l.s.helpers : module documentation

Return a new string with text replaced according to the dict provided.

The keys of the dict are substrings to find, the values are what to replace found substrings with.

>>> text_replaced('', {'a':'b'})
''
>>> text_replaced('a', {'a':'c'})
'c'
>>> text_replaced('faa bar baz', {'a': 'A', 'aa': 'X'})
'fX bAr bAz'
>>> text_replaced('1 2 3 4', {'1': '2', '2': '1'})
'2 1 3 4'

Unicode strings work too.

>>> text_replaced(u'1 2 3 4', {u'1': u'2', u'2': u'1'})
u'2 1 3 4'

The argument _cache is used as a cache of replacements that were requested before, so we only compute regular expressions once.

Parameters	text	An unicode or str to do the replacement.
	replacements	A dictionary with the replacements that should be done

Return a copy of the string, with non-ASCII characters rendered as xNN or uNNNN. Used to test data containing typographical quotes etc.

Convert a binary string containing a tar file into a tar file obj.

Run a command, give it input on its standard input, and capture its standard output.

Returns the data from standard output.

This function is needed to avoid certain deadlock situations. For example, if you popen2() a command, write its standard input, then read its standard output, this can deadlock due to the parent process blocking on writing to the child, while the child process is simultaneously blocking on writing to its parent. This function avoids that problem by using subprocess.Popen.communicate().

Return a listified version of sequence.

If <sequence> has more than <longest_expected> items, a warning is issued.

>>> shortlist([1, 2])
[1, 2]

>>> shortlist([1, 2, 3], 2) #doctest: +NORMALIZE_WHITESPACE
Traceback (most recent call last):
...
UserWarning: shortlist() should not be used here. It's meant to listify
sequences with no more than 2 items.  There were 3 items.

>>> shortlist([1, 2, 3, 4], hardlimit=2)
Traceback (most recent call last):
...
ShortListTooBigError: Hard limit of 2 exceeded.

>>> shortlist(
...     [1, 2, 3, 4], 2, hardlimit=4) #doctest: +NORMALIZE_WHITESPACE
Traceback (most recent call last):
...
UserWarning: shortlist() should not be used here. It's meant to listify
sequences with no more than 2 items.  There were 4 items.

It works on iterable also which don't support the extended slice protocol.

>>> iter(range(5))[:1] #doctest: +ELLIPSIS
Traceback (most recent call last):
...
TypeError: ...

>>> shortlist(iter(range(10)), 5, hardlimit=8) #doctest: +ELLIPSIS
Traceback (most recent call last):
...
ShortListTooBigError: ...

Check whether a filename looks like a filename that belongs to a tar file, possibly one compressed somehow.

Generate a string indicating the difference between expected and actual values in a test.

Return the a ContentType-like entry for arbitrary filenames

deb files

>>> filenameToContentType('test.deb')
'application/x-debian-package'

text files

>>> filenameToContentType('test.txt')
'text/plain'

Not recognized format

>>> filenameToContentType('test.tgz')
'application/octet-stream'

Build logs >>> filenameToContentType('buildlog.txt.gz') 'text/plain'

Various compressed files

>>> filenameToContentType('Packages.gz')
'application/x-gzip'
>>> filenameToContentType('Packages.bz2')
'application/x-bzip2'
>>> filenameToContentType('Packages.xz')
'application/x-xz'

Return int(value) or 0 if the conversion fails.

>>> intOrZero('1.23')
0
>>> intOrZero('1.ab')
0
>>> intOrZero('2')
2
>>> intOrZero(None)
0
>>> intOrZero(1)
1
>>> intOrZero(-9)
-9

Return a version of string no longer than max_length characters.

Tries not to cut off the text mid-word.

Return all the items concatenated into a English-style string.

Follows the advice given in The Elements of Style, chapter I, section 2:

"In a series of three or more terms with a single conjunction, use

a comma after each term except the last."

Beware that this is US English and is wrong for non-US.

Return input as unicode. None is passed through unharmed.

Do not use this method. This method exists only to help migration of legacy code where str objects were being passed into contexts where unicode objects are required. All invokations of ensure_unicode() should eventually be removed.

This differs from the builtin unicode() function, as a TypeError exception will be raised if the parameter is not a basestring or if a raw string is not ASCII.

>>> ensure_unicode(u'hello')
u'hello'

>>> ensure_unicode('hello')
u'hello'

>>> ensure_unicode(u'A'.encode('utf-16')) # Not ASCII
Traceback (most recent call last):
...
TypeError: '\xff\xfeA\x00' is not US-ASCII

>>> ensure_unicode(42)
Traceback (most recent call last):
...
TypeError: 42 is not a basestring (<type 'int'>)

>>> ensure_unicode(None) is None
True