#
# 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 1.4"""
from os import access, F_OK
import os.path
from parser_tools import check_token, find_token, \
- get_value, del_token, is_nonempty_line, \
+ get_value, is_nonempty_line, \
find_tokens, find_end_of, find_beginning_of, find_token_exact, find_tokens_exact, \
find_re, find_tokens_backwards
from sys import stdin
-from string import replace, split, find, strip, join
from lyx_0_12 import update_latexaccents
def get_layout(line, default_layout):
" Get layout, if empty return the default layout."
- tokens = split(line)
+ tokens = line.split()
if len(tokens) > 1:
return tokens[1]
return default_layout
"Finds the matching \end_inset"
return find_end_of(lines, i, "\\begin_inset", "\\end_inset")
+def del_token(lines, token, start, end):
+ """ del_token(lines, token, start, end) -> int
+
+ Find the lower line in lines where token is the first element and
+ delete that line.
+
+ Returns the number of lines remaining."""
+
+ k = find_token_exact(lines, token, start, end)
+ if k == -1:
+ return end
+ else:
+ del lines[k]
+ return end - 1
+
# End of helper functions
####################################################################
i = find_token(document.body, "\\color default", i)
if i == -1:
return
- document.body[i] = replace(document.body[i], "\\color default",
- "\\color inherit")
+ document.body[i] = document.body[i].replace("\\color default",
+ "\\color inherit")
def add_end_header(document):
if i == -1:
document.warning("Malformed LyX document: Missing '\\use_amsmath'.")
return
- tokens = split(document.header[i])
+ tokens = document.header[i].split()
if len(tokens) != 2:
document.warning("Malformed LyX document: Could not parse line '%s'." % document.header[i])
use_amsmath = '0'
if i == -1:
document.warning("Malformed LyX document: Missing '\\use_amsmath'.")
return
- tokens = split(document.header[i])
+ tokens = document.header[i].split()
if len(tokens) != 2:
document.warning("Malformed LyX document: Could not parse line '%s'." % document.header[i])
use_amsmath = '0'
def convert_spaces(document):
" \SpecialChar ~ -> \InsetSpace ~"
for i in range(len(document.body)):
- document.body[i] = replace(document.body[i],"\\SpecialChar ~","\\InsetSpace ~")
+ document.body[i] = document.body[i].replace("\\SpecialChar ~",
+ "\\InsetSpace ~")
def revert_spaces(document):
""" \InsetSpace \, -> \InsetSpace \thinspace{}
\InsetSpace \space -> \InsetSpace \space{}"""
for i in range(len(document.body)):
- document.body[i] = replace(document.body[i],"\\InsetSpace \\space","\\InsetSpace \\space{}")
- document.body[i] = replace(document.body[i],"\\InsetSpace \,","\\InsetSpace \\thinspace{}")
+ document.body[i] = document.body[i].replace("\\InsetSpace \\space",
+ "\\InsetSpace \\space{}")
+ document.body[i] = document.body[i].replace("\\InsetSpace \,",
+ "\\InsetSpace \\thinspace{}")
def revert_space_names(document):
""" \InsetSpace \thinspace{} -> \InsetSpace \,
\InsetSpace \space{} -> \InsetSpace \space"""
for i in range(len(document.body)):
- document.body[i] = replace(document.body[i],"\\InsetSpace \\space{}","\\InsetSpace \\space")
- document.body[i] = replace(document.body[i],"\\InsetSpace \\thinspace{}","\\InsetSpace \\,")
+ document.body[i] = document.body[i].replace("\\InsetSpace \\space{}",
+ "\\InsetSpace \\space")
+ document.body[i] = document.body[i].replace("\\InsetSpace \\thinspace{}",
+ "\\InsetSpace \\,")
def lyx_support_escape(lab):
- " Equivalent to lyx::support::escape()"
+ " Equivalent to pre-unicode lyx::support::escape()"
hexdigit = ['0', '1', '2', '3', '4', '5', '6', '7',
'8', '9', 'A', 'B', 'C', 'D', 'E', 'F']
enc = ""
def convert_bibtex(document):
" Convert BibTeX changes."
for i in range(len(document.body)):
- document.body[i] = replace(document.body[i],"\\begin_inset LatexCommand \\BibTeX",
- "\\begin_inset LatexCommand \\bibtex")
+ document.body[i] = document.body[i].replace("\\begin_inset LatexCommand \\BibTeX",
+ "\\begin_inset LatexCommand \\bibtex")
def revert_bibtex(document):
" Revert BibTeX changes."
for i in range(len(document.body)):
- document.body[i] = replace(document.body[i], "\\begin_inset LatexCommand \\bibtex",
- "\\begin_inset LatexCommand \\BibTeX")
+ document.body[i] = document.body[i].replace("\\begin_inset LatexCommand \\bibtex",
+ "\\begin_inset LatexCommand \\BibTeX")
def remove_insetparent(document):
if i == -1:
break
- template = split(document.body[i+1])
+ template = document.body[i+1].split()
template.reverse()
del document.body[i+1]
- filename = split(document.body[i+1])
+ filename = document.body[i+1].split()
filename.reverse()
del document.body[i+1]
- params = split(document.body[i+1])
+ params = document.body[i+1].split()
params.reverse()
if document.body[i+1]: del document.body[i+1]
- document.body[i] = document.body[i] + " " + template[0]+ ', "' + filename[0] + '", " '+ join(params[1:]) + '"'
+ document.body[i] = document.body[i] + " " + template[0]+ ', "' + filename[0] + '", " '+ " ".join(params[1:]) + '"'
i = i + 1
#but if this happens deal with it greacefully adding
#the missing \end_deeper.
i = len(document.body) - 1
- document.body[i:i] = ["\end_deeper",""]
+ document.body[i:i] = ["\\end_deeper",""]
return
else:
del document.body[i]
i = i + 1
continue
- if find(document.body[i], comment) == -1:
+ if document.body[i].find(comment) == -1:
document.body[i:i] = ["\\end_inset"]
i = i + 1
break
"\\begin_deeper", "\\end_deeper", "\\the_end"], i)
if i != -1:
- token = split(document.body[i])[0]
+ token = document.body[i].split()[0]
else:
document.warning("Truncated document.")
i = len(document.body)
document.body.insert(i,"")
document.body.insert(i,"\\end_layout")
i = i + 3
+ # consecutive begin_deeper only insert one end_layout
+ while document.body[i].startswith('\\begin_deeper'):
+ i += 1
struct_stack.append(token)
continue
if i == -1:
return
- document.body[i] = replace(document.body[i], '\\layout', '\\begin_layout')
+ document.body[i] = document.body[i].replace('\\layout', '\\begin_layout')
i = i + 1
if i == -1:
return
- document.body[i] = replace(document.body[i], '\\begin_layout', '\\layout')
+ document.body[i] = document.body[i].replace('\\begin_layout', '\\layout')
i = i + 1
'valignment="center" -> valignment="middle"'
for i in range(start, end):
if re.search('^<(column|cell) .*valignment="center".*>$', body[i]):
- body[i] = replace(body[i], 'valignment="center"', 'valignment="middle"')
+ body[i] = body[i].replace('valignment="center"', 'valignment="middle"')
def convert_table_valignment_middle(document):
" valignment, middle -> center"
for i in range(start, end):
if re.search('^<(column|cell) .*valignment="middle".*>$', body[i]):
- body[i] = replace(body[i], 'valignment="middle"', 'valignment="center"')
+ body[i] = body[i].replace('valignment="middle"', 'valignment="center"')
def revert_valignment_middle(document):
# Merge all paragraph parameters into a single line
# We cannot check for '\\' only because paragraphs may start e.g.
# with '\\backslash'
- while document.body[i + 1][:1] == '\\' and split(document.body[i + 1][1:])[0] in par_params:
+ while document.body[i + 1][:1] == '\\' and document.body[i + 1][1:].split()[0] in par_params:
document.body[i] = document.body[i + 1] + ' ' + document.body[i]
del document.body[i+1]
- line_top = find(document.body[i],"\\line_top")
- line_bot = find(document.body[i],"\\line_bottom")
- pb_top = find(document.body[i],"\\pagebreak_top")
- pb_bot = find(document.body[i],"\\pagebreak_bottom")
- vspace_top = find(document.body[i],"\\added_space_top")
- vspace_bot = find(document.body[i],"\\added_space_bottom")
+ line_top = document.body[i].find("\\line_top")
+ line_bot = document.body[i].find("\\line_bottom")
+ pb_top = document.body[i].find("\\pagebreak_top")
+ pb_bot = document.body[i].find("\\pagebreak_bottom")
+ vspace_top = document.body[i].find("\\added_space_top")
+ vspace_bot = document.body[i].find("\\added_space_bottom")
if line_top == -1 and line_bot == -1 and pb_bot == -1 and pb_top == -1 and vspace_top == -1 and vspace_bot == -1:
continue
# inherit font sizes.
nonstandard = 0
if (not document.is_default_layout(layout) or
- find(document.body[i],"\\align") != -1 or
- find(document.body[i],"\\labelwidthstring") != -1 or
- find(document.body[i],"\\noindent") != -1):
+ document.body[i].find("\\align") != -1 or
+ document.body[i].find("\\labelwidthstring") != -1 or
+ document.body[i].find("\\noindent") != -1):
nonstandard = 1
# get the font size of the beginning of this paragraph, since we need
while not is_nonempty_line(document.body[j]):
j = j + 1
size_top = ""
- if find(document.body[j], "\\size") != -1:
- size_top = split(document.body[j])[1]
+ if document.body[j].find("\\size") != -1:
+ size_top = document.body[j].split()[1]
for tag in "\\line_top", "\\line_bottom", "\\pagebreak_top", "\\pagebreak_bottom":
- document.body[i] = replace(document.body[i], tag, "")
+ document.body[i] = document.body[i].replace(tag, "")
if vspace_top != -1:
# the position could be change because of the removal of other
# paragraph properties above
- vspace_top = find(document.body[i],"\\added_space_top")
- tmp_list = split(document.body[i][vspace_top:])
+ vspace_top = document.body[i].find("\\added_space_top")
+ tmp_list = document.body[i][vspace_top:].split()
vspace_top_value = tmp_list[1]
- document.body[i] = document.body[i][:vspace_top] + join(tmp_list[2:])
+ document.body[i] = document.body[i][:vspace_top] + " ".join(tmp_list[2:])
if vspace_bot != -1:
# the position could be change because of the removal of other
# paragraph properties above
- vspace_bot = find(document.body[i],"\\added_space_bottom")
- tmp_list = split(document.body[i][vspace_bot:])
+ vspace_bot = document.body[i].find("\\added_space_bottom")
+ tmp_list = document.body[i][vspace_bot:].split()
vspace_bot_value = tmp_list[1]
- document.body[i] = document.body[i][:vspace_bot] + join(tmp_list[2:])
+ document.body[i] = document.body[i][:vspace_bot] + " ".join(tmp_list[2:])
- document.body[i] = strip(document.body[i])
+ document.body[i] = document.body[i].strip()
i = i + 1
# Create an empty paragraph or paragraph fragment for line and
size_bot = size_top
j = i + 1
while j < k:
- if find(document.body[j], "\\size") != -1:
- size_bot = split(document.body[j])[1]
+ if document.body[j].find("\\size") != -1:
+ size_bot = document.body[j].split()[1]
j = j + 1
- elif find(document.body[j], "\\begin_inset") != -1:
+ elif document.body[j].find("\\begin_inset") != -1:
# skip insets
j = find_end_of_inset(document.body, j)
else:
# We cannot check for '\\' only because paragraphs may start e.g.
# with '\\backslash'
params = ''
- while lines[i][:1] == '\\' and split(lines[i][1:])[0] in par_params:
- params = params + ' ' + strip(lines[i])
+ while lines[i][:1] == '\\' and lines[i][1:].split()[0] in par_params:
+ params = params + ' ' + lines[i].strip()
i = i + 1
- return strip(params)
+ return params.strip()
def lyxsize2latexsize(lyxsize):
if i == -1:
defskipamount = 'medskip'
else:
- defskipamount = split(document.header[i])[1]
+ defskipamount = document.header[i].split()[1]
keys = {"\\begin_inset" : "vspace", "\\lyxline" : "lyxline",
"\\newpage" : "newpage"}
size = "normal"
# Paragraph parameters may be on one or more lines.
# Find the start of the real paragraph text.
- while document.body[start][:1] == '\\' and split(document.body[start])[0] in params:
+ while document.body[start][:1] == '\\' and document.body[start].split()[0] in params:
start = start + 1
for k in range(start, i):
- if find(document.body[k], "\\size") != -1:
+ if document.body[k].find("\\size") != -1:
# store font size
- size = split(document.body[k])[1]
+ size = document.body[k].split()[1]
elif is_nonempty_line(document.body[k]):
paragraph_start = 0
break
while k < next_par:
if find_tokens(document.body, tokens, k) == k:
# inset to convert
- lines.append(split(document.body[k]))
+ lines.append(document.body[k].split())
insets.append(keys[lines[n][0]])
del_lines.append([k, k])
top.append(0)
sizes.append(size)
n = n + 1
inset_end = k
- elif find(document.body[k], "\\size") != -1:
+ elif document.body[k].find("\\size") != -1:
# store font size
- size = split(document.body[k])[1]
+ size = document.body[k].split()[1]
elif find_token(document.body, "\\begin_inset ERT", k) == k:
ert_begin = find_token(document.body, "\\layout", k) + 1
if ert_begin == 0:
# determine font size
prev_size = "normal"
k = prev_par + 1
- while document.body[k][:1] == '\\' and split(document.body[k])[0] in prev_params:
+ while document.body[k][:1] == '\\' and document.body[k].split()[0] in prev_params:
k = k + 1
while k < this_par:
- if find(document.body[k], "\\size") != -1:
- prev_size = split(document.body[k])[1]
+ if document.body[k].find("\\size") != -1:
+ prev_size = document.body[k].split()[1]
break
- elif find(document.body[k], "\\begin_inset") != -1:
+ elif document.body[k].find("\\begin_inset") != -1:
# skip insets
k = find_end_of_inset(document.body, k)
elif is_nonempty_line(document.body[k]):
if next_par > 0 and not after:
next_params = get_par_params(document.body, next_par + 1)
ert = 0
- while document.body[k][:1] == '\\' and split(document.body[k])[0] in next_params:
+ while document.body[k][:1] == '\\' and document.body[k].split()[0] in next_params:
k = k + 1
# determine font size
next_size = "normal"
k = next_par + 1
while k < this_par:
- if find(document.body[k], "\\size") != -1:
- next_size = split(document.body[k])[1]
+ if document.body[k].find("\\size") != -1:
+ next_size = document.body[k].split()[1]
break
elif is_nonempty_line(document.body[k]):
break
# Convert LyX units to LaTeX units
for unit in units.keys():
- if find(len, unit) != -1:
+ if len.find(unit) != -1:
len = '%f' % (len2value(len) / 100) + units[unit]
break
'special':'none', 'height':'1in',
'height_special':'totalheight', 'collapsed':'false'}
for key in params.keys():
- value = replace(get_value(document.body, key, i, j), '"', '')
+ value = get_value(document.body, key, i, j).replace('"', '')
if value != "":
if key == 'position':
# convert new to old position: 'position "t"' -> 0
i = find_token(document.header, "\\branch", i)
if i == -1:
break
- document.warning("Removing branch %s." % split(document.header[i])[1])
+ document.warning("Removing branch %s." % document.header[i].split()[1])
j = find_token(document.header, "\\end_branch", i)
if j == -1:
document.warning("Malformed LyX document: Missing '\\end_branch'.")
def revert_float(document):
- " Revert sideway floats. "
+ " Revert sideways floats. "
i = 0
while 1:
i = find_token_exact(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)
+ floattype = m.group(1)
+ if floattype != "figure" and floattype != "table":
+ i = 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
continue
if get_value(document.body, 'sideways', i, j) != "false":
- document.warning("Conversion of 'sideways true' not yet implemented.")
- # Don't remove 'sideways' so that people will get warnings by lyx
+ 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] = '\\layout Standard\n\\begin_inset ERT\nstatus Collapsed\n\n' \
+ '\\layout Standard\n\n\n\\backslash\n' \
+ 'end{sideways' + floattype + '}\n\n\\end_inset\n'
+ del document.body[i+1:l-1]
+ document.body[i] = '\\begin_inset ERT\nstatus Collapsed\n\n' \
+ '\\layout Standard\n\n\n\\backslash\n' \
+ 'begin{sideways' + floattype + '}\n\n\\end_inset\n\n'
+ add_to_preamble(document,
+ ['\\usepackage{rotfloat}\n'])
i = i + 1
continue
del_token(document.body, 'sideways', i, j)
if j == -1:
return
i = i + 1
- filename = split(document.body[j])[1]
+ filename = document.body[j].split()[1]
absname = os.path.normpath(os.path.join(document.dir, filename))
if document.input == stdin and not os.path.isabs(filename):
# We don't know the directory and cannot check the document.
# We could use a heuristic and take the current directory,
# and we could try to find out if documentname has an extension,
# but that would be just guesses and could be wrong.
- document.warning("""Warning: Can not determine whether document
+ document.warning("""Warning: Cannot determine whether document
%s
needs an extension when reading from standard input.
You may need to correct the document manually or run
if access(absname, F_OK):
continue
if access(absname + ".ps", F_OK):
- document.body[j] = replace(document.body[j], filename, filename + ".ps")
+ document.body[j] = document.body[j].replace(filename, filename + ".ps")
continue
if access(absname + ".eps", F_OK):
- document.body[j] = replace(document.body[j], filename, filename + ".eps")
+ document.body[j] = document.body[j].replace(filename, filename + ".eps")
def convert_names(document):
document.warning("Malformed lyx document: Missing '\\use_jurabib'.")
return
- use_natbib = int(split(document.header[a])[1])
- use_numerical_citations = int(split(document.header[b])[1])
- use_jurabib = int(split(document.header[c])[1])
+ use_natbib = int(document.header[a].split()[1])
+ use_numerical_citations = int(document.header[b].split()[1])
+ use_jurabib = int(document.header[c].split()[1])
cite_engine = "basic"
if use_natbib:
document.warning("Malformed lyx document: Missing '\\cite_engine'.")
return
- cite_engine = split(document.header[i])[1]
+ cite_engine = document.header[i].split()[1]
use_natbib = '0'
use_numerical = '0'
return
packages = {'default':'none','a4':'none', 'a4wide':'a4', 'widemarginsa4':'a4wide'}
- if len(split(document.header[i])) > 1:
- paperpackage = split(document.header[i])[1]
- document.header[i] = replace(document.header[i], paperpackage, packages[paperpackage])
+ if len(document.header[i].split()) > 1:
+ paperpackage = document.header[i].split()[1]
+ document.header[i] = document.header[i].replace(paperpackage, packages[paperpackage])
else:
document.header[i] = document.header[i] + ' widemarginsa4'
packages = {'none':'a4', 'a4':'a4wide', 'a4wide':'widemarginsa4',
'widemarginsa4':'', 'default': 'default'}
- if len(split(document.header[i])) > 1:
- paperpackage = split(document.header[i])[1]
+ if len(document.header[i].split()) > 1:
+ paperpackage = document.header[i].split()[1]
else:
paperpackage = 'default'
- document.header[i] = replace(document.header[i], paperpackage, packages[paperpackage])
+ document.header[i] = document.header[i].replace(paperpackage, packages[paperpackage])
def convert_bullets(document):
if i == -1:
return
if document.header[i][:12] == '\\bulletLaTeX':
- document.header[i] = document.header[i] + ' ' + strip(document.header[i+1])
+ document.header[i] = document.header[i] + ' ' + document.header[i+1].strip()
n = 3
else:
- document.header[i] = document.header[i] + ' ' + strip(document.header[i+1]) +\
- ' ' + strip(document.header[i+2]) + ' ' + strip(document.header[i+3])
+ document.header[i] = document.header[i] + ' ' + document.header[i+1].strip() +\
+ ' ' + document.header[i+2].strip() + ' ' + document.header[i+3].strip()
n = 5
del document.header[i+1:i + n]
i = i + 1
if i == -1:
return
if document.header[i][:12] == '\\bulletLaTeX':
- n = find(document.header[i], '"')
+ n = document.header[i].find('"')
if n == -1:
document.warning("Malformed header.")
return
document.header[i:i+1] = [document.header[i][:n-1],'\t' + document.header[i][n:], '\\end_bullet']
i = i + 3
else:
- frag = split(document.header[i])
+ frag = document.header[i].split()
if len(frag) != 5:
document.warning("Malformed header.")
return
if i == -1:
return
- tmp = split(document.header[i])
+ tmp = document.header[i].split()
if tmp[1] == "Default":
document.header[i] = '\\papersize default'
return
if i == -1:
return
- tmp = split(document.header[i])
+ tmp = document.header[i].split()
if tmp[1] == "custom":
document.header[i] = '\\papersize Custom'
" Strip spaces at end of command line. "
for i in range(len(document.body)):
if document.body[i][:1] == '\\':
- document.body[i] = strip(document.body[i])
+ document.body[i] = document.body[i].strip()
def use_x_boolean(document):
i = find_token(document.header, use, 0)
if i == -1:
continue
- decompose = split(document.header[i])
+ decompose = document.header[i].split()
document.header[i] = decompose[0] + ' ' + bin2bool[decompose[1]]
i = find_token(document.header, use, 0)
if i == -1:
continue
- decompose = split(document.header[i])
+ decompose = document.header[i].split()
document.header[i] = decompose[0] + ' ' + bool2bin[decompose[1]]
i = i + 1
while 1:
- if strip(body[i]) and split(body[i])[0] not in allowed_parameters:
+ if body[i].strip() and body[i].split()[0] not in allowed_parameters:
break
- j = find(body[i],'\\', 1)
+ j = body[i].find('\\', 1)
if j != -1:
- body[i:i+1] = [strip(body[i][:j]), body[i][j:]]
+ body[i:i+1] = [body[i][:j].strip(), body[i][j:]]
i = i + 1
# remove all paragraph parameters and font settings
k = i
while k < j:
- if (strip(document.body[k]) and
- split(document.body[k])[0] in forbidden_settings):
+ if (document.body[k].strip() and
+ document.body[k].split()[0] in forbidden_settings):
del document.body[k]
j = j - 1
else:
if k == -1:
break
document.body[k:k+1] = ["\\end_layout", "", '\\begin_layout %s' % document.default_layout]
- k = k + 4
- j = j + 3
+ k = k + 3
+ j = j + 2
# We need an empty line if document.default_layout == ''
- if document.body[k-1] != '':
- document.body.insert(k-1, '')
+ if document.body[k] != '':
+ document.body.insert(k, '')
k = k + 1
j = j + 1
i = i + 1
l = k + 1
while document.body[l] == "":
l = l + 1
- if strip(document.body[l]) and split(document.body[l])[0] == "\\newline":
+ if document.body[l].strip() and document.body[l].split()[0] == "\\newline":
document.body[k:l+1] = ["\\end_layout", "",
'\\begin_layout %s' % document.default_layout]
j = j - l + k + 2
if i == -1:
return
- paperpackage = split(document.header[i])[1]
+ paperpackage = document.header[i].split()[1]
del document.header[i]
i = i + 10
##
-# Convertion hub
+# Conversion hub
#
supported_versions = ["1.4.%d" % i for i in range(3)] + ["1.4"]
projects / lyx.git / blobdiff