Malformed LyX document: Missing \\index_command.")
j = find_end_of_inset(document.body, i)
l = find_token(document.body, "width", i, j)
if l == -1:
- document.warning("Can't find width option for nomencl_print!")
i = j
continue
width = get_quoted_value(document.body, "width", i, j)
if j == -1:
i += 1
continue
- # FIXME Is this correct? It wipes out everything after the
+ # FIXME Is this correct? It wipes out everything after the
# one we found.
document.body[fline] = document.body[fline][:j - 1] + '>'
- # since there could be a tabular inside this one, we
+ # since there could be a tabular inside this one, we
# cannot jump to end.
i += 1
i = 0
anum = 1
re_author = re.compile(r'(\\author) (\".*\")\s*(.*)$')
-
+
while True:
i = find_token(document.header, "\\author", i)
if i == -1:
document.header[i] = "\\author %i %s %s" % (anum, name, email)
anum += 1
i += 1
-
+
i = 0
while True:
i = find_token(document.body, "\\change_", i)
" Remove the author_id from the \\author definition "
i = 0
anum = 0
- rx = re.compile(r'(\\author)\s+(\d+)\s+(\".*\")\s*(.*)$')
+ rx = re.compile(r'(\\author)\s+(-?\d+)\s+(\".*\")\s*(.*)$')
idmap = dict()
while True:
del document.header[i]
+def convert_mhchem(document):
+ "Set mhchem to off for versions older than 1.6.x"
+ if document.initial_format < 277:
+ # LyX 1.5.x and older did never load mhchem.
+ # Therefore we must switch it off: Documents that use mhchem have
+ # a manual \usepackage anyway, and documents not using mhchem but
+ # custom macros with the same names as mhchem commands might get
+ # corrupted if mhchem is automatically loaded.
+ mhchem = 0 # off
+ else:
+ # LyX 1.6.x did always load mhchem automatically.
+ mhchem = 1 # auto
+ i = find_token(document.header, "\\use_esint", 0)
+ if i == -1:
+ # pre-1.5.x document
+ i = find_token(document.header, "\\use_amsmath", 0)
+ if i == -1:
+ document.warning("Malformed LyX document: Could not find amsmath os esint setting.")
+ return
+ document.header.insert(i + 1, "\\use_mhchem %d" % mhchem)
+
+
def revert_mhchem(document):
"Revert mhchem loading to preamble code"
if mhchem == "off":
# don't load case
- return
+ return
if mhchem == "auto":
i = 0
i += 1
if mhchem == "on":
- pre = ["\\PassOptionsToPackage{version=3}{mhchem}",
+ pre = ["\\PassOptionsToPackage{version=3}{mhchem}",
"\\usepackage{mhchem}"]
- insert_to_preamble(document, pre)
+ insert_to_preamble(document, pre)
def revert_fontenc(document):
"Verteiler": "cc",
"Gruss": "Closing"}
i = 0
- while 1:
+ while True:
i = find_token(document.body, "\\begin_layout", i)
if i == -1:
break
document.body[i] = "\\begin_layout " + obsoletedby[layout]
i += 1
-
+
document.textclass = "g-brief"
document.set_textclass()
def revert_multirow(document):
" Revert multirow cells in tables to TeX-code"
- i = 0
- begin_table = 0
- bottom_index = 0
- multirow = False
+
+ # first, let's find out if we need to do anything
# cell type 3 is multirow begin cell
- i = find_token(document.body, '<cell multirow="3"', i)
+ i = find_token(document.body, '<cell multirow="3"', 0)
if i == -1:
return
- # a multirow cell was found
- multirow = True
+
+ add_to_preamble(document, ["\\usepackage{multirow}"])
+
+ begin_table = 0
while True:
- # find begin/end of table
- begin_table = find_token(document.body, '<lyxtabular version=', begin_table)
- if begin_table == -1:
- break
- end_table = find_token(document.body, '</lyxtabular>', begin_table)
- if end_table == -1:
- document.warning("Malformed LyX document: Could not find end of table.")
- break
- # store the number of rows
- begin = document.body[begin_table].find('rows="')
- end = document.body[begin_table].find('" ', begin)
- rows = document.body[begin_table][begin + 6:end]
- # store the number of columns
- begin = document.body[begin_table].find('columns="')
- end = document.body[begin_table].find('">', begin)
- columns = document.body[begin_table][begin + 9:end]
- # find column number of the multirows
- begin_row = begin_table
- for row in range(int(rows)):
- begin_row = find_token(document.body, '<row>', begin_row)
- begin_cell = begin_row
- for column in range(int(columns)):
- begin_cell = find_token(document.body, '<cell ', begin_cell)
- i = document.body[begin_cell].find('multirow="3"')
- if i <> -1:
- # we found a multirow in column number column - 1
- # store column width
- begin = document.body[begin_table + 2 + column].find('width="')
- end = document.body[begin_table + 2 + column].find('">', begin)
- col_width = document.body[begin_table + 2 + column][begin + 7:end]
+ # find begin/end of table
+ begin_table = find_token(document.body, '<lyxtabular version=', begin_table)
+ if begin_table == -1:
+ break
+ end_table = find_end_of(document.body, begin_table, '<lyxtabular', '</lyxtabular>')
+ if end_table == -1:
+ document.warning("Malformed LyX document: Could not find end of table.")
+ begin_table += 1
+ continue
+ # does this table have multirow?
+ i = find_token(document.body, '<cell multirow="3"', begin_table, end_table)
+ if i == -1:
+ begin_table = end_table
+ continue
+
+ # store the number of rows and columns
+ numrows = get_option_value(document.body[begin_table], "rows")
+ numcols = get_option_value(document.body[begin_table], "columns")
+ try:
+ numrows = int(numrows)
+ numcols = int(numcols)
+ except:
+ document.warning("Unable to determine rows and columns!")
+ begin_table = end_table
+ continue
+
+ mrstarts = []
+ multirows = []
+ # collect info on rows and columns of this table.
+ begin_row = begin_table
+ for row in range(numrows):
+ begin_row = find_token(document.body, '<row>', begin_row, end_table)
+ if begin_row == -1:
+ document.warning("Can't find row " + str(row + 1))
+ break
+ end_row = find_end_of(document.body, begin_row, '<row>', '</row>')
+ if end_row == -1:
+ document.warning("Can't find end of row " + str(row + 1))
+ break
+ begin_cell = begin_row
+ multirows.append([])
+ for column in range(numcols):
+ begin_cell = find_token(document.body, '<cell ', begin_cell, end_row)
+ if begin_cell == -1:
+ document.warning("Can't find column " + str(column + 1) + \
+ "in row " + str(row + 1))
+ break
+ # NOTE
+ # this will fail if someone puts "</cell>" in a cell, but
+ # that seems fairly unlikely.
+ end_cell = find_end_of(document.body, begin_cell, '<cell', '</cell>')
+ if end_cell == -1:
+ document.warning("Can't find end of column " + str(column + 1) + \
+ "in row " + str(row + 1))
+ break
+ multirows[row].append([begin_cell, end_cell, 0])
+ if document.body[begin_cell].find('multirow="3"') != -1:
+ multirows[row][column][2] = 3 # begin multirow
+ mrstarts.append([row, column])
+ elif document.body[begin_cell].find('multirow="4"') != -1:
+ multirows[row][column][2] = 4 # in multirow
+ begin_cell = end_cell
+ begin_row = end_row
+ # end of table info collection
+
+ # work from the back to avoid messing up numbering
+ mrstarts.reverse()
+ for m in mrstarts:
+ row = m[0]
+ col = m[1]
+ # get column width
+ col_width = get_option_value(document.body[begin_table + 2 + col], "width")
# "0pt" means that no width is specified
- if col_width == "0pt":
+ if not col_width or col_width == "0pt":
col_width = "*"
# determine the number of cells that are part of the multirow
- begin_row2 = begin_table
- # we have the multirow begin cell as minimal span
- span = 1
- # go one row behind the row where we found the multirow begin
- row2 = -1
- while row2 < row:
- begin_row2 = find_token(document.body, '<row>', begin_row2)
- row2 += 1
- begin_row2 += 1
- # step now through all rows until we reach the last table row
- rows2 = row
- while row2 < rows:
- begin_row2 = find_token(document.body, '<row>', begin_row2)
- begin_cell2 = begin_row2
- # look now for the column-th cell in the row
- column2 = 0
- while column2 < column:
- begin_cell2 = find_token(document.body, '<cell ', begin_cell2)
- column2 += 1
- begin_cell2 += 1
- # cell type 4 is multirow-part cell
- begin_cell2 = find_token(document.body, '<cell ', begin_cell2)
- j = document.body[begin_cell2].find('multirow="4"')
- if j <> -1:
- span += 1
+ nummrs = 1
+ for r in range(row + 1, numrows):
+ if multirows[r][col][2] != 4:
+ break
+ nummrs += 1
+ # take the opportunity to revert this line
+ lineno = multirows[r][col][0]
+ document.body[lineno] = document.body[lineno].\
+ replace(' multirow="4" ', ' ').\
+ replace('valignment="middle"', 'valignment="top"').\
+ replace(' topline="true" ', ' ')
# remove bottom line of previous multirow-part cell
- if span > 1:
- document.body[bottom_index] = document.body[bottom_index].replace(' bottomline="true" ', ' ')
- # save index to be later able to remove bottom line
- bottom_index = begin_cell2
- else:
- break
- begin_row2 = find_token(document.body, '</row>', begin_row2)
- # remove the multirow tag, set the valignment to top
- # and remove the bottom line
- document.body[begin_cell] = document.body[begin_cell].replace(' multirow="3" ', ' ')
- document.body[begin_cell] = document.body[begin_cell].replace('valignment="middle"', 'valignment="top"')
- # only remove the bottom line if it is not a single cell as multirow
- if span > 1:
- document.body[begin_cell] = document.body[begin_cell].replace(' bottomline="true" ', ' ')
- # write ERT to create the multirow cell
- cend = find_token(document.body, "</cell>", begin_cell)
- if cend == -1:
- document.warning("Malformed LyX document: Could not find end of tabular cell.")
- i += 1
- continue
- blay = find_token(document.body, "\\begin_layout", begin_cell, cend)
+ lineno = multirows[r-1][col][0]
+ document.body[lineno] = document.body[lineno].replace(' bottomline="true" ', ' ')
+ # revert beginning cell
+ bcell = multirows[row][col][0]
+ ecell = multirows[row][col][1]
+ document.body[bcell] = document.body[bcell].\
+ replace(' multirow="3" ', ' ').\
+ replace('valignment="middle"', 'valignment="top"')
+ blay = find_token(document.body, "\\begin_layout", bcell, ecell)
if blay == -1:
document.warning("Can't find layout for cell!")
- i = j
continue
bend = find_end_of_layout(document.body, blay)
if bend == -1:
document.warning("Can't find end of layout for cell!")
- i = cend
continue
# do the later one first, so as not to mess up the numbering
# we are wrapping the whole cell in this ert
# so before the end of the layout...
document.body[bend:bend] = put_cmd_in_ert("}")
# ...and after the beginning
- document.body[blay + 1:blay + 1] = put_cmd_in_ert("\\multirow{" + str(span) + "}{" + col_width + "}{")
- # set index to end of current cell (that is now much bigger due to the ERT addition)
- begin_cell = find_token(document.body, '</cell>', begin_cell) + 1
- begin_row = find_token(document.body, '</row>', begin_row) + 1
-
- # revert now all multirow part cells
- while True:
- # cell type 4 is multirow part cell
- k = find_token(document.body, '<cell multirow="4"', begin_table)
- if k == -1:
- break
- # remove the multirow tag, set the valignment to top
- # and remove the top line
- document.body[k] = document.body[k].replace(' multirow="4" ', ' ')
- document.body[k] = document.body[k].replace('valignment="middle"', 'valignment="top"')
- document.body[k] = document.body[k].replace(' topline="true" ', ' ')
- k += 1
- # go to the next table
- begin_table = end_table + 1
+ document.body[blay + 1:blay + 1] = \
+ put_cmd_in_ert("\\multirow{" + str(nummrs) + "}{" + col_width + "}{")
- if multirow == True:
- add_to_preamble(document, ["\\usepackage{multirow}"])
+ begin_table = end_table
def convert_math_output(document):
else:
document.warning("Unable to match " + document.header[i])
document.header[i] = "\\html_use_mathml " + newval
-
+
def revert_inset_preview(document):
document.warning("Malformed LyX document: Could not find end of Preview inset.")
i += 1
continue
-
+
# This has several issues.
# We need to do something about the layouts inside InsetPreview.
# If we just leave the first one, then we have something like:
# \begin_layout Standard
# and we get a "no \end_layout" error. So something has to be done.
# Ideally, we would check if it is the same as the layout we are in.
- # If so, we just remove it; if not, we end the active one. But it is
+ # If so, we just remove it; if not, we end the active one. But it is
# not easy to know what layout we are in, due to depth changes, etc,
# and it is not clear to me how much work it is worth doing. In most
# cases, the layout will probably be the same.
- #
+ #
# For the same reason, we have to remove the \end_layout tag at the
# end of the last layout in the inset. Again, that will sometimes be
# wrong, but it will usually be right. To know what to do, we would
# again have to know what layout the inset is in.
-
+
blay = find_token(document.body, "\\begin_layout", i, iend)
if blay == -1:
document.warning("Can't find layout for preview inset!")
# This is where we would check what layout we are in.
# The check for Standard is definitely wrong.
- #
+ #
# lay = document.body[blay].split(None, 1)[1]
# if lay != oldlayout:
# # record a boolean to tell us what to do later....
# # better to do it later, since (a) it won't mess up
# # the numbering and (b) we only modify at the end.
-
+
# we want to delete the last \\end_layout in this inset, too.
# note that this may not be the \\end_layout that goes with blay!!
bend = find_end_of_layout(document.body, blay)
del document.body[bend]
del document.body[i:blay + 1]
# we do not need to reset i
-
+
def revert_equalspacing_xymatrix(document):
" Revert a Formula with xymatrix@! to an ERT inset "
document.warning("Malformed LyX document: Could not find end of Formula inset.")
i += 1
continue
-
+
for curline in range(i,j):
found = document.body[curline].find("\\xymatrix@!")
if found != -1:
break
-
+
if found != -1:
has_equal_spacing = True
content = [document.body[i][21:]]
has_preamble = True;
break;
i = j + 1
-
+
if has_equal_spacing and not has_preamble:
add_to_preamble(document, ['\\usepackage[all]{xy}'])
# are there any grey notes?
if find_token(document.body, "\\begin_inset Note Greyedout", 0) == -1:
- # no need to do anything else, and \renewcommand will throw
+ # no need to do anything else, and \renewcommand will throw
# an error since lyxgreyedout will not exist.
return
def revert_turkmen(document):
- "Set language Turkmen to English"
+ "Set language Turkmen to English"
- if document.language == "turkmen":
- document.language = "english"
- i = find_token(document.header, "\\language", 0)
- if i != -1:
- document.header[i] = "\\language english"
+ if document.language == "turkmen":
+ 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 turkmen", j)
- if j == -1:
- return
- document.body[j] = document.body[j].replace("\\lang turkmen", "\\lang english")
- j += 1
+ j = 0
+ while True:
+ j = find_token(document.body, "\\lang turkmen", j)
+ if j == -1:
+ return
+ document.body[j] = document.body[j].replace("\\lang turkmen", "\\lang english")
+ j += 1
def revert_fontcolor(document):
pass
i = 0
- while 1:
+ while True:
i = find_token(document.body, '\\begin_inset Info', i)
if i == -1:
return
m = l.match(line)
if m:
document.header[i] = "\\html_latex_start " + m.group(1)
-
+
i = find_token(document.header, '\\html_latex_end', 0)
if i != -1:
line = document.header[i]
m = l.match(line)
if m:
document.header[i] = "\\html_latex_end " + m.group(1)
-
+
def revert_html_quotes(document):
" Remove quotes around html_latex_start and html_latex_end "
-
+
i = find_token(document.header, '\\html_latex_start', 0)
if i != -1:
line = document.header[i]
del document.header[i]
else:
document.header[i] = "\\html_latex_start \"" + m.group(1) + "\""
-
+
i = find_token(document.header, '\\html_latex_end', 0)
if i != -1:
line = document.header[i]
def convert_optarg(document):
" Convert \\begin_inset OptArg to \\begin_inset Argument "
i = 0
- while 1:
+ while True:
i = find_token(document.body, '\\begin_inset OptArg', i)
if i == -1:
return
def revert_argument(document):
" Convert \\begin_inset Argument to \\begin_inset OptArg "
i = 0
- while 1:
+ while True:
i = find_token(document.body, '\\begin_inset Argument', i)
if i == -1:
return
def revert_makebox(document):
" Convert \\makebox to TeX code "
i = 0
- while 1:
+ while True:
i = find_token(document.body, '\\begin_inset Box', i)
if i == -1:
break
if j == -1:
i = z
continue
-
+
if not check_token(document.body[i], "\\begin_inset Box Frameless") \
or get_value(document.body, 'use_makebox', j) != 1:
del document.body[j]
def convert_use_makebox(document):
" Adds use_makebox option for boxes "
i = 0
- while 1:
+ while True:
i = find_token(document.body, '\\begin_inset Box', i)
if i == -1:
return
i = z
continue
document.body.insert(k + 1, "use_makebox 0")
- i = z + 1
+ i = blay + 1 # not z + 1 (box insets may be nested)
def revert_IEEEtran(document):
" Convert IEEEtran layouts and styles to TeX code "
+
if document.textclass != "IEEEtran":
return
+
revert_flex_inset(document.body, "IEEE membership", "\\IEEEmembership")
revert_flex_inset(document.body, "Lowercase", "\\MakeLowercase")
+
layouts = ("Special Paper Notice", "After Title Text", "Publication ID",
"Page headings", "Biography without photo")
latexcmd = {"Special Paper Notice": "\\IEEEspecialpapernotice",
"Publication ID": "\\IEEEpubid"}
obsoletedby = {"Page headings": "MarkBoth",
"Biography without photo": "BiographyNoPhoto"}
+
for layout in layouts:
i = 0
while True:
document.warning("Malformed LyX document: Can't find end of " + layout + " layout.")
i += 1
continue
- if layout in obsoletedby:
+ if layout in list(obsoletedby.keys()):
document.body[i] = "\\begin_layout " + obsoletedby[layout]
i = j
continue
document.body[k] = "LatexCommand formatted"
i = j + 1
document.header.insert(-1, "\\use_refstyle 0")
-
-
+
+
def revert_refstyle(document):
" Reverts neutral formatted refs to prettyref "
re_ref = re.compile("^reference\s+\"(\w+):(\S+)\"")
i = find_token(document.header, "\\use_refstyle", 0)
if i != -1:
document.header.pop(i)
-
+
def revert_nameref(document):
" Convert namerefs to regular references "
for cmd in cmds:
i = 0
oldcmd = "LatexCommand " + cmd
- while 1:
+ while True:
# It seems better to look for this, as most of the reference
# insets won't be ones we care about.
i = find_token(document.body, oldcmd, i)
document.body[stins:endins + 1] = newcontent
if foundone:
- add_to_preamble(document, ["\usepackage{nameref}"])
+ add_to_preamble(document, ["\\usepackage{nameref}"])
def remove_Nameref(document):
" Convert Nameref commands to nameref commands "
i = 0
- while 1:
+ while True:
# It seems better to look for this, as most of the reference
# insets won't be ones we care about.
i = find_token(document.body, "LatexCommand Nameref" , i)
break
cmdloc = i
i += 1
-
+
# Make sure it is actually in an inset!
val = is_in_inset(document.body, cmdloc, \
"\\begin_inset CommandInset ref")
def convert_flexnames(document):
"Convert \\begin_inset Flex Custom:Style to \\begin_inset Flex Style and similarly for CharStyle and Element."
-
+
i = 0
rx = re.compile(r'^\\begin_inset Flex (?:Custom|CharStyle|Element):(.+)$')
while True:
flexlist = flex_insets
else:
flexlist = flex_elements
-
+
rx = re.compile(r'^\\begin_inset Flex\s+(.+)$')
i = 0
while True:
def convert_mathdots(document):
" Load mathdots automatically "
- i = find_token(document.header, "\\use_esint" , 0)
- if i != -1:
- document.header.insert(i + 1, "\\use_mathdots 1")
+ i = find_token(document.header, "\\use_mhchem" , 0)
+ if i == -1:
+ i = find_token(document.header, "\\use_esint" , 0)
+ if i == -1:
+ document.warning("Malformed LyX document: Can't find \\use_mhchem.")
+ return;
+ j = find_token(document.preamble, "\\usepackage{mathdots}", 0)
+ if j == -1:
+ document.header.insert(i + 1, "\\use_mathdots 0")
+ else:
+ document.header.insert(i + 1, "\\use_mathdots 2")
+ del document.preamble[j]
def revert_mathdots(document):
mathdots = find_token(document.header, "\\use_mathdots" , 0)
if mathdots == -1:
- document.warning("No \\usemathdots line. Assuming auto.")
+ document.warning("No \\use_mathdots line. Assuming auto.")
else:
val = get_value(document.header, "\\use_mathdots", mathdots)
del document.header[mathdots]
return
if usedots == 2:
# force load case
- add_to_preamble(["\\usepackage{mathdots}"])
+ add_to_preamble(document, ["\\usepackage{mathdots}"])
return
-
+
# so we are in the auto case. we want to load mathdots if \iddots is used.
i = 0
while True:
def convert_rule(document):
" Convert \\lyxline to CommandInset line. "
i = 0
-
+
inset = ['\\begin_inset CommandInset line',
'LatexCommand rule',
'offset "0.5ex"',
def revert_rule(document):
" Revert line insets to Tex code "
i = 0
- while 1:
+ while True:
i = find_token(document.body, "\\begin_inset CommandInset line" , i)
if i == -1:
return
j = find_end_of_inset(document.body, i)
if j == -1:
document.warning("Malformed LyX document: Can't find end of Formula inset.")
- return
+ return
lines = "\n".join(document.body[i:j])
if lines.find("\\Diagram") == -1:
i = j
# only need to do it once!
return
+chapters = ("amsbook", "book", "docbook-book", "elsart", "extbook", "extreport",
+ "jbook", "jreport", "jsbook", "literate-book", "literate-report", "memoir",
+ "mwbk", "mwrep", "recipebook", "report", "scrbook", "scrreprt", "svmono",
+ "svmult", "tbook", "treport", "tufte-book")
def convert_bibtex_clearpage(document):
" insert a clear(double)page bibliographystyle if bibtotoc option is used "
+ if document.textclass not in chapters:
+ return
+
i = find_token(document.header, '\\papersides', 0)
sides = 0
if i == -1:
document.warning("Can't find options for bibliography inset at line " + str(j))
j = k
continue
-
+
if val.find("bibtotoc") == -1:
j = k
continue
-
+
# so we want to insert a new page right before the paragraph that
- # this bibliography thing is in.
+ # this bibliography thing is in.
lay = find_token_backwards(document.body, "\\begin_layout", j)
if lay == -1:
document.warning("Can't find layout containing bibliography inset at line " + str(j))
" http://www.mail-archive.com/lyx-devel@lists.lyx.org/msg161298.html "
if not check_passthru:
return
-
+
rx = re.compile("\\\\begin_layout \s*(\w+)")
beg = 0
for lay in ["Chunk", "Scrap"]:
document.warning("Can't find end of layout at line " + str(beg))
beg += 1
continue
+
# we are now going to replace newline insets within this layout
# by new instances of this layout. so we have repeated layouts
# instead of newlines.
+
+ # if the paragraph has any customization, however, we do not want to
+ # do the replacement.
+ if document.body[beg + 1].startswith("\\"):
+ beg = end + 1
+ continue
+
ns = beg
while True:
ns = find_token(document.body, "\\begin_inset Newline newline", ns, end)
# do ns properly, too
newlines = (ne - ns) - len(subst)
ns += newlines + 2
- end += newlines + 1
+ end += newlines + 2
+
# ok, we now want to find out if the next layout is the
# same as this one. if so, we will insert an extra copy of it
didit = False
beg = end + 1
if didit:
beg += 4 # for the extra layout
-
+
def revert_passthru(document):
" http://www.mail-archive.com/lyx-devel@lists.lyx.org/msg161298.html "
document.warning("Can't find end of layout at line " + str(beg))
beg += 1
continue
-
+
# we now want to find out if the next layout is the
# same as this one. but we will need to do this over and
# over again.
# but first let's check and make sure there is no content between the
# two layouts. i'm not sure if that can happen or not.
for l in range(end + 1, next):
- document.warning("c'" + document.body[l] + "'")
if document.body[l] != "":
document.warning("Found content between adjacent " + lay + " layouts!")
break
break
empty = True
for l in range(next + 1, nextend):
- document.warning("e'" + document.body[l] + "'")
if document.body[l] != "":
empty = False
break
" Revert multirow cells with offset in tables to TeX-code"
# this routine is the same as the revert_multirow routine except that
# it checks additionally for the offset
- i = 0
- begin_table = 0
- bottom_index = 0
- multirowOffset = False
- # cell type 3 is multirow begin cell
- i = find_token(document.body, '<cell multirow="3" mroffset=', i)
+
+ # first, let's find out if we need to do anything
+ i = find_token(document.body, '<cell multirow="3" mroffset=', 0)
if i == -1:
return
- # a multirow cell with offset was found
- multirowOffset = True
+
+ add_to_preamble(document, ["\\usepackage{multirow}"])
+
+ rgx = re.compile(r'mroffset="[^"]+?"')
+ begin_table = 0
+
while True:
- # find begin/end of table
- begin_table = find_token(document.body, '<lyxtabular version=', begin_table)
- if begin_table == -1:
- break
- end_table = find_token(document.body, '</lyxtabular>', begin_table)
- if end_table == -1:
- document.warning("Malformed LyX document: Could not find end of table.")
- break
- # store the number of rows
- begin = document.body[begin_table].find('rows="')
- end = document.body[begin_table].find('" ', begin)
- rows = document.body[begin_table][begin + 6:end]
- # store the number of columns
- begin = document.body[begin_table].find('columns="')
- end = document.body[begin_table].find('">', begin)
- columns = document.body[begin_table][begin + 9:end]
- # find column number of the multirows with offset
- begin_row = begin_table
- for row in range(int(rows)):
- begin_row = find_token(document.body, '<row>', begin_row)
- begin_cell = begin_row
- for column in range(int(columns)):
- begin_cell = find_token(document.body, '<cell ', begin_cell)
- i = document.body[begin_cell].find('multirow="3" mroffset=')
- if i <> -1:
- # we found a multirow in column number column - 1
- # store column width
- begin = document.body[begin_table + 2 + column].find('width="')
- end = document.body[begin_table + 2 + column].find('">', begin)
- col_width = document.body[begin_table + 2 + column][begin + 7:end]
+ # find begin/end of table
+ begin_table = find_token(document.body, '<lyxtabular version=', begin_table)
+ if begin_table == -1:
+ break
+ end_table = find_end_of(document.body, begin_table, '<lyxtabular', '</lyxtabular>')
+ if end_table == -1:
+ document.warning("Malformed LyX document: Could not find end of table.")
+ begin_table += 1
+ continue
+ # does this table have multirow?
+ i = find_token(document.body, '<cell multirow="3"', begin_table, end_table)
+ if i == -1:
+ begin_table = end_table
+ continue
+
+ # store the number of rows and columns
+ numrows = get_option_value(document.body[begin_table], "rows")
+ numcols = get_option_value(document.body[begin_table], "columns")
+ try:
+ numrows = int(numrows)
+ numcols = int(numcols)
+ except:
+ document.warning("Unable to determine rows and columns!")
+ begin_table = end_table
+ continue
+
+ mrstarts = []
+ multirows = []
+ # collect info on rows and columns of this table.
+ begin_row = begin_table
+ for row in range(numrows):
+ begin_row = find_token(document.body, '<row>', begin_row, end_table)
+ if begin_row == -1:
+ document.warning("Can't find row " + str(row + 1))
+ break
+ end_row = find_end_of(document.body, begin_row, '<row>', '</row>')
+ if end_row == -1:
+ document.warning("Can't find end of row " + str(row + 1))
+ break
+ begin_cell = begin_row
+ multirows.append([])
+ for column in range(numcols):
+ begin_cell = find_token(document.body, '<cell ', begin_cell, end_row)
+ if begin_cell == -1:
+ document.warning("Can't find column " + str(column + 1) + \
+ "in row " + str(row + 1))
+ break
+ # NOTE
+ # this will fail if someone puts "</cell>" in a cell, but
+ # that seems fairly unlikely.
+ end_cell = find_end_of(document.body, begin_cell, '<cell', '</cell>')
+ if end_cell == -1:
+ document.warning("Can't find end of column " + str(column + 1) + \
+ "in row " + str(row + 1))
+ break
+ multirows[row].append([begin_cell, end_cell, 0])
+ if document.body[begin_cell].find('multirow="3" mroffset=') != -1:
+ multirows[row][column][2] = 3 # begin multirow
+ mrstarts.append([row, column])
+ elif document.body[begin_cell].find('multirow="4"') != -1:
+ multirows[row][column][2] = 4 # in multirow
+ begin_cell = end_cell
+ begin_row = end_row
+ # end of table info collection
+
+ # work from the back to avoid messing up numbering
+ mrstarts.reverse()
+ for m in mrstarts:
+ row = m[0]
+ col = m[1]
+ # get column width
+ col_width = get_option_value(document.body[begin_table + 2 + col], "width")
# "0pt" means that no width is specified
- if col_width == "0pt":
+ if not col_width or col_width == "0pt":
col_width = "*"
# determine the number of cells that are part of the multirow
- begin_row2 = begin_table
- # we have the multirow begin cell as minimal span
- span = 1
- # go one row behind the row where we found the multirow begin
- row2 = -1
- while row2 < row:
- begin_row2 = find_token(document.body, '<row>', begin_row2)
- row2 += 1
- begin_row2 += 1
- # step now through all rows until we reach the last table row
- rows2 = row
- while row2 < rows:
- begin_row2 = find_token(document.body, '<row>', begin_row2)
- begin_cell2 = begin_row2
- # look now for the column-th cell in the row
- column2 = 0
- while column2 < column:
- begin_cell2 = find_token(document.body, '<cell ', begin_cell2)
- column2 += 1
- begin_cell2 += 1
- # cell type 4 is multirow-part cell
- begin_cell2 = find_token(document.body, '<cell ', begin_cell2)
- j = document.body[begin_cell2].find('multirow="4"')
- if j <> -1:
- span += 1
+ nummrs = 1
+ for r in range(row + 1, numrows):
+ if multirows[r][col][2] != 4:
+ break
+ nummrs += 1
+ # take the opportunity to revert this line
+ lineno = multirows[r][col][0]
+ document.body[lineno] = document.body[lineno].\
+ replace(' multirow="4" ', ' ').\
+ replace('valignment="middle"', 'valignment="top"').\
+ replace(' topline="true" ', ' ')
# remove bottom line of previous multirow-part cell
- if span > 1:
- document.body[bottom_index] = document.body[bottom_index].replace(' bottomline="true" ', ' ')
- # save index to be later able to remove bottom line
- bottom_index = begin_cell2
- # remove the multirow tag, set the valignment to top
- # and remove the top line
- document.body[begin_cell2] = document.body[begin_cell2].replace(' multirow="4" ', ' ')
- document.body[begin_cell2] = document.body[begin_cell2].replace('valignment="middle"', 'valignment="top"')
- document.body[begin_cell2] = document.body[begin_cell2].replace(' topline="true" ', ' ')
- else:
- break
- begin_row2 = find_token(document.body, '</row>', begin_row2)
- # remove the multirow tag, set the valignment to top,
- # remove the the bottom line and offset
- document.body[begin_cell] = document.body[begin_cell].replace(' multirow="3" ', ' ')
- document.body[begin_cell] = document.body[begin_cell].replace('valignment="middle"', 'valignment="top"')
- # only remove the bottom line if it is not a single cell as multirow
- if span > 1:
- document.body[begin_cell] = document.body[begin_cell].replace(' bottomline="true" ', ' ')
- document.body[begin_cell] = document.body[begin_cell].replace(' mroffset=', '')
- # store the offset and remove it
- begin = document.body[begin_cell].find('"')
- end = document.body[begin_cell].find('" ', begin)
- offset = document.body[begin_cell][begin + 1:end]
- document.body[begin_cell] = document.body[begin_cell].replace(document.body[begin_cell][begin:end + 1], '')
- # write ERT to create the multirow cell
- cend = find_token(document.body, "</cell>", begin_cell)
- if cend == -1:
- document.warning("Malformed LyX document: Could not find end of tabular cell.")
- i += 1
- continue
- blay = find_token(document.body, "\\begin_layout", begin_cell, cend)
+ lineno = multirows[r-1][col][0]
+ document.body[lineno] = document.body[lineno].replace(' bottomline="true" ', ' ')
+ # revert beginning cell
+ bcell = multirows[row][col][0]
+ ecell = multirows[row][col][1]
+ offset = get_option_value(document.body[bcell], "mroffset")
+ document.body[bcell] = document.body[bcell].\
+ replace(' multirow="3" ', ' ').\
+ replace('valignment="middle"', 'valignment="top"')
+ # remove mroffset option
+ document.body[bcell] = rgx.sub('', document.body[bcell])
+
+ blay = find_token(document.body, "\\begin_layout", bcell, ecell)
if blay == -1:
document.warning("Can't find layout for cell!")
- i = j
continue
bend = find_end_of_layout(document.body, blay)
if bend == -1:
document.warning("Can't find end of layout for cell!")
- i = cend
continue
# do the later one first, so as not to mess up the numbering
# we are wrapping the whole cell in this ert
# so before the end of the layout...
document.body[bend:bend] = put_cmd_in_ert("}")
# ...and after the beginning
- document.body[blay + 1:blay + 1] = put_cmd_in_ert("\\multirow{" + str(span) + "}{" + col_width + "}[" + offset + "]{")
- # set index to end of current cell (that is now much bigger due to the ERT addition)
- begin_cell = find_token(document.body, '</cell>', begin_cell) + 1
- begin_row = find_token(document.body, '</row>', begin_row) + 1
+ document.body[blay + 1:blay + 1] = \
+ put_cmd_in_ert("\\multirow{" + str(nummrs) + "}{" + col_width + "}[" \
+ + offset + "]{")
+
+ # on to the next table
+ begin_table = end_table
+
+
+def revert_script(document):
+ " Convert subscript/superscript inset to TeX code "
+ i = 0
+ foundsubscript = False
+ while True:
+ i = find_token(document.body, '\\begin_inset script', i)
+ if i == -1:
+ break
+ z = find_end_of_inset(document.body, i)
+ if z == -1:
+ document.warning("Malformed LyX document: Can't find end of script inset.")
+ i += 1
+ continue
+ blay = find_token(document.body, "\\begin_layout", i, z)
+ if blay == -1:
+ document.warning("Malformed LyX document: Can't find layout in script inset.")
+ i = z
+ continue
+
+ if check_token(document.body[i], "\\begin_inset script subscript"):
+ subst = '\\textsubscript{'
+ foundsubscript = True
+ elif check_token(document.body[i], "\\begin_inset script superscript"):
+ subst = '\\textsuperscript{'
+ else:
+ document.warning("Malformed LyX document: Unknown type of script inset.")
+ i = z
+ continue
+ bend = find_end_of_layout(document.body, blay)
+ if bend == -1 or bend > z:
+ document.warning("Malformed LyX document: Can't find end of layout in script inset.")
+ i = z
+ continue
+ # remove the \end_layout \end_inset pair
+ document.body[bend:z + 1] = put_cmd_in_ert("}")
+ document.body[i:blay + 1] = put_cmd_in_ert(subst)
+ i += 1
+ # these classes provide a \textsubscript command:
+ # FIXME: Would be nice if we could use the information of the .layout file here
+ classes = ["memoir", "scrartcl", "scrbook", "scrlttr2", "scrreprt"]
+ if foundsubscript and find_token_exact(classes, document.textclass, 0) == -1:
+ add_to_preamble(document, ['\\usepackage{subscript}'])
+
+
+def convert_use_xetex(document):
+ " convert \\use_xetex to \\use_non_tex_fonts "
+ i = find_token(document.header, "\\use_xetex", 0)
+ if i == -1:
+ document.header.insert(-1, "\\use_non_tex_fonts 0")
+ else:
+ val = get_value(document.header, "\\use_xetex", 0)
+ document.header[i] = "\\use_non_tex_fonts " + val
+
+
+def revert_use_xetex(document):
+ " revert \\use_non_tex_fonts to \\use_xetex "
+ i = 0
+ i = find_token(document.header, "\\use_non_tex_fonts", 0)
+ if i == -1:
+ document.warning("Malformed document. No \\use_non_tex_fonts param!")
+ return
+
+ val = get_value(document.header, "\\use_non_tex_fonts", 0)
+ document.header[i] = "\\use_xetex " + val
+
+
+def revert_labeling(document):
+ koma = ("scrartcl", "scrarticle-beamer", "scrbook", "scrlettr",
+ "scrlttr2", "scrreprt")
+ if document.textclass in koma:
+ return
+ i = 0
+ while True:
+ i = find_token_exact(document.body, "\\begin_layout Labeling", i)
+ if i == -1:
+ return
+ document.body[i] = "\\begin_layout List"
+
- # go to the next table
- begin_table = end_table + 1
+def revert_langpack(document):
+ " revert \\language_package parameter "
+ i = 0
+ i = find_token(document.header, "\\language_package", 0)
+ if i == -1:
+ document.warning("Malformed document. No \\language_package param!")
+ return
+
+ del document.header[i]
+
+
+def convert_langpack(document):
+ " Add \\language_package parameter "
+ i = find_token(document.header, "\language" , 0)
+ if i == -1:
+ document.warning("Malformed document. No \\language defined!")
+ return
+
+ document.header.insert(i + 1, "\\language_package default")
+
+
+def revert_tabularwidth(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)
+ if j == -1:
+ document.warning("Unable to find end of Tabular inset at line " + str(i))
+ i += 1
+ continue
+ i += 1
+ features = find_token(document.body, "<features", i, j)
+ if features == -1:
+ document.warning("Can't find any features in Tabular inset at line " + str(i))
+ i = j
+ continue
+ if document.body[features].find('alignment="tabularwidth"') != -1:
+ remove_option(document.body, features, 'tabularwidth')
- if multirowOffset == True:
- add_to_preamble(document, ["\\usepackage{multirow}"])
+def revert_html_css_as_file(document):
+ if not del_token(document.header, '\\html_css_as_file', 0):
+ document.warning("Malformed LyX document: Missing \\html_css_as_file.")
##
[368, []],
[369, [convert_author_id]],
[370, []],
- [371, []],
+ [371, [convert_mhchem]],
[372, []],
[373, [merge_gbrief]],
[374, []],
[404, [convert_prettyref]],
[405, []],
[406, [convert_passthru]],
- [407, []]
+ [407, []],
+ [408, []],
+ [409, [convert_use_xetex]],
+ [410, []],
+ [411, [convert_langpack]],
+ [412, []],
+ [413, []]
]
-revert = [[406, [revert_multirowOffset]],
+revert = [[412, [revert_html_css_as_file]],
+ [411, [revert_tabularwidth]],
+ [410, [revert_langpack]],
+ [409, [revert_labeling]],
+ [408, [revert_use_xetex]],
+ [407, [revert_script]],
+ [406, [revert_multirowOffset]],
[405, [revert_passthru]],
[404, []],
[403, [revert_refstyle]],
projects / lyx.git / blobdiff