summaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
...
* | Change version in master branch to 4.7.0a0.lxml-4.7.0-preStefan Behnel2021-11-051-1/+1
| |
* | Prepare release of 4.6.4.Stefan Behnel2021-10-151-1/+1
| |
* | Implement a dedicated int/float parser for XML (schema) values in ↵Stefan Behnel2021-08-142-9/+179
| | | | | | | | | | | | | | | | | | lxml.objectify. This disables support for "_" in numbers, which are allowed by Python but not by XMLSchema. We keep a few additional literals, such as "+NaN", simply because they shouldn't hurt. See https://mail.python.org/archives/list/lxml@python.org/thread/6F7VIDKWZTJ6LB6VOX6IJNNWICYHFPNR/
* | Use Cython's autowrapping feature for cdef functions to keep internal ↵Stefan Behnel2021-08-121-12/+12
| | | | | | | | utility functions out of the objectify module dict.
* | _tofilelikeC14N: Always close output buffer (GH-322)Petr Viktorin2021-07-291-8/+10
| | | | | | | | | | | | | | | | | | If `with writer.error_log` raises an exception, `c_buffer` would leak. It seems that currently, it can't actually raise (it's uses small and tight `cdef` functions), but there's no guarantee they'll remain exception-free in the future. But there's one more thing that potentially could leak (at least Cython generates an `unlikely` `goto` block for it): the lookup of `__exit__` that happens at the start of the `with` block. Put the `xmlOutputBufferClose` call into a `finally` block to make this safer.
* | Rewrite Unicode chunk parsing by directly encoding to UTF-8.Stefan Behnel2021-07-182-49/+114
|/ | | | Previously, we required Py_UNICODE strings, which is inefficient since most strings in Py3 use the PEP-393 memory layout.
* Implement "__rXXX__" special methods in objectify elements to support proper ↵Stefan Behnel2021-07-161-12/+87
| | | | Python semantics in Cython 3.
* Avoid direct C-API call.Stefan Behnel2021-05-191-2/+1
|
* Adapt a test to a behavioural change in libxml2 2.9.11+.Stefan Behnel2021-05-191-1/+4
|
* Clean up fuzzer test.Stefan Behnel2021-05-081-1/+3
|
* Add initial Atheris fuzzer. (GH-313)DavidKorczynski2021-05-081-0/+23
|
* Enable access to the system_url of DTD entity declarations (GH-317)Joel2021-05-082-0/+13
|
* Prepare release of lxml 4.6.3.lxml-4.6.3Stefan Behnel2021-03-211-1/+1
|
* Add HTML-5 "formaction" attribute to "defs.link_attrs" (GH-316)Kevin Chung2021-03-212-0/+17
| | | | Resolves https://bugs.launchpad.net/lxml/+bug/1888153 See https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-28957
* Work around Py2's lack of "re.ASCII".lxml-4.6.2Stefan Behnel2020-11-261-2/+4
|
* Prepare release of 4.6.2.Stefan Behnel2020-11-261-1/+1
|
* Prevent combinations of <math/svg> and <style> to sneak JavaScript through ↵Stefan Behnel2020-11-263-11/+39
| | | | the HTML cleaner.
* Prepare release of lxml 4.6.1.lxml-4.6.1Stefan Behnel2020-10-181-1/+1
|
* Prevent combinations of <noscript> and <style> to sneak JavaScript through ↵Stefan Behnel2020-10-182-0/+13
| | | | the HTML cleaner.
* Prepare release of lxml 4.6.0.Stefan Behnel2020-10-171-1/+1
|
* Exclude a test in Py 3.9.0 due to ET bug https://bugs.python.org/issue41900Stefan Behnel2020-10-151-0/+12
|
* Disable test in unfixed ET versions <= 3.8.6.Stefan Behnel2020-10-011-0/+4
|
* LP#1869455: C14N 2.0 serialisation failed for unprefixed attributes when a ↵Stefan Behnel2020-10-012-1/+14
| | | | default namespace was defined.
* Remove dead imports.Stefan Behnel2020-08-132-2/+2
|
* Remove dead code.Stefan Behnel2020-08-131-9/+0
|
* Disable gc.collect() calls after each test run since there haven't been ↵Stefan Behnel2020-08-131-1/+10
| | | | proxy crashes for a very long time now and it considerably slows down the test runs (~factor 6).
* Remove dead code.Stefan Behnel2020-08-131-1/+1
|
* html: Add InputGetter.items() method and make .keys() return the field names ↵Stefan Behnel2020-08-122-8/+47
| | | | in document order.
* html: Avoid XPath in InputGetter where fast and simple iteration is enough.Stefan Behnel2020-08-121-20/+19
|
* html: Simplify and speed up InputGetter.__iter__() and __len__().Stefan Behnel2020-08-121-6/+3
|
* Implement __len__() on InputGetter which is expected by ↵AidanWoolley2020-08-121-0/+3
| | | | FormElement/FieldsDict (GH-310)
* Use sphinx-apidoc to create API reference (GH-309)Chris Mayo2020-08-046-98/+102
| | | | | | | | | | | | | * Add some missing files to .gitignore * Remove duplicate open_in_browser from lxml.html.__all__ * Make ETreeXMLSchemaTestCase docstring Sphinx autodoc friendly * Fix outdated codespeak.net links in docstrings * Convert html/defs.py comment to be the module docstring * Use sphinx-apidoc to create the API reference instead of epydoc Epydoc is Python 2 only and unmaintained. sphinx-apidoc is run before the build step, to avoid duplicate entries being created. * Include the elements from html.builder in the API reference * Use Python 3.8 for coverage Travis job * Build html documentation in Travis
* Fix an import in Py3.Stefan Behnel2020-08-041-1/+1
|
* Remove dead code.Stefan Behnel2020-08-031-1/+0
|
* Raise XMLSyntaxError instead of plain AssertionError when calling ↵Stefan Behnel2020-07-281-3/+15
| | | | | | TreeBuilder.close() in an inconsistent state. Uses a subclass XMLSyntaxAssertionError that also inherits from AssertionError to keep up backwards compatibility.
* Prepare release of 4.5.2.lxml-4.5.2Stefan Behnel2020-07-091-1/+1
|
* Cleaner: Catch bad arg combo in constructor (GH-301)Mike Lissner2020-06-202-0/+21
| | | Fixes https://bugs.launchpad.net/lxml/+bug/1882606
* Improve compilation of clean.py (e.g. dict iteration) by switching to ↵Stefan Behnel2020-06-191-1/+1
| | | | language_level=3str.
* Avoid calling hasattr when we need the attribute anyway, and validate the ↵Stefan Behnel2020-06-191-4/+10
| | | | argument names passed into Cleaner() along the way.
* Extend C14N2 tests to cover comment handling and "strip_text" together.Stefan Behnel2020-06-141-16/+21
|
* LP#1882606: ``Cleaner.clean_html()`` discarded comments and PIs regardless ↵Stefan Behnel2020-06-133-4/+49
| | | | of the corresponding configuration option, if "remove_unknown_tags=True" was set.
* Use a bound method instead of looking it up on each element.Stefan Behnel2020-06-131-2/+2
|
* Fix a test after moving it to a different test module.Stefan Behnel2020-05-261-2/+2
|
* Move some ElementTree compatibility tests over to the etree-only tests since ↵Stefan Behnel2020-05-262-248/+252
| | | | the features were removed in Py3.9.
* Avoid globally overriding the libxml2 external entity resolver and instead ↵Stefan Behnel2020-05-237-11/+51
| | | | | | set it for each parser run. This improves the interoperability with other users of libxml2 in the system, such as libxmlsec.
* Prepare release of 4.5.1.lxml-4.5.1Stefan Behnel2020-05-191-1/+1
|
* Make it less likely that the serialisation of large documents (> MAX_INT) is ↵Stefan Behnel2020-05-112-4/+6
| | | | considered a failure due to C integer wrap-around.
* Make iter() work with qnames (GH-298)xmo-odoo2020-03-032-0/+26
| | | | | "QName" is supposed to be usable anywhere a tag name is expected and iter() should take any number of tag names for filtering, but before this change passing a QName to iter() results in an exception.
* Prepare release of lxml 4.5.0.lxml-4.5.0Stefan Behnel2020-01-291-1/+1
|
* LP#1857794: Tail text of nodes that get removed from a document using item ↵Stefan Behnel2020-01-022-1/+28
| | | | deletion disappeared silently instead of sticking with the node that was removed.
View Lorry Controller Status