1
# Copyright (C) 2011 Canonical Ltd
3
# UTextWrapper._handle_long_word, UTextWrapper._wrap_chunks,
4
# UTextWrapper._fix_sentence_endings, wrap and fill is copied from Python's
5
# textwrap module (under PSF license) and modified for support CJK.
6
# Original Copyright for these functions:
8
# Copyright (C) 1999-2001 Gregory P. Ward.
9
# Copyright (C) 2002, 2003 Python Software Foundation.
11
# Written by Greg Ward <gward@python.net>
12
# This program is free software; you can redistribute it and/or modify
13
# it under the terms of the GNU General Public License as published by
14
# the Free Software Foundation; either version 2 of the License, or
15
# (at your option) any later version.
17
# This program is distributed in the hope that it will be useful,
18
# but WITHOUT ANY WARRANTY; without even the implied warranty of
19
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
20
# GNU General Public License for more details.
22
# You should have received a copy of the GNU General Public License
23
# along with this program; if not, write to the Free Software
24
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
26
from __future__ import absolute_import
30
from unicodedata import east_asian_width as _eawidth
32
from bzrlib import osutils
34
__all__ = ["UTextWrapper", "fill", "wrap"]
36
class UTextWrapper(textwrap.TextWrapper):
38
Extend TextWrapper for Unicode.
40
This textwrapper handles east asian double width and split word
41
even if !break_long_words when word contains double width
44
:param ambiguous_width: (keyword argument) width for character when
45
unicodedata.east_asian_width(c) == 'A'
49
* expand_tabs doesn't fixed. It uses len() for calculating width
50
of string on left of TAB.
51
* Handles one codeunit as a single character having 1 or 2 width.
52
This is not correct when there are surrogate pairs, combined
53
characters or zero-width characters.
54
* Treats all asian character are line breakable. But it is not
55
true because line breaking is prohibited around some characters.
56
(For example, breaking before punctation mark is prohibited.)
57
See UAX # 14 "UNICODE LINE BREAKING ALGORITHM"
60
def __init__(self, width=None, **kwargs):
62
width = (osutils.terminal_width() or
63
osutils.default_terminal_width) - 1
65
ambi_width = kwargs.pop('ambiguous_width', 1)
67
self._east_asian_doublewidth = 'FW'
69
self._east_asian_doublewidth = 'FWA'
71
raise ValueError("ambiguous_width should be 1 or 2")
73
# No drop_whitespace param before Python 2.6 it was always dropped
74
if sys.version_info < (2, 6):
75
self.drop_whitespace = kwargs.pop("drop_whitespace", True)
76
if not self.drop_whitespace:
77
raise ValueError("TextWrapper version must drop whitespace")
78
textwrap.TextWrapper.__init__(self, width, **kwargs)
80
def _unicode_char_width(self, uc):
81
"""Return width of character `uc`.
83
:param: uc Single unicode character.
85
# 'A' means width of the character is not be able to determine.
86
# We assume that it's width is 2 because longer wrap may over
87
# terminal width but shorter wrap may be acceptable.
88
return (_eawidth(uc) in self._east_asian_doublewidth and 2) or 1
91
"""Returns width for s.
93
When s is unicode, take care of east asian width.
94
When s is bytes, treat all byte is single width character.
96
charwidth = self._unicode_char_width
97
return sum(charwidth(c) for c in s)
99
def _cut(self, s, width):
100
"""Returns head and rest of s. (head+rest == s)
102
Head is large as long as _width(head) <= width.
105
charwidth = self._unicode_char_width
106
for pos, c in enumerate(s):
109
return s[:pos], s[pos:]
112
def _fix_sentence_endings(self, chunks):
113
"""_fix_sentence_endings(chunks : [string])
115
Correct for sentence endings buried in 'chunks'. Eg. when the
116
original text contains "... foo.\nBar ...", munge_whitespace()
117
and split() will convert that to [..., "foo.", " ", "Bar", ...]
118
which has one too few spaces; this method simply changes the one
121
Note: This function is copied from textwrap.TextWrap and modified
122
to use unicode always.
126
patsearch = self.sentence_end_re.search
128
if chunks[i+1] == u" " and patsearch(chunks[i]):
134
def _handle_long_word(self, chunks, cur_line, cur_len, width):
135
# Figure out when indent is larger than the specified width, and make
136
# sure at least one character is stripped off on every pass
138
space_left = chunks[-1] and self._width(chunks[-1][0]) or 1
140
space_left = width - cur_len
142
# If we're allowed to break long words, then do so: put as much
143
# of the next chunk onto the current line as will fit.
144
if self.break_long_words:
145
head, rest = self._cut(chunks[-1], space_left)
146
cur_line.append(head)
152
# Otherwise, we have to preserve the long word intact. Only add
153
# it to the current line if there's nothing already there --
154
# that minimizes how much we violate the width constraint.
156
cur_line.append(chunks.pop())
158
# If we're not allowed to break long words, and there's already
159
# text on the current line, do nothing. Next time through the
160
# main loop of _wrap_chunks(), we'll wind up here again, but
161
# cur_len will be zero, so the next line will be entirely
162
# devoted to the long word that we can't handle right now.
164
def _wrap_chunks(self, chunks):
167
raise ValueError("invalid width %r (must be > 0)" % self.width)
169
# Arrange in reverse order so items can be efficiently popped
170
# from a stack of chucks.
175
# Start the list of chunks that will make up the current line.
176
# cur_len is just the length of all the chunks in cur_line.
180
# Figure out which static string will prefix this line.
182
indent = self.subsequent_indent
184
indent = self.initial_indent
186
# Maximum width for this line.
187
width = self.width - len(indent)
189
# First chunk on line is whitespace -- drop it, unless this
190
# is the very beginning of the text (ie. no lines started yet).
191
if self.drop_whitespace and chunks[-1].strip() == '' and lines:
195
# Use _width instead of len for east asian width
196
l = self._width(chunks[-1])
198
# Can at least squeeze this chunk onto the current line.
199
if cur_len + l <= width:
200
cur_line.append(chunks.pop())
203
# Nope, this line is full.
207
# The current line is full, and the next chunk is too big to
208
# fit on *any* line (not just this one).
209
if chunks and self._width(chunks[-1]) > width:
210
self._handle_long_word(chunks, cur_line, cur_len, width)
212
# If the last chunk on this line is all whitespace, drop it.
213
if self.drop_whitespace and cur_line and not cur_line[-1].strip():
216
# Convert current line back to a string and store it in list
217
# of all lines (return value).
219
lines.append(indent + u''.join(cur_line))
223
def _split(self, text):
224
chunks = textwrap.TextWrapper._split(self, unicode(text))
225
cjk_split_chunks = []
228
for pos, char in enumerate(chunk):
229
if self._unicode_char_width(char) == 2:
231
cjk_split_chunks.append(chunk[prev_pos:pos])
232
cjk_split_chunks.append(char)
234
if prev_pos < len(chunk):
235
cjk_split_chunks.append(chunk[prev_pos:])
236
return cjk_split_chunks
238
def wrap(self, text):
239
# ensure text is unicode
240
return textwrap.TextWrapper.wrap(self, unicode(text))
242
# -- Convenience interface ---------------------------------------------
244
def wrap(text, width=None, **kwargs):
245
"""Wrap a single paragraph of text, returning a list of wrapped lines.
247
Reformat the single paragraph in 'text' so it fits in lines of no
248
more than 'width' columns, and return a list of wrapped lines. By
249
default, tabs in 'text' are expanded with string.expandtabs(), and
250
all other whitespace characters (including newline) are converted to
251
space. See TextWrapper class for available keyword args to customize
254
return UTextWrapper(width=width, **kwargs).wrap(text)
256
def fill(text, width=None, **kwargs):
257
"""Fill a single paragraph of text, returning a new string.
259
Reformat the single paragraph in 'text' to fit in lines of no more
260
than 'width' columns, and return a new string containing the entire
261
wrapped paragraph. As with wrap(), tabs are expanded and other
262
whitespace characters converted to space. See TextWrapper class for
263
available keyword args to customize wrapping behaviour.
265
return UTextWrapper(width=width, **kwargs).fill(text)