lyx_mirror/development/tools/unicodesymbols.py
Georg Baum feb7895965 Add machinery to output arbitrary unicode characters with LaTeX commands
read from a text file.

	* src/encoding.[Ch]

	(Encoding::latexChar): New, output a character to LaTeX
	(Encoding::validate): New, add needed preamble stuff for a character
	(Encodings::read): Read new unicodesymbols file
	(Encodings::isCombiningChar): New, is a character a combining char?

	* src/paragraph_pimpl.C
	(isEncoding): Delete, no longer needed
	(getEncoding): New, get the real encoding of a font
	(Paragraph::Pimpl::latexSurrogatePair): New, output a surrogate pair
	to LaTeX
	(Paragraph::Pimpl::simpleTeXBlanks): Use latexSurrogatePair if needed
	(Paragraph::Pimpl::simpleTeXSpecialChars): Ditto, and replace several
	hardcoded characters with a call of encoding.latexChar()
	(Paragraph::Pimpl::validate): replace several hardcoded characters
	with a call of encoding.validate()

	* src/support/debugstream.h
	(basic_debugstream::disable): New, disable the stream completely
	(basic_debugstream::enable): New, reenable the stream

	* src/lyx_main.[Ch]: Adjust to changes above

	* src/paragraph.C: Ditto

	* lib/unicodesymbols: New file with UCS4 -> LaTeX command mapping.
	It is far from complete yet, but contains most accents on latin
	characters.

	* lib/Makefile.am: add lib/unicodesymbols

	* development/scons/scons_manifest.py: ditto

	* development/tools/unicodesymbols.py: Helper script to update
	lib/unicodesymbols with new symbols


git-svn-id: svn://svn.lyx.org/lyx/lyx-devel/trunk@16920 a592a061-630c-0410-9148-cb99ea01b6c8
2007-01-28 21:27:45 +00:00

120 lines
2.9 KiB
Python

#! /usr/bin/env python
# -*- coding: utf-8 -*-
# file unciodesymbols.py
# This file is part of LyX, the document processor.
# Licence details can be found in the file COPYING.
# author Georg Baum
# Full author contact details are available in file CREDITS
# This script reads a unicode symbol file and completes it in the given range
import os, re, string, sys, unicodedata
def usage(prog_name):
return ("Usage: %s start stop inputfile outputfile\n" % prog_name +
"or %s start stop <inputfile >outputfile" % prog_name)
def error(message):
sys.stderr.write(message + '\n')
sys.exit(1)
def trim_eol(line):
" Remove end of line char(s)."
if line[-2:-1] == '\r':
return line[:-2]
elif line[-1:] == '\r' or line[-1:] == '\n':
return line[:-1]
else:
# file with no EOL in last line
return line
def read(input):
" Read input file and strip lineendings."
lines = list()
while 1:
line = input.readline()
if not line:
break
line = trim_eol(line)
tokens = line.split()
char = -1
if len(tokens) > 0:
if tokens[0][0:2] == "0x":
char = int(tokens[0][2:], 16)
elif tokens[0][0:3] == "#0x":
char = int(tokens[0][3:], 16)
lines.append([char, line])
return lines
def write(output, lines):
" Write output file with native lineendings."
for line in lines:
output.write(line[1] + os.linesep)
def complete(lines, start, stop):
l = 0
for i in range(start, stop):
# This catches both comments (lines[l][0] == -1) and code points less than i
while l < len(lines) and lines[l][0] < i:
print lines[l]
l = l + 1
continue
if l >= len(lines) or lines[l][0] != i:
c = unichr(i)
name = unicodedata.name(c, "")
if name != "":
if unicodedata.combining(c):
combining = "combining"
else:
combining = ""
line = [i, '#0x%04x "" "" "%s" # %s' % (i, combining, name)]
lines.insert(l, line)
print lines[l]
l = l + 1
def main(argv):
# Open files
if len(argv) == 3:
input = sys.stdin
output = sys.stdout
elif len(argv) == 5:
input = open(argv[3], 'rb')
output = open(argv[4], 'wb')
else:
error(usage(argv[0]))
if argv[1][:2] == "0x":
start = int(argv[1][2:], 16)
else:
start = int(argv[1])
if argv[2][:2] == "0x":
stop = int(argv[2][2:], 16)
else:
stop = int(argv[2])
# Do the real work
lines = read(input)
complete(lines, start, stop)
write(output, lines)
# Close files
if len(argv) == 3:
input.close()
output.close()
return 0
if __name__ == "__main__":
main(sys.argv)