summaryrefslogtreecommitdiff
path: root/src/lxml/apihelpers.pxi
Commit message (Collapse)AuthorAgeFilesLines
* Implement indent() function for in-place pretty-printing of XML trees.Stefan Behnel2019-08-131-0/+13
|
* LP#1838252: Keep the order provided by an OrderedDict that gets passed as ↵Stefan Behnel2019-07-291-10/+5
| | | | attrib mapping during element creation. This was broken in 4.4.0.
* Fix some misbehaviour in slice selection and deletion: Large step sizes ↵Stefan Behnel2019-04-211-0/+2
| | | | could lead to long running stupid loops.
* Fix some misbehaviour in slice assignments:Stefan Behnel2019-04-211-1/+9
| | | | | | - Large step sizes could lead to long running stupid loops. - ValueError was not raised when assigning extended slices of the wrong size. - Slices with negative step size could be inserted in the wrong place, too far on the left.
* Refactor duplicate code.Stefan Behnel2019-04-051-0/+18
|
* Exclude absolute Windows (C:\...) file paths from URL escaping since libxml2 ↵Stefan Behnel2019-03-271-4/+11
| | | | does not recognise them as file paths and thus does not unescape them.
* Fix C compiler warning about comparing signed to unsigned integers.Stefan Behnel2019-03-271-1/+1
|
* Keep the original dict insertion order in Py3.6+ when setting attributes or ↵Stefan Behnel2019-03-011-3/+14
| | | | | | namespaces from a user provided dict. This follows the ElementTree change in Py3.8, see https://bugs.python.org/issue34160.
* Fix crash due to incorrect dict handling for text nodes.Stefan Behnel2019-02-281-1/+3
| | | | The C doc link needs to be set after removing text from the dict and before putting it there. Thus, it is best to separate the adaptations into two traversals again.
* LP#1814522: Fix a crash when appending a child subtree that contains ↵Stefan Behnel2019-02-051-2/+19
| | | | | | | unsubstituted entity references. This is a work-around for a (supposed) bug in libxml2 (https://gitlab.gnome.org/GNOME/libxml2/issues/42), which crashes by running into an infinite recursive loop while traversing the child nodes of the entity reference. A lucky side effect is that the previously duplicated cleanup traversal to a) update the .doc pointers in libxml2 and b) update the dict names in lxml is now replaced by a single traversal, which should speed things up for large subtrees.
* Speed up ascii/non-ascii string detection in isutf8() and funicode() helper ↵Stefan Behnel2018-10-201-6/+42
| | | | functions.
* Simplify isinstanceHugo2018-08-261-1/+1
|
* Remove unnecessary backslashHugo2018-08-251-2/+2
|
* Remove redundant code for Python <= 2.6Hugo2018-08-251-3/+2
|
* Use f-strings for all string formatting for which it makes sense (i.e. does ↵Stefan Behnel2018-01-251-22/+12
| | | | not look unreadable).
* fix C compiler warning about "const" type mismatchStefan Behnel2017-08-111-0/+1
|
* replace IS_PYTHON3 C macro with IS_PYTHON2 before it starts looking funny in ↵Stefan Behnel2016-09-031-2/+2
| | | | Python 4.x :)
* use PyUnicode_FromFormat() also in Py2.x (but not in PyPy which doesn't have it)Stefan Behnel2016-09-031-2/+2
|
* allow None as attribute value only for HTML treesStefan Behnel2016-07-291-2/+3
|
* allow None as an attribute value, for HTML boolean attributesDaniel Holth2016-07-281-4/+7
|
* re-implement "keep_ns_prefixes" option in cleanup_namespaces()Stefan Behnel2015-09-041-8/+10
|
* Merge pull request #167 from tomkralidis/cleanup_namespaces-keep_nsmapscoder2015-09-041-3/+8
|\ | | | | add keep_nsmap argument to cleanup_namespaces
| * test on listTom Kralidis2015-06-141-1/+1
| |
| * pass only ns prefixesTom Kralidis2015-06-141-4/+4
| |
| * update implemenation, testsTom Kralidis2015-05-181-1/+2
| |
| * doc changeTom Kralidis2015-05-151-1/+1
| |
| * update everything but the actual implementation change to reflect agreed ↵Tom Kralidis2015-05-151-1/+5
| | | | | | | | upon semantics
* | improve error messageStefan Behnel2015-08-211-1/+1
|/
* use PyUnicode_FromFormat in pypy3Stefan Behnel2015-04-101-1/+1
|
* minor code cleanupStefan Behnel2015-04-041-1/+1
|
* speed up removal of unused namespaces by avoiding to traverse ↵Stefan Behnel2015-03-011-4/+5
| | | | elements/attributes for which there is nothing to remove
* for better symmetry, use lxml_malloc()/lxml_free() instead of ↵Stefan Behnel2015-03-011-4/+4
| | | | lxml_malloc()/PyMem_Free()
* make nsmap evaluation safer by preventing to sort on None and string values ↵Stefan Behnel2015-03-011-36/+46
| | | | (fails in Py3)
* make attribute creation order in xmlfile.element() reproducible (same as for ↵Stefan Behnel2015-03-011-10/+18
| | | | Element creation)
* make sure we always validate 'node_ns_utf' in _setNodeNamespaces() by ↵Stefan Behnel2015-02-281-37/+36
| | | | removing only partially redundant code
* remove useless list size changingStefan Behnel2015-02-271-4/+6
| | | | | --HG-- extra : transplant_source : %95%FEhL%84%84V%8B%F4%AB6%09e%C6eg%BB%7EOZ
* declare types that Cython couldn't inferStefan Behnel2015-02-271-2/+2
|
* refactor duplicate code in _removeUnusedNamespaceDeclarations()Stefan Behnel2015-02-271-45/+34
|
* rename helper function to match more general applicability and fix incorrect ↵Stefan Behnel2015-02-271-7/+7
| | | | docstring
* clean up and refactor last commitsStefan Behnel2015-02-211-20/+23
|
* Update name validation.Olli Pottonen2015-02-211-2/+2
| | | | | Validate names accordingy to XML 1.0 standard fifth edition, instead of the outdated fourth edition. (See also Namespaces in XML 1.0 third edition.)
* On python 2, __repr__ shall return str, not unicode.Olli Pottonen2015-02-211-0/+9
|
* Check all input strings are in XML character range.Olli Pottonen2015-02-211-19/+38
|
* Make _ElementTree.deepcopy() properly handle comments/PI's before doctype.Olli Pottonen2015-02-181-5/+13
|
* accept CDATA wrapped content for tail textStefan Behnel2014-11-231-16/+16
|
* merge lxml-3.3 branch into masterStefan Behnel2014-08-281-26/+24
|\
| * refactor some duplicated codeStefan Behnel2014-08-281-18/+16
| |
| * prevent tree cycle creation when adding siblingsStefan Behnel2014-08-281-14/+14
| |
| * PyPy fixStefan Behnel2014-03-101-1/+2
| |
* | simplify some codeStefan Behnel2014-04-041-4/+2
| |
View Lorry Controller Status