diff options
author | Pavel Aharoni <pa0916@att.com> | 2017-03-29 13:35:45 +0300 |
---|---|---|
committer | Pavel Aharoni <pa0916@att.com> | 2017-03-29 13:35:45 +0300 |
commit | e2cc2530fc6d54ebc975c01a4ff887ce12f0a736 (patch) | |
tree | 38385867295c8a09fb0d7f8eaf5fa78179e5b13a /jython-tosca-parser/src/main/resources/Lib/site-packages/pip/_vendor/html5lib/filters/whitespace.py | |
parent | bccebaa9888906f8ff78172f62ec592956066d82 (diff) |
[SDC-6] sdc-distribution-client 1707 rebasing
Change-Id: I322a05fd79beb6ba4fee4d32afffecf531b86e98
Signed-off-by: Pavel Aharoni <pa0916@att.com>
Diffstat (limited to 'jython-tosca-parser/src/main/resources/Lib/site-packages/pip/_vendor/html5lib/filters/whitespace.py')
-rw-r--r-- | jython-tosca-parser/src/main/resources/Lib/site-packages/pip/_vendor/html5lib/filters/whitespace.py | 38 |
1 files changed, 38 insertions, 0 deletions
diff --git a/jython-tosca-parser/src/main/resources/Lib/site-packages/pip/_vendor/html5lib/filters/whitespace.py b/jython-tosca-parser/src/main/resources/Lib/site-packages/pip/_vendor/html5lib/filters/whitespace.py new file mode 100644 index 0000000..dfc60ee --- /dev/null +++ b/jython-tosca-parser/src/main/resources/Lib/site-packages/pip/_vendor/html5lib/filters/whitespace.py @@ -0,0 +1,38 @@ +from __future__ import absolute_import, division, unicode_literals + +import re + +from . import _base +from ..constants import rcdataElements, spaceCharacters +spaceCharacters = "".join(spaceCharacters) + +SPACES_REGEX = re.compile("[%s]+" % spaceCharacters) + + +class Filter(_base.Filter): + + spacePreserveElements = frozenset(["pre", "textarea"] + list(rcdataElements)) + + def __iter__(self): + preserve = 0 + for token in _base.Filter.__iter__(self): + type = token["type"] + if type == "StartTag" \ + and (preserve or token["name"] in self.spacePreserveElements): + preserve += 1 + + elif type == "EndTag" and preserve: + preserve -= 1 + + elif not preserve and type == "SpaceCharacters" and token["data"]: + # Test on token["data"] above to not introduce spaces where there were not + token["data"] = " " + + elif not preserve and type == "Characters": + token["data"] = collapse_spaces(token["data"]) + + yield token + + +def collapse_spaces(text): + return SPACES_REGEX.sub(' ', text) |