mirror of
https://git.lyx.org/repos/lyx.git
synced 2024-12-22 05:16:21 +00:00
Fix lyx2lyx dash conversion and make it faster.
This commit is contained in:
Günter Milde
parent
62f32992ab
commit
a151b274bf
@ -1557,10 +1557,11 @@ def convert_latexargs(document):
|
||||
"theorems-chap-bytype", "theorems-chap", "theorems-named", "theorems-sec-bytype",
|
||||
"theorems-sec", "theorems-starred", "theorems-std", "todonotes"]
|
||||
# Modules we need to take care of
|
||||
caveat_modules = ["initials"]
|
||||
caveat_modules = ["initials"] # TODO: , "graphicboxes", "bicaption"]
|
||||
# information about the relevant styles in caveat_modules (number of opt and req args)
|
||||
# use this if we get more caveat_modules. For now, use hard coding (see below).
|
||||
# initials = [{'Layout' : 'Initial', 'opt' : 1, 'req' : 1}]
|
||||
# graphicboxes = { ... }
|
||||
|
||||
# Is this a known safe layout?
|
||||
safe_layout = document.textclass in safe_layouts
|
||||
|
||||
@ -37,7 +37,7 @@ from lyx2lyx_tools import (add_to_preamble, put_cmd_in_ert, get_ert,
|
||||
from parser_tools import (check_token, del_complete_lines,
|
||||
find_end_of_inset, find_end_of_layout, find_nonempty_line, find_re,
|
||||
find_token, find_token_backwards, get_containing_layout,
|
||||
get_value, is_in_inset)
|
||||
get_containing_inset, get_value, is_in_inset)
|
||||
|
||||
|
||||
####################################################################
|
||||
@ -622,41 +622,40 @@ def convert_dashes(document):
|
||||
while i+1 < len(lines):
|
||||
i += 1
|
||||
line = lines[i]
|
||||
words = line.split()
|
||||
if (len(words) > 1 and words[0] == "\\begin_inset"
|
||||
and (words[1] in ["CommandInset", "ERT", "External", "Formula",
|
||||
"FormulaMacro", "Graphics", "IPA", "listings"]
|
||||
or line.endswith("Flex Code"))):
|
||||
# must not replace anything in insets that store LaTeX contents in .lyx files
|
||||
# (math and command insets without overridden read() and write() methods
|
||||
# filtering out IPA makes Text::readParToken() more simple
|
||||
# skip ERT as well since it is not needed there
|
||||
# Flex Code is logical markup, typically rendered as typewriter
|
||||
j = find_end_of_inset(lines, i)
|
||||
if j == -1:
|
||||
document.warning("Malformed LyX document: Can't find end of " +
|
||||
words[1] + " inset at line " + str(i))
|
||||
else:
|
||||
i = j
|
||||
continue
|
||||
if lines[i] == "\\begin_layout LyX-Code":
|
||||
j = find_end_of_layout(lines, i)
|
||||
if j == -1:
|
||||
document.warning("Malformed LyX document: "
|
||||
"Can't find end of %s layout at line %d" % (words[1],i))
|
||||
else:
|
||||
i = j
|
||||
if "--" not in line:
|
||||
continue
|
||||
# skip label width string (bug 10243):
|
||||
if line.startswith("\\labelwidthstring"):
|
||||
# skip label width string (bug 10243)
|
||||
continue
|
||||
|
||||
if "--" in line:
|
||||
# We can have an arbitrary number of consecutive hyphens.
|
||||
# Replace as LaTeX does: First try emdash, then endash
|
||||
line = line.replace("---", "\\threehyphens\n")
|
||||
line = line.replace("--", "\\twohyphens\n")
|
||||
lines[i:i+1] = line.splitlines()
|
||||
# Do not touch hyphens in some insets:
|
||||
try:
|
||||
value, start, end = get_containing_inset(lines, i)
|
||||
except TypeError:
|
||||
# False means no (or malformed) containing inset
|
||||
value, start, end = "no inset", -1, -1
|
||||
# We must not replace anything in insets that store LaTeX contents in .lyx files
|
||||
# (math and command insets without overridden read() and write() methods.
|
||||
# Filtering out IPA and ERT makes Text::readParToken() more simple,
|
||||
# Flex Code is logical markup, typically rendered as typewriter
|
||||
if (value.split()[0] in ["CommandInset", "ERT", "External", "Formula",
|
||||
"FormulaMacro", "Graphics", "IPA", "listings"]
|
||||
or value in ["Flex Code", "Flex URL"]):
|
||||
i = end
|
||||
continue
|
||||
try:
|
||||
layout, start, end, j = get_containing_layout(lines, i)
|
||||
except TypeError: # no (or malformed) containing layout
|
||||
document.warning("Malformed LyX document: "
|
||||
"Can't find layout at line %d" % i)
|
||||
continue
|
||||
if layout == "LyX-Code":
|
||||
i = end
|
||||
continue
|
||||
# We can have an arbitrary number of consecutive hyphens.
|
||||
# Replace as LaTeX does: First try emdash, then endash
|
||||
line = line.replace("---", "\\threehyphens\n")
|
||||
line = line.replace("--", "\\twohyphens\n")
|
||||
lines[i:i+1] = line.splitlines()
|
||||
|
||||
# remove ligature breaks between dashes
|
||||
i = 1
|
||||
@ -672,40 +671,40 @@ def convert_dashes(document):
|
||||
|
||||
def revert_dashes(document):
|
||||
"""
|
||||
Remove preamble code from 2.3->2.2 conversion.
|
||||
Prevent ligatures of existing --- and --.
|
||||
Revert \\twohyphens and \\threehyphens to -- and ---.
|
||||
Remove preamble code from 2.3->2.2 conversion.
|
||||
"""
|
||||
del_complete_lines(document.preamble,
|
||||
['% Added by lyx2lyx',
|
||||
r'\renewcommand{\textendash}{--}',
|
||||
r'\renewcommand{\textemdash}{---}'])
|
||||
|
||||
# Insert ligature breaks to prevent ligation of hyphens to dashes:
|
||||
lines = document.body
|
||||
i = 0
|
||||
while i+1 < len(lines):
|
||||
i += 1
|
||||
line = lines[i]
|
||||
if "--" not in line:
|
||||
continue
|
||||
# skip label width string (bug 10243):
|
||||
if line.startswith("\\labelwidthstring"):
|
||||
continue
|
||||
# do not touch hyphens in some insets (cf. convert_dashes):
|
||||
if line.startswith("\\begin_inset"):
|
||||
try:
|
||||
if line.split()[1] in ["CommandInset", "ERT", "External",
|
||||
"Formula", "FormulaMacro", "Graphics",
|
||||
"IPA", "listings"]:
|
||||
j = find_end_of_inset(lines, i)
|
||||
if j == -1:
|
||||
document.warning("Malformed LyX document: Can't find "
|
||||
"end of %s inset at line %d." % (itype, i))
|
||||
continue
|
||||
i = j
|
||||
except IndexError:
|
||||
continue
|
||||
if "--" in line:
|
||||
line = line.replace("--", "-\\SpecialChar \\textcompwordmark{}\n-")
|
||||
document.body[i:i+1] = line.split('\n')
|
||||
try:
|
||||
value, start, end = get_containing_inset(lines, i)
|
||||
except TypeError:
|
||||
# False means no (or malformed) containing inset
|
||||
value, start, end = "no inset", -1, -1
|
||||
if (value.split()[0] in ["CommandInset", "ERT", "External", "Formula",
|
||||
"FormulaMacro", "Graphics", "IPA", "listings"]
|
||||
or value == "Flex URL"):
|
||||
i = end
|
||||
continue
|
||||
line = line.replace("--", "-\\SpecialChar \\textcompwordmark{}\n-")
|
||||
document.body[i:i+1] = line.split('\n')
|
||||
|
||||
# Revert \twohyphens and \threehyphens:
|
||||
i = 1
|
||||
while i < len(lines):
|
||||
|
||||
@ -26,7 +26,7 @@ import sys, os
|
||||
|
||||
from parser_tools import (del_token, del_value, del_complete_lines,
|
||||
find_complete_lines, find_end_of, find_end_of_layout, find_end_of_inset,
|
||||
find_re, find_token, find_token_backwards,
|
||||
find_re, find_token, find_token_backwards, get_containing_inset,
|
||||
get_containing_layout, get_bool_value, get_value, get_quoted_value)
|
||||
# find_tokens, find_token_exact, is_in_inset,
|
||||
# check_token, get_option_value
|
||||
@ -1853,46 +1853,54 @@ def convert_dashligatures(document):
|
||||
if use_dash_ligatures is None:
|
||||
# Look for dashes (Documents by LyX 2.1 or older have "\twohyphens\n"
|
||||
# or "\threehyphens\n" as interim representation for -- an ---.)
|
||||
has_literal_dashes = False
|
||||
has_ligature_dashes = False
|
||||
j = 0
|
||||
for i, line in enumerate(document.body):
|
||||
# Skip some document parts where dashes are not converted
|
||||
if (i < j) or line.startswith("\\labelwidthstring"):
|
||||
lines = document.body
|
||||
has_literal_dashes = has_ligature_dashes = False
|
||||
i = j = 0
|
||||
while i+1 < len(lines):
|
||||
i += 1
|
||||
line = lines[i]
|
||||
# skip lines without any dashes:
|
||||
if not re.search(u"[\u2013\u2014]|\\twohyphens|\\threehyphens", line):
|
||||
continue
|
||||
if line.startswith("\\begin_inset"):
|
||||
try:
|
||||
it = line.split()[1]
|
||||
except IndexError:
|
||||
continue
|
||||
if (it in ["CommandInset", "ERT", "External", "Formula",
|
||||
"FormulaMacro", "Graphics", "IPA", "listings"]
|
||||
or line.endswith("Flex Code")):
|
||||
j = find_end_of_inset(document.body, i)
|
||||
if j == -1:
|
||||
document.warning("Malformed LyX document: Can't "
|
||||
"find end of %s inset at line %d." % (itype, i))
|
||||
continue
|
||||
if line == "\\begin_layout LyX-Code":
|
||||
j = find_end_of_layout(document.body, i)
|
||||
if j == -1:
|
||||
document.warning("Malformed LyX document: "
|
||||
"Can't find end of %s layout at line %d" % (words[1],i))
|
||||
# skip label width string (see bug 10243):
|
||||
if line.startswith("\\labelwidthstring"):
|
||||
continue
|
||||
# do not touch hyphens in some insets (cf. lyx_2_2.convert_dashes):
|
||||
try:
|
||||
value, start, end = get_containing_inset(lines, i)
|
||||
except TypeError: # no containing inset
|
||||
value, start, end = "no inset", -1, -1
|
||||
if (value.split()[0] in
|
||||
["CommandInset", "ERT", "External", "Formula",
|
||||
"FormulaMacro", "Graphics", "IPA", "listings"]
|
||||
or value == "Flex Code"):
|
||||
i = end
|
||||
continue
|
||||
try:
|
||||
layout, start, end, j = get_containing_layout(lines, i)
|
||||
except TypeError: # no (or malformed) containing layout
|
||||
document.warning("Malformed LyX document: "
|
||||
"Can't find layout at line %d" % i)
|
||||
continue
|
||||
if layout == "LyX-Code":
|
||||
i = end
|
||||
continue
|
||||
|
||||
# literal dash followed by a word or no-break space:
|
||||
if re.search(u"[\u2013\u2014]([\w\u00A0]|$)", line,
|
||||
flags=re.UNICODE):
|
||||
has_literal_dashes = True
|
||||
# ligature dash followed by word or no-break space on next line:
|
||||
if (re.search(r"(\\twohyphens|\\threehyphens)", line) and
|
||||
re.match(u"[\w\u00A0]", document.body[i+1], flags=re.UNICODE)):
|
||||
re.match(u"[\w\u00A0]", lines[i+1], flags=re.UNICODE)):
|
||||
has_ligature_dashes = True
|
||||
if has_literal_dashes and has_ligature_dashes:
|
||||
# TODO: insert a warning note in the document?
|
||||
document.warning('This document contained both literal and '
|
||||
'"ligature" dashes.\n Line breaks may have changed. '
|
||||
'See UserGuide chapter 3.9.1 for details.')
|
||||
elif has_literal_dashes:
|
||||
if has_literal_dashes and has_ligature_dashes:
|
||||
# TODO: insert a warning note in the document?
|
||||
document.warning('This document contained both literal and '
|
||||
'"ligature" dashes.\n Line breaks may have changed. '
|
||||
'See UserGuide chapter 3.9.1 for details.')
|
||||
break
|
||||
if has_literal_dashes:
|
||||
use_dash_ligatures = False
|
||||
elif has_ligature_dashes:
|
||||
use_dash_ligatures = True
|
||||
|
||||
Loading…
Reference in New Issue
Block a user