2002-08-22 15:37:50 -03:00
|
|
|
#
|
2004-06-02 22:59:41 -03:00
|
|
|
# Test suite for the textwrap module.
|
2002-08-22 15:37:50 -03:00
|
|
|
#
|
|
|
|
# Original tests written by Greg Ward <gward@python.net>.
|
|
|
|
# Converted to PyUnit by Peter Hansen <peter@engcorp.com>.
|
|
|
|
# Currently maintained by Greg Ward.
|
|
|
|
#
|
|
|
|
# $Id$
|
|
|
|
#
|
|
|
|
|
2002-08-22 15:11:10 -03:00
|
|
|
import unittest
|
2002-08-22 15:35:49 -03:00
|
|
|
from test import test_support
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2003-05-07 22:58:26 -03:00
|
|
|
from textwrap import TextWrapper, wrap, fill, dedent
|
2002-08-22 15:11:10 -03:00
|
|
|
|
|
|
|
|
2002-08-22 15:57:26 -03:00
|
|
|
class BaseTestCase(unittest.TestCase):
|
2002-08-22 15:11:10 -03:00
|
|
|
'''Parent class with utility methods for textwrap tests.'''
|
|
|
|
|
|
|
|
def show(self, textin):
|
|
|
|
if isinstance(textin, list):
|
|
|
|
result = []
|
|
|
|
for i in range(len(textin)):
|
|
|
|
result.append(" %d: %r" % (i, textin[i]))
|
|
|
|
result = '\n'.join(result)
|
2003-09-17 02:50:59 -03:00
|
|
|
elif isinstance(textin, basestring):
|
2002-08-22 15:11:10 -03:00
|
|
|
result = " %s\n" % repr(textin)
|
|
|
|
return result
|
|
|
|
|
|
|
|
|
|
|
|
def check(self, result, expect):
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(result, expect,
|
2002-08-22 16:47:27 -03:00
|
|
|
'expected:\n%s\nbut got:\n%s' % (
|
|
|
|
self.show(expect), self.show(result)))
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-10-02 12:47:32 -03:00
|
|
|
def check_wrap(self, text, width, expect, **kwargs):
|
|
|
|
result = wrap(text, width, **kwargs)
|
2002-08-22 15:55:38 -03:00
|
|
|
self.check(result, expect)
|
|
|
|
|
2002-10-31 12:11:18 -04:00
|
|
|
def check_split(self, text, expect):
|
|
|
|
result = self.wrapper._split(text)
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(result, expect,
|
|
|
|
"\nexpected %r\n"
|
|
|
|
"but got %r" % (expect, result))
|
2002-08-22 18:16:25 -03:00
|
|
|
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 15:57:26 -03:00
|
|
|
class WrapTestCase(BaseTestCase):
|
2002-08-22 15:11:10 -03:00
|
|
|
|
|
|
|
def setUp(self):
|
2004-05-12 22:53:10 -03:00
|
|
|
self.wrapper = TextWrapper(width=45)
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 15:35:49 -03:00
|
|
|
def test_simple(self):
|
2002-08-22 17:13:47 -03:00
|
|
|
# Simple case: just words, spaces, and a bit of punctuation
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 15:55:38 -03:00
|
|
|
text = "Hello there, how are you this fine day? I'm glad to hear it!"
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 15:55:38 -03:00
|
|
|
self.check_wrap(text, 12,
|
|
|
|
["Hello there,",
|
|
|
|
"how are you",
|
|
|
|
"this fine",
|
|
|
|
"day? I'm",
|
|
|
|
"glad to hear",
|
|
|
|
"it!"])
|
|
|
|
self.check_wrap(text, 42,
|
|
|
|
["Hello there, how are you this fine day?",
|
|
|
|
"I'm glad to hear it!"])
|
|
|
|
self.check_wrap(text, 80, [text])
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2012-09-08 14:42:01 -03:00
|
|
|
def test_empty_string(self):
|
|
|
|
# Check that wrapping the empty string returns an empty list.
|
|
|
|
self.check_wrap("", 6, [])
|
|
|
|
self.check_wrap("", 6, [], drop_whitespace=False)
|
|
|
|
|
|
|
|
def test_empty_string_with_initial_indent(self):
|
|
|
|
# Check that the empty string is not indented.
|
|
|
|
self.check_wrap("", 6, [], initial_indent="++")
|
|
|
|
self.check_wrap("", 6, [], initial_indent="++", drop_whitespace=False)
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 15:35:49 -03:00
|
|
|
def test_whitespace(self):
|
2002-08-22 17:13:47 -03:00
|
|
|
# Whitespace munging and end-of-sentence detection
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 15:55:38 -03:00
|
|
|
text = """\
|
2002-08-22 15:11:10 -03:00
|
|
|
This is a paragraph that already has
|
|
|
|
line breaks. But some of its lines are much longer than the others,
|
|
|
|
so it needs to be wrapped.
|
|
|
|
Some lines are \ttabbed too.
|
|
|
|
What a mess!
|
|
|
|
"""
|
|
|
|
|
2002-08-22 15:55:38 -03:00
|
|
|
expect = ["This is a paragraph that already has line",
|
|
|
|
"breaks. But some of its lines are much",
|
|
|
|
"longer than the others, so it needs to be",
|
|
|
|
"wrapped. Some lines are tabbed too. What a",
|
|
|
|
"mess!"]
|
|
|
|
|
2004-05-12 22:53:10 -03:00
|
|
|
wrapper = TextWrapper(45, fix_sentence_endings=True)
|
|
|
|
result = wrapper.wrap(text)
|
2002-08-22 15:11:10 -03:00
|
|
|
self.check(result, expect)
|
|
|
|
|
2004-05-12 22:53:10 -03:00
|
|
|
result = wrapper.fill(text)
|
2002-08-22 15:11:10 -03:00
|
|
|
self.check(result, '\n'.join(expect))
|
|
|
|
|
2004-05-12 22:53:10 -03:00
|
|
|
def test_fix_sentence_endings(self):
|
|
|
|
wrapper = TextWrapper(60, fix_sentence_endings=True)
|
|
|
|
|
|
|
|
# SF #847346: ensure that fix_sentence_endings=True does the
|
|
|
|
# right thing even on input short enough that it doesn't need to
|
|
|
|
# be wrapped.
|
|
|
|
text = "A short line. Note the single space."
|
|
|
|
expect = ["A short line. Note the single space."]
|
|
|
|
self.check(wrapper.wrap(text), expect)
|
|
|
|
|
|
|
|
# Test some of the hairy end cases that _fix_sentence_endings()
|
|
|
|
# is supposed to handle (the easy stuff is tested in
|
|
|
|
# test_whitespace() above).
|
|
|
|
text = "Well, Doctor? What do you think?"
|
|
|
|
expect = ["Well, Doctor? What do you think?"]
|
|
|
|
self.check(wrapper.wrap(text), expect)
|
|
|
|
|
|
|
|
text = "Well, Doctor?\nWhat do you think?"
|
|
|
|
self.check(wrapper.wrap(text), expect)
|
|
|
|
|
|
|
|
text = 'I say, chaps! Anyone for "tennis?"\nHmmph!'
|
|
|
|
expect = ['I say, chaps! Anyone for "tennis?" Hmmph!']
|
|
|
|
self.check(wrapper.wrap(text), expect)
|
|
|
|
|
|
|
|
wrapper.width = 20
|
|
|
|
expect = ['I say, chaps!', 'Anyone for "tennis?"', 'Hmmph!']
|
|
|
|
self.check(wrapper.wrap(text), expect)
|
|
|
|
|
|
|
|
text = 'And she said, "Go to hell!"\nCan you believe that?'
|
|
|
|
expect = ['And she said, "Go to',
|
|
|
|
'hell!" Can you',
|
|
|
|
'believe that?']
|
|
|
|
self.check(wrapper.wrap(text), expect)
|
|
|
|
|
|
|
|
wrapper.width = 60
|
|
|
|
expect = ['And she said, "Go to hell!" Can you believe that?']
|
|
|
|
self.check(wrapper.wrap(text), expect)
|
2004-07-08 01:22:35 -03:00
|
|
|
|
2008-04-25 13:59:09 -03:00
|
|
|
text = 'File stdio.h is nice.'
|
|
|
|
expect = ['File stdio.h is nice.']
|
|
|
|
self.check(wrapper.wrap(text), expect)
|
|
|
|
|
2002-08-22 15:35:49 -03:00
|
|
|
def test_wrap_short(self):
|
2002-08-22 17:13:47 -03:00
|
|
|
# Wrapping to make short lines longer
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 15:55:38 -03:00
|
|
|
text = "This is a\nshort paragraph."
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 15:55:38 -03:00
|
|
|
self.check_wrap(text, 20, ["This is a short",
|
|
|
|
"paragraph."])
|
|
|
|
self.check_wrap(text, 40, ["This is a short paragraph."])
|
2002-08-22 15:11:10 -03:00
|
|
|
|
|
|
|
|
2002-10-02 12:47:32 -03:00
|
|
|
def test_wrap_short_1line(self):
|
|
|
|
# Test endcases
|
|
|
|
|
|
|
|
text = "This is a short line."
|
|
|
|
|
|
|
|
self.check_wrap(text, 30, ["This is a short line."])
|
|
|
|
self.check_wrap(text, 30, ["(1) This is a short line."],
|
|
|
|
initial_indent="(1) ")
|
|
|
|
|
|
|
|
|
2002-08-22 15:35:49 -03:00
|
|
|
def test_hyphenated(self):
|
2002-08-22 17:13:47 -03:00
|
|
|
# Test breaking hyphenated words
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 15:55:38 -03:00
|
|
|
text = ("this-is-a-useful-feature-for-"
|
|
|
|
"reformatting-posts-from-tim-peters'ly")
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 15:55:38 -03:00
|
|
|
self.check_wrap(text, 40,
|
|
|
|
["this-is-a-useful-feature-for-",
|
|
|
|
"reformatting-posts-from-tim-peters'ly"])
|
|
|
|
self.check_wrap(text, 41,
|
|
|
|
["this-is-a-useful-feature-for-",
|
|
|
|
"reformatting-posts-from-tim-peters'ly"])
|
|
|
|
self.check_wrap(text, 42,
|
|
|
|
["this-is-a-useful-feature-for-reformatting-",
|
|
|
|
"posts-from-tim-peters'ly"])
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2005-03-04 22:53:17 -04:00
|
|
|
def test_hyphenated_numbers(self):
|
|
|
|
# Test that hyphenated numbers (eg. dates) are not broken like words.
|
|
|
|
text = ("Python 1.0.0 was released on 1994-01-26. Python 1.0.1 was\n"
|
|
|
|
"released on 1994-02-15.")
|
|
|
|
|
2008-12-13 19:12:30 -04:00
|
|
|
self.check_wrap(text, 35, ['Python 1.0.0 was released on',
|
2005-03-04 22:53:17 -04:00
|
|
|
'1994-01-26. Python 1.0.1 was',
|
|
|
|
'released on 1994-02-15.'])
|
|
|
|
self.check_wrap(text, 40, ['Python 1.0.0 was released on 1994-01-26.',
|
|
|
|
'Python 1.0.1 was released on 1994-02-15.'])
|
|
|
|
|
|
|
|
text = "I do all my shopping at 7-11."
|
|
|
|
self.check_wrap(text, 25, ["I do all my shopping at",
|
|
|
|
"7-11."])
|
|
|
|
self.check_wrap(text, 27, ["I do all my shopping at",
|
|
|
|
"7-11."])
|
|
|
|
self.check_wrap(text, 29, ["I do all my shopping at 7-11."])
|
|
|
|
|
2002-08-22 16:47:27 -03:00
|
|
|
def test_em_dash(self):
|
2002-08-22 17:13:47 -03:00
|
|
|
# Test text with em-dashes
|
2002-08-22 16:47:27 -03:00
|
|
|
text = "Em-dashes should be written -- thus."
|
|
|
|
self.check_wrap(text, 25,
|
|
|
|
["Em-dashes should be",
|
|
|
|
"written -- thus."])
|
|
|
|
|
|
|
|
# Probe the boundaries of the properly written em-dash,
|
|
|
|
# ie. " -- ".
|
|
|
|
self.check_wrap(text, 29,
|
|
|
|
["Em-dashes should be written",
|
|
|
|
"-- thus."])
|
|
|
|
expect = ["Em-dashes should be written --",
|
|
|
|
"thus."]
|
|
|
|
self.check_wrap(text, 30, expect)
|
|
|
|
self.check_wrap(text, 35, expect)
|
|
|
|
self.check_wrap(text, 36,
|
|
|
|
["Em-dashes should be written -- thus."])
|
2002-08-22 17:13:47 -03:00
|
|
|
|
2002-08-22 16:47:27 -03:00
|
|
|
# The improperly written em-dash is handled too, because
|
|
|
|
# it's adjacent to non-whitespace on both sides.
|
|
|
|
text = "You can also do--this or even---this."
|
|
|
|
expect = ["You can also do",
|
|
|
|
"--this or even",
|
|
|
|
"---this."]
|
|
|
|
self.check_wrap(text, 15, expect)
|
|
|
|
self.check_wrap(text, 16, expect)
|
|
|
|
expect = ["You can also do--",
|
|
|
|
"this or even---",
|
|
|
|
"this."]
|
|
|
|
self.check_wrap(text, 17, expect)
|
|
|
|
self.check_wrap(text, 19, expect)
|
|
|
|
expect = ["You can also do--this or even",
|
|
|
|
"---this."]
|
|
|
|
self.check_wrap(text, 29, expect)
|
|
|
|
self.check_wrap(text, 31, expect)
|
|
|
|
expect = ["You can also do--this or even---",
|
|
|
|
"this."]
|
|
|
|
self.check_wrap(text, 32, expect)
|
|
|
|
self.check_wrap(text, 35, expect)
|
|
|
|
|
|
|
|
# All of the above behaviour could be deduced by probing the
|
|
|
|
# _split() method.
|
|
|
|
text = "Here's an -- em-dash and--here's another---and another!"
|
|
|
|
expect = ["Here's", " ", "an", " ", "--", " ", "em-", "dash", " ",
|
|
|
|
"and", "--", "here's", " ", "another", "---",
|
|
|
|
"and", " ", "another!"]
|
2002-10-31 12:11:18 -04:00
|
|
|
self.check_split(text, expect)
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 18:27:05 -03:00
|
|
|
text = "and then--bam!--he was gone"
|
|
|
|
expect = ["and", " ", "then", "--", "bam!", "--",
|
|
|
|
"he", " ", "was", " ", "gone"]
|
2002-10-31 12:11:18 -04:00
|
|
|
self.check_split(text, expect)
|
2002-08-22 18:27:05 -03:00
|
|
|
|
|
|
|
|
2002-08-22 18:10:07 -03:00
|
|
|
def test_unix_options (self):
|
|
|
|
# Test that Unix-style command-line options are wrapped correctly.
|
|
|
|
# Both Optik (OptionParser) and Docutils rely on this behaviour!
|
|
|
|
|
|
|
|
text = "You should use the -n option, or --dry-run in its long form."
|
|
|
|
self.check_wrap(text, 20,
|
|
|
|
["You should use the",
|
|
|
|
"-n option, or --dry-",
|
|
|
|
"run in its long",
|
|
|
|
"form."])
|
|
|
|
self.check_wrap(text, 21,
|
|
|
|
["You should use the -n",
|
|
|
|
"option, or --dry-run",
|
|
|
|
"in its long form."])
|
|
|
|
expect = ["You should use the -n option, or",
|
|
|
|
"--dry-run in its long form."]
|
|
|
|
self.check_wrap(text, 32, expect)
|
|
|
|
self.check_wrap(text, 34, expect)
|
|
|
|
self.check_wrap(text, 35, expect)
|
|
|
|
self.check_wrap(text, 38, expect)
|
|
|
|
expect = ["You should use the -n option, or --dry-",
|
|
|
|
"run in its long form."]
|
|
|
|
self.check_wrap(text, 39, expect)
|
|
|
|
self.check_wrap(text, 41, expect)
|
|
|
|
expect = ["You should use the -n option, or --dry-run",
|
|
|
|
"in its long form."]
|
|
|
|
self.check_wrap(text, 42, expect)
|
|
|
|
|
2002-08-22 18:12:54 -03:00
|
|
|
# Again, all of the above can be deduced from _split().
|
|
|
|
text = "the -n option, or --dry-run or --dryrun"
|
|
|
|
expect = ["the", " ", "-n", " ", "option,", " ", "or", " ",
|
|
|
|
"--dry-", "run", " ", "or", " ", "--dryrun"]
|
2002-10-31 12:11:18 -04:00
|
|
|
self.check_split(text, expect)
|
|
|
|
|
|
|
|
def test_funky_hyphens (self):
|
|
|
|
# Screwy edge cases cooked up by David Goodger. All reported
|
|
|
|
# in SF bug #596434.
|
|
|
|
self.check_split("what the--hey!", ["what", " ", "the", "--", "hey!"])
|
|
|
|
self.check_split("what the--", ["what", " ", "the--"])
|
|
|
|
self.check_split("what the--.", ["what", " ", "the--."])
|
|
|
|
self.check_split("--text--.", ["--text--."])
|
|
|
|
|
2003-05-06 22:19:22 -03:00
|
|
|
# When I first read bug #596434, this is what I thought David
|
|
|
|
# was talking about. I was wrong; these have always worked
|
|
|
|
# fine. The real problem is tested in test_funky_parens()
|
|
|
|
# below...
|
2002-10-31 12:11:18 -04:00
|
|
|
self.check_split("--option", ["--option"])
|
|
|
|
self.check_split("--option-opt", ["--option-", "opt"])
|
2003-05-06 22:19:22 -03:00
|
|
|
self.check_split("foo --option-opt bar",
|
|
|
|
["foo", " ", "--option-", "opt", " ", "bar"])
|
|
|
|
|
2004-06-02 22:59:41 -03:00
|
|
|
def test_punct_hyphens(self):
|
|
|
|
# Oh bother, SF #965425 found another problem with hyphens --
|
|
|
|
# hyphenated words in single quotes weren't handled correctly.
|
|
|
|
# In fact, the bug is that *any* punctuation around a hyphenated
|
|
|
|
# word was handled incorrectly, except for a leading "--", which
|
|
|
|
# was special-cased for Optik and Docutils. So test a variety
|
|
|
|
# of styles of punctuation around a hyphenated word.
|
|
|
|
# (Actually this is based on an Optik bug report, #813077).
|
|
|
|
self.check_split("the 'wibble-wobble' widget",
|
|
|
|
['the', ' ', "'wibble-", "wobble'", ' ', 'widget'])
|
|
|
|
self.check_split('the "wibble-wobble" widget',
|
|
|
|
['the', ' ', '"wibble-', 'wobble"', ' ', 'widget'])
|
|
|
|
self.check_split("the (wibble-wobble) widget",
|
|
|
|
['the', ' ', "(wibble-", "wobble)", ' ', 'widget'])
|
|
|
|
self.check_split("the ['wibble-wobble'] widget",
|
|
|
|
['the', ' ', "['wibble-", "wobble']", ' ', 'widget'])
|
|
|
|
|
2003-05-06 22:19:22 -03:00
|
|
|
def test_funky_parens (self):
|
|
|
|
# Second part of SF bug #596434: long option strings inside
|
|
|
|
# parentheses.
|
|
|
|
self.check_split("foo (--option) bar",
|
|
|
|
["foo", " ", "(--option)", " ", "bar"])
|
|
|
|
|
|
|
|
# Related stuff -- make sure parens work in simpler contexts.
|
|
|
|
self.check_split("foo (bar) baz",
|
|
|
|
["foo", " ", "(bar)", " ", "baz"])
|
|
|
|
self.check_split("blah (ding dong), wubba",
|
|
|
|
["blah", " ", "(ding", " ", "dong),",
|
|
|
|
" ", "wubba"])
|
2002-08-22 18:12:54 -03:00
|
|
|
|
2012-09-08 14:42:01 -03:00
|
|
|
def test_drop_whitespace_false(self):
|
|
|
|
# Check that drop_whitespace=False preserves whitespace.
|
|
|
|
# SF patch #1581073
|
|
|
|
text = " This is a sentence with much whitespace."
|
|
|
|
self.check_wrap(text, 10,
|
|
|
|
[" This is a", " ", "sentence ",
|
|
|
|
"with ", "much white", "space."],
|
|
|
|
drop_whitespace=False)
|
|
|
|
|
|
|
|
def test_drop_whitespace_false_whitespace_only(self):
|
|
|
|
# Check that drop_whitespace=False preserves a whitespace-only string.
|
|
|
|
self.check_wrap(" ", 6, [" "], drop_whitespace=False)
|
|
|
|
|
|
|
|
def test_drop_whitespace_false_whitespace_only_with_indent(self):
|
|
|
|
# Check that a whitespace-only string gets indented (when
|
|
|
|
# drop_whitespace is False).
|
|
|
|
self.check_wrap(" ", 6, [" "], drop_whitespace=False,
|
|
|
|
initial_indent=" ")
|
|
|
|
|
|
|
|
def test_drop_whitespace_whitespace_only(self):
|
|
|
|
# Check drop_whitespace on a whitespace-only string.
|
|
|
|
self.check_wrap(" ", 6, [])
|
|
|
|
|
|
|
|
def test_drop_whitespace_leading_whitespace(self):
|
|
|
|
# Check that drop_whitespace does not drop leading whitespace (if
|
|
|
|
# followed by non-whitespace).
|
2002-12-09 12:27:15 -04:00
|
|
|
# SF bug #622849 reported inconsistent handling of leading
|
|
|
|
# whitespace; let's test that a bit, shall we?
|
|
|
|
text = " This is a sentence with leading whitespace."
|
|
|
|
self.check_wrap(text, 50,
|
|
|
|
[" This is a sentence with leading whitespace."])
|
|
|
|
self.check_wrap(text, 30,
|
|
|
|
[" This is a sentence with", "leading whitespace."])
|
|
|
|
|
2012-09-08 14:42:01 -03:00
|
|
|
def test_drop_whitespace_whitespace_line(self):
|
|
|
|
# Check that drop_whitespace skips the whole line if a non-leading
|
|
|
|
# line consists only of whitespace.
|
|
|
|
text = "abcd efgh"
|
|
|
|
# Include the result for drop_whitespace=False for comparison.
|
|
|
|
self.check_wrap(text, 6, ["abcd", " ", "efgh"],
|
2007-03-13 15:15:41 -03:00
|
|
|
drop_whitespace=False)
|
2012-09-08 14:42:01 -03:00
|
|
|
self.check_wrap(text, 6, ["abcd", "efgh"])
|
|
|
|
|
|
|
|
def test_drop_whitespace_whitespace_only_with_indent(self):
|
|
|
|
# Check that initial_indent is not applied to a whitespace-only
|
|
|
|
# string. This checks a special case of the fact that dropping
|
|
|
|
# whitespace occurs before indenting.
|
|
|
|
self.check_wrap(" ", 6, [], initial_indent="++")
|
|
|
|
|
|
|
|
def test_drop_whitespace_whitespace_indent(self):
|
|
|
|
# Check that drop_whitespace does not drop whitespace indents.
|
|
|
|
# This checks a special case of the fact that dropping whitespace
|
|
|
|
# occurs before indenting.
|
|
|
|
self.check_wrap("abcd efgh", 6, [" abcd", " efgh"],
|
|
|
|
initial_indent=" ", subsequent_indent=" ")
|
2007-03-13 15:15:41 -03:00
|
|
|
|
2005-08-03 14:09:04 -03:00
|
|
|
if test_support.have_unicode:
|
|
|
|
def test_unicode(self):
|
|
|
|
# *Very* simple test of wrapping Unicode strings. I'm sure
|
|
|
|
# there's more to it than this, but let's at least make
|
|
|
|
# sure textwrap doesn't crash on Unicode input!
|
|
|
|
text = u"Hello there, how are you today?"
|
|
|
|
self.check_wrap(text, 50, [u"Hello there, how are you today?"])
|
|
|
|
self.check_wrap(text, 20, [u"Hello there, how are", "you today?"])
|
|
|
|
olines = self.wrapper.wrap(text)
|
2010-01-24 12:58:36 -04:00
|
|
|
self.assertIsInstance(olines, list)
|
|
|
|
self.assertIsInstance(olines[0], unicode)
|
2005-08-03 14:09:04 -03:00
|
|
|
otext = self.wrapper.fill(text)
|
2010-01-24 12:58:36 -04:00
|
|
|
self.assertIsInstance(otext, unicode)
|
2002-12-09 12:32:41 -04:00
|
|
|
|
2008-12-13 19:12:30 -04:00
|
|
|
def test_no_split_at_umlaut(self):
|
|
|
|
text = u"Die Empf\xe4nger-Auswahl"
|
|
|
|
self.check_wrap(text, 13, [u"Die", u"Empf\xe4nger-", u"Auswahl"])
|
|
|
|
|
|
|
|
def test_umlaut_followed_by_dash(self):
|
|
|
|
text = u"aa \xe4\xe4-\xe4\xe4"
|
|
|
|
self.check_wrap(text, 7, [u"aa \xe4\xe4-", u"\xe4\xe4"])
|
|
|
|
|
2002-08-22 15:11:10 -03:00
|
|
|
def test_split(self):
|
2002-08-22 17:13:47 -03:00
|
|
|
# Ensure that the standard _split() method works as advertised
|
|
|
|
# in the comments
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 15:55:38 -03:00
|
|
|
text = "Hello there -- you goof-ball, use the -b option!"
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 15:55:38 -03:00
|
|
|
result = self.wrapper._split(text)
|
2002-08-22 15:11:10 -03:00
|
|
|
self.check(result,
|
|
|
|
["Hello", " ", "there", " ", "--", " ", "you", " ", "goof-",
|
|
|
|
"ball,", " ", "use", " ", "the", " ", "-b", " ", "option!"])
|
|
|
|
|
2008-05-11 07:42:28 -03:00
|
|
|
def test_break_on_hyphens(self):
|
|
|
|
# Ensure that the break_on_hyphens attributes work
|
|
|
|
text = "yaba daba-doo"
|
|
|
|
self.check_wrap(text, 10, ["yaba daba-", "doo"],
|
|
|
|
break_on_hyphens=True)
|
|
|
|
self.check_wrap(text, 10, ["yaba", "daba-doo"],
|
|
|
|
break_on_hyphens=False)
|
|
|
|
|
2003-05-06 21:54:42 -03:00
|
|
|
def test_bad_width(self):
|
|
|
|
# Ensure that width <= 0 is caught.
|
|
|
|
text = "Whatever, it doesn't matter."
|
|
|
|
self.assertRaises(ValueError, wrap, text, 0)
|
|
|
|
self.assertRaises(ValueError, wrap, text, -1)
|
|
|
|
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 16:02:37 -03:00
|
|
|
class LongWordTestCase (BaseTestCase):
|
|
|
|
def setUp(self):
|
|
|
|
self.wrapper = TextWrapper()
|
2002-12-09 12:27:15 -04:00
|
|
|
self.text = '''\
|
2002-08-22 15:11:10 -03:00
|
|
|
Did you say "supercalifragilisticexpialidocious?"
|
|
|
|
How *do* you spell that odd word, anyways?
|
|
|
|
'''
|
2002-08-22 16:02:37 -03:00
|
|
|
|
|
|
|
def test_break_long(self):
|
2002-08-22 17:13:47 -03:00
|
|
|
# Wrap text with long words and lots of punctuation
|
2002-08-22 16:02:37 -03:00
|
|
|
|
|
|
|
self.check_wrap(self.text, 30,
|
2002-08-22 15:55:38 -03:00
|
|
|
['Did you say "supercalifragilis',
|
|
|
|
'ticexpialidocious?" How *do*',
|
|
|
|
'you spell that odd word,',
|
|
|
|
'anyways?'])
|
2002-08-22 16:02:37 -03:00
|
|
|
self.check_wrap(self.text, 50,
|
2002-08-22 15:55:38 -03:00
|
|
|
['Did you say "supercalifragilisticexpialidocious?"',
|
|
|
|
'How *do* you spell that odd word, anyways?'])
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2003-08-30 11:43:55 -03:00
|
|
|
# SF bug 797650. Prevent an infinite loop by making sure that at
|
|
|
|
# least one character gets split off on every pass.
|
|
|
|
self.check_wrap('-'*10+'hello', 10,
|
|
|
|
['----------',
|
|
|
|
' h',
|
|
|
|
' e',
|
|
|
|
' l',
|
|
|
|
' l',
|
|
|
|
' o'],
|
|
|
|
subsequent_indent = ' '*15)
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2008-01-19 15:48:19 -04:00
|
|
|
# bug 1146. Prevent a long word to be wrongly wrapped when the
|
|
|
|
# preceding word is exactly one character shorter than the width
|
|
|
|
self.check_wrap(self.text, 12,
|
|
|
|
['Did you say ',
|
|
|
|
'"supercalifr',
|
|
|
|
'agilisticexp',
|
|
|
|
'ialidocious?',
|
|
|
|
'" How *do*',
|
|
|
|
'you spell',
|
|
|
|
'that odd',
|
|
|
|
'word,',
|
|
|
|
'anyways?'])
|
|
|
|
|
2002-08-22 17:13:47 -03:00
|
|
|
def test_nobreak_long(self):
|
|
|
|
# Test with break_long_words disabled
|
2002-08-22 15:11:10 -03:00
|
|
|
self.wrapper.break_long_words = 0
|
|
|
|
self.wrapper.width = 30
|
2002-08-22 15:55:38 -03:00
|
|
|
expect = ['Did you say',
|
|
|
|
'"supercalifragilisticexpialidocious?"',
|
|
|
|
'How *do* you spell that odd',
|
|
|
|
'word, anyways?'
|
2002-08-22 17:13:47 -03:00
|
|
|
]
|
2002-08-22 16:02:37 -03:00
|
|
|
result = self.wrapper.wrap(self.text)
|
2002-08-22 15:11:10 -03:00
|
|
|
self.check(result, expect)
|
|
|
|
|
|
|
|
# Same thing with kwargs passed to standalone wrap() function.
|
2002-08-22 16:02:37 -03:00
|
|
|
result = wrap(self.text, width=30, break_long_words=0)
|
2002-08-22 15:11:10 -03:00
|
|
|
self.check(result, expect)
|
|
|
|
|
|
|
|
|
2002-08-22 15:57:26 -03:00
|
|
|
class IndentTestCases(BaseTestCase):
|
2002-08-22 15:11:10 -03:00
|
|
|
|
|
|
|
# called before each test method
|
|
|
|
def setUp(self):
|
2002-08-22 16:06:45 -03:00
|
|
|
self.text = '''\
|
2002-08-22 15:11:10 -03:00
|
|
|
This paragraph will be filled, first without any indentation,
|
|
|
|
and then with some (including a hanging indent).'''
|
|
|
|
|
|
|
|
|
2002-08-22 15:35:49 -03:00
|
|
|
def test_fill(self):
|
2002-08-22 17:13:47 -03:00
|
|
|
# Test the fill() method
|
2002-08-22 15:11:10 -03:00
|
|
|
|
|
|
|
expect = '''\
|
|
|
|
This paragraph will be filled, first
|
|
|
|
without any indentation, and then with
|
|
|
|
some (including a hanging indent).'''
|
|
|
|
|
2002-08-22 16:06:45 -03:00
|
|
|
result = fill(self.text, 40)
|
2002-08-22 15:11:10 -03:00
|
|
|
self.check(result, expect)
|
|
|
|
|
|
|
|
|
2002-08-22 15:35:49 -03:00
|
|
|
def test_initial_indent(self):
|
2002-08-22 17:13:47 -03:00
|
|
|
# Test initial_indent parameter
|
2002-08-22 15:11:10 -03:00
|
|
|
|
2002-08-22 16:06:45 -03:00
|
|
|
expect = [" This paragraph will be filled,",
|
|
|
|
"first without any indentation, and then",
|
|
|
|
"with some (including a hanging indent)."]
|
|
|
|
result = wrap(self.text, 40, initial_indent=" ")
|
2002-08-22 15:11:10 -03:00
|
|
|
self.check(result, expect)
|
|
|
|
|
2002-08-22 16:06:45 -03:00
|
|
|
expect = "\n".join(expect)
|
|
|
|
result = fill(self.text, 40, initial_indent=" ")
|
2002-08-22 15:11:10 -03:00
|
|
|
self.check(result, expect)
|
|
|
|
|
|
|
|
|
2002-08-22 15:35:49 -03:00
|
|
|
def test_subsequent_indent(self):
|
2002-08-22 17:13:47 -03:00
|
|
|
# Test subsequent_indent parameter
|
2002-08-22 15:11:10 -03:00
|
|
|
|
|
|
|
expect = '''\
|
|
|
|
* This paragraph will be filled, first
|
|
|
|
without any indentation, and then
|
|
|
|
with some (including a hanging
|
|
|
|
indent).'''
|
|
|
|
|
2002-08-22 16:06:45 -03:00
|
|
|
result = fill(self.text, 40,
|
|
|
|
initial_indent=" * ", subsequent_indent=" ")
|
2002-08-22 15:11:10 -03:00
|
|
|
self.check(result, expect)
|
|
|
|
|
|
|
|
|
2003-05-07 22:58:26 -03:00
|
|
|
# Despite the similar names, DedentTestCase is *not* the inverse
|
|
|
|
# of IndentTestCase!
|
|
|
|
class DedentTestCase(unittest.TestCase):
|
|
|
|
|
2006-06-10 21:40:49 -03:00
|
|
|
def assertUnchanged(self, text):
|
|
|
|
"""assert that dedent() has no effect on 'text'"""
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(text, dedent(text))
|
2006-06-10 21:40:49 -03:00
|
|
|
|
2003-05-07 22:58:26 -03:00
|
|
|
def test_dedent_nomargin(self):
|
|
|
|
# No lines indented.
|
|
|
|
text = "Hello there.\nHow are you?\nOh good, I'm glad."
|
2006-06-10 21:40:49 -03:00
|
|
|
self.assertUnchanged(text)
|
2003-05-07 22:58:26 -03:00
|
|
|
|
|
|
|
# Similar, with a blank line.
|
|
|
|
text = "Hello there.\n\nBoo!"
|
2006-06-10 21:40:49 -03:00
|
|
|
self.assertUnchanged(text)
|
2003-05-07 22:58:26 -03:00
|
|
|
|
|
|
|
# Some lines indented, but overall margin is still zero.
|
|
|
|
text = "Hello there.\n This is indented."
|
2006-06-10 21:40:49 -03:00
|
|
|
self.assertUnchanged(text)
|
2003-05-07 22:58:26 -03:00
|
|
|
|
|
|
|
# Again, add a blank line.
|
|
|
|
text = "Hello there.\n\n Boo!\n"
|
2006-06-10 21:40:49 -03:00
|
|
|
self.assertUnchanged(text)
|
2003-05-07 22:58:26 -03:00
|
|
|
|
|
|
|
def test_dedent_even(self):
|
|
|
|
# All lines indented by two spaces.
|
|
|
|
text = " Hello there.\n How are ya?\n Oh good."
|
|
|
|
expect = "Hello there.\nHow are ya?\nOh good."
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(expect, dedent(text))
|
2003-05-07 22:58:26 -03:00
|
|
|
|
|
|
|
# Same, with blank lines.
|
|
|
|
text = " Hello there.\n\n How are ya?\n Oh good.\n"
|
|
|
|
expect = "Hello there.\n\nHow are ya?\nOh good.\n"
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(expect, dedent(text))
|
2003-05-07 22:58:26 -03:00
|
|
|
|
|
|
|
# Now indent one of the blank lines.
|
|
|
|
text = " Hello there.\n \n How are ya?\n Oh good.\n"
|
|
|
|
expect = "Hello there.\n\nHow are ya?\nOh good.\n"
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(expect, dedent(text))
|
2003-05-07 22:58:26 -03:00
|
|
|
|
|
|
|
def test_dedent_uneven(self):
|
|
|
|
# Lines indented unevenly.
|
|
|
|
text = '''\
|
|
|
|
def foo():
|
|
|
|
while 1:
|
|
|
|
return foo
|
|
|
|
'''
|
|
|
|
expect = '''\
|
|
|
|
def foo():
|
|
|
|
while 1:
|
|
|
|
return foo
|
|
|
|
'''
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(expect, dedent(text))
|
2003-05-07 22:58:26 -03:00
|
|
|
|
|
|
|
# Uneven indentation with a blank line.
|
|
|
|
text = " Foo\n Bar\n\n Baz\n"
|
|
|
|
expect = "Foo\n Bar\n\n Baz\n"
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(expect, dedent(text))
|
2003-05-07 22:58:26 -03:00
|
|
|
|
|
|
|
# Uneven indentation with a whitespace-only line.
|
|
|
|
text = " Foo\n Bar\n \n Baz\n"
|
|
|
|
expect = "Foo\n Bar\n\n Baz\n"
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(expect, dedent(text))
|
2006-06-10 21:40:49 -03:00
|
|
|
|
|
|
|
# dedent() should not mangle internal tabs
|
|
|
|
def test_dedent_preserve_internal_tabs(self):
|
|
|
|
text = " hello\tthere\n how are\tyou?"
|
|
|
|
expect = "hello\tthere\nhow are\tyou?"
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(expect, dedent(text))
|
2006-06-10 21:40:49 -03:00
|
|
|
|
|
|
|
# make sure that it preserves tabs when it's not making any
|
|
|
|
# changes at all
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(expect, dedent(expect))
|
2006-06-10 21:40:49 -03:00
|
|
|
|
|
|
|
# dedent() should not mangle tabs in the margin (i.e.
|
|
|
|
# tabs and spaces both count as margin, but are *not*
|
|
|
|
# considered equivalent)
|
|
|
|
def test_dedent_preserve_margin_tabs(self):
|
|
|
|
text = " hello there\n\thow are you?"
|
|
|
|
self.assertUnchanged(text)
|
|
|
|
|
|
|
|
# same effect even if we have 8 spaces
|
|
|
|
text = " hello there\n\thow are you?"
|
|
|
|
self.assertUnchanged(text)
|
|
|
|
|
|
|
|
# dedent() only removes whitespace that can be uniformly removed!
|
|
|
|
text = "\thello there\n\thow are you?"
|
|
|
|
expect = "hello there\nhow are you?"
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(expect, dedent(text))
|
2006-06-10 21:40:49 -03:00
|
|
|
|
|
|
|
text = " \thello there\n \thow are you?"
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(expect, dedent(text))
|
2006-06-10 21:40:49 -03:00
|
|
|
|
|
|
|
text = " \t hello there\n \t how are you?"
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(expect, dedent(text))
|
2006-06-11 16:42:51 -03:00
|
|
|
|
2006-06-10 21:40:49 -03:00
|
|
|
text = " \thello there\n \t how are you?"
|
|
|
|
expect = "hello there\n how are you?"
|
2010-11-21 09:34:58 -04:00
|
|
|
self.assertEqual(expect, dedent(text))
|
2003-05-07 22:58:26 -03:00
|
|
|
|
|
|
|
|
2002-08-22 15:35:49 -03:00
|
|
|
def test_main():
|
2003-05-07 22:58:26 -03:00
|
|
|
test_support.run_unittest(WrapTestCase,
|
|
|
|
LongWordTestCase,
|
|
|
|
IndentTestCases,
|
|
|
|
DedentTestCase)
|
2002-08-22 15:35:49 -03:00
|
|
|
|
2002-08-22 15:11:10 -03:00
|
|
|
if __name__ == '__main__':
|
2002-08-22 15:35:49 -03:00
|
|
|
test_main()
|