summaryrefslogtreecommitdiffstats
path: root/Lib/textwrap.py
diff options
context:
space:
mode:
authorAlexandre Vassalotti <alexandre@peadrop.com>2008-05-16 00:03:33 (GMT)
committerAlexandre Vassalotti <alexandre@peadrop.com>2008-05-16 00:03:33 (GMT)
commit5f8ced2b959f306b3a1747a1f3daa4274b8c78b8 (patch)
tree75172061d8b246659f822876d22a10f7bcebeb0b /Lib/textwrap.py
parenta79e33e039530e5ca62f5dc19981cf76f4dcf824 (diff)
downloadcpython-5f8ced2b959f306b3a1747a1f3daa4274b8c78b8.zip
cpython-5f8ced2b959f306b3a1747a1f3daa4274b8c78b8.tar.gz
cpython-5f8ced2b959f306b3a1747a1f3daa4274b8c78b8.tar.bz2
Merged revisions 62998-63003,63005-63006,63009-63012,63014-63017,63019-63020,63022-63024,63026-63029,63031-63041,63043-63045,63047-63054,63056-63062 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/trunk ........ r62998 | andrew.kuchling | 2008-05-10 15:51:55 -0400 (Sat, 10 May 2008) | 7 lines #1858 from Tarek Ziade: Allow multiple repositories in .pypirc; see http://wiki.python.org/moin/EnhancedPyPI for discussion. The patch is slightly revised from Tarek's last patch: I've simplified the PyPIRCCommand.finalize_options() method to not look at sys.argv. Tests still pass. ........ r63000 | alexandre.vassalotti | 2008-05-10 15:59:16 -0400 (Sat, 10 May 2008) | 5 lines Cleaned up io._BytesIO.write(). I am amazed that the old code, for inserting null-bytes, actually worked. Who wrote that thing? Oh, it is me... doh. ........ r63002 | brett.cannon | 2008-05-10 16:52:01 -0400 (Sat, 10 May 2008) | 2 lines Revert r62998 as it broke the build (seems distutils.config is missing). ........ r63014 | andrew.kuchling | 2008-05-10 18:12:38 -0400 (Sat, 10 May 2008) | 1 line #1858: add distutils.config module ........ r63027 | brett.cannon | 2008-05-10 21:09:32 -0400 (Sat, 10 May 2008) | 2 lines Flesh out the 3.0 deprecation to suggest using the ctypes module. ........ r63028 | skip.montanaro | 2008-05-10 22:59:30 -0400 (Sat, 10 May 2008) | 4 lines Copied two versions of the example from the interactive session. Delete one. ........ r63037 | georg.brandl | 2008-05-11 03:02:17 -0400 (Sun, 11 May 2008) | 2 lines reload() takes the module itself. ........ r63038 | alexandre.vassalotti | 2008-05-11 03:06:04 -0400 (Sun, 11 May 2008) | 4 lines Added test framework for handling module renames. Factored the import guard in test_py3kwarn.TestStdlibRemovals into a context manager, namely test_support.CleanImport. ........ r63039 | georg.brandl | 2008-05-11 03:06:05 -0400 (Sun, 11 May 2008) | 2 lines #2742: ``''`` is not converted to NULL in getaddrinfo. ........ r63040 | alexandre.vassalotti | 2008-05-11 03:08:12 -0400 (Sun, 11 May 2008) | 2 lines Fixed typo in a comment of test_support.CleanImport. ........ r63041 | alexandre.vassalotti | 2008-05-11 03:10:25 -0400 (Sun, 11 May 2008) | 2 lines Removed a dead line of code. ........ r63043 | georg.brandl | 2008-05-11 04:47:53 -0400 (Sun, 11 May 2008) | 2 lines #2812: document property.getter/setter/deleter. ........ r63049 | georg.brandl | 2008-05-11 05:06:30 -0400 (Sun, 11 May 2008) | 2 lines #1153769: document PEP 237 changes to string formatting. ........ r63050 | georg.brandl | 2008-05-11 05:11:40 -0400 (Sun, 11 May 2008) | 2 lines #2809: elaborate str.split docstring a bit. ........ r63051 | georg.brandl | 2008-05-11 06:13:59 -0400 (Sun, 11 May 2008) | 2 lines Fix typo. ........ r63052 | georg.brandl | 2008-05-11 06:33:27 -0400 (Sun, 11 May 2008) | 2 lines #2709: clarification. ........ r63053 | georg.brandl | 2008-05-11 06:42:28 -0400 (Sun, 11 May 2008) | 2 lines #2659: add ``break_on_hyphens`` to TextWrapper. ........ r63057 | georg.brandl | 2008-05-11 06:59:39 -0400 (Sun, 11 May 2008) | 2 lines #2741: clarification of value range for address_family. ........ r63058 | georg.brandl | 2008-05-11 07:09:35 -0400 (Sun, 11 May 2008) | 2 lines #2452: timeout is used for all blocking operations. ........ r63059 | andrew.kuchling | 2008-05-11 09:33:56 -0400 (Sun, 11 May 2008) | 2 lines #1792: Improve performance of marshal.dumps() on large objects by increasing the size of the buffer more quickly. ........ r63060 | andrew.kuchling | 2008-05-11 10:00:00 -0400 (Sun, 11 May 2008) | 1 line #1858: re-apply patch for this, adding the missing files ........ r63061 | benjamin.peterson | 2008-05-11 10:13:25 -0400 (Sun, 11 May 2008) | 2 lines Add the "until" command to pdb ........ r63062 | georg.brandl | 2008-05-11 10:17:13 -0400 (Sun, 11 May 2008) | 2 lines Add some sentence endings. ........
Diffstat (limited to 'Lib/textwrap.py')
-rw-r--r--Lib/textwrap.py28
1 files changed, 24 insertions, 4 deletions
diff --git a/Lib/textwrap.py b/Lib/textwrap.py
index b5f87efc..6a2021d 100644
--- a/Lib/textwrap.py
+++ b/Lib/textwrap.py
@@ -55,6 +55,10 @@ class TextWrapper:
break_long_words (default: true)
Break words longer than 'width'. If false, those words will not
be broken, and some lines might be longer than 'width'.
+ break_on_hyphens (default: true)
+ Allow breaking hyphenated words. If true, wrapping will occur
+ preferably on whitespaces and right after hyphens part of
+ compound words.
drop_whitespace (default: true)
Drop leading and trailing whitespace from lines.
"""
@@ -75,11 +79,18 @@ class TextWrapper:
r'[^\s\w]*\w+[a-zA-Z]-(?=\w+[a-zA-Z])|' # hyphenated words
r'(?<=[\w\!\"\'\&\.\,\?])-{2,}(?=\w))') # em-dash
- # XXX this is not locale-aware
+ # This less funky little regex just split on recognized spaces. E.g.
+ # "Hello there -- you goof-ball, use the -b option!"
+ # splits into
+ # Hello/ /there/ /--/ /you/ /goof-ball,/ /use/ /the/ /-b/ /option!/
+ wordsep_simple_re = re.compile(r'(\s+)')
+
+ # XXX this is not locale- or charset-aware -- string.lowercase
+ # is US-ASCII only (and therefore English-only)
sentence_end_re = re.compile(r'[a-z]' # lowercase letter
r'[\.\!\?]' # sentence-ending punct.
r'[\"\']?' # optional end-of-quote
- r'\Z') # end of chunk
+ r'\Z') # end of chunk
def __init__(self,
@@ -90,7 +101,8 @@ class TextWrapper:
replace_whitespace=True,
fix_sentence_endings=False,
break_long_words=True,
- drop_whitespace=True):
+ drop_whitespace=True,
+ break_on_hyphens=True):
self.width = width
self.initial_indent = initial_indent
self.subsequent_indent = subsequent_indent
@@ -99,6 +111,7 @@ class TextWrapper:
self.fix_sentence_endings = fix_sentence_endings
self.break_long_words = break_long_words
self.drop_whitespace = drop_whitespace
+ self.break_on_hyphens = break_on_hyphens
# -- Private methods -----------------------------------------------
@@ -128,8 +141,15 @@ class TextWrapper:
breaks into the following chunks:
'Look,', ' ', 'goof-', 'ball', ' ', '--', ' ',
'use', ' ', 'the', ' ', '-b', ' ', 'option!'
+ if break_on_hyphens is True, or in:
+ 'Look,', ' ', 'goof-ball', ' ', '--', ' ',
+ 'use', ' ', 'the', ' ', '-b', ' ', option!'
+ otherwise.
"""
- chunks = self.wordsep_re.split(text)
+ if self.break_on_hyphens is True:
+ chunks = self.wordsep_re.split(text)
+ else:
+ chunks = self.wordsep_simple_re.split(text)
chunks = [c for c in chunks if c]
return chunks