rpms
/
python-lxml
2
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Update to 4.5.1 

Signed-off-by: Igor Raits <ignatenkobrain@fedoraproject.org>
This commit is contained in:
Igor Raits 2020-06-01 20:13:00 +02:00
parent 7af357f1ba
commit e646c1352a
4 changed files with 286 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
.gitignore vendored
View File

@ -53,3 +53,4 @@ lxml-2.2.7.tar.gz.asc
/lxml-4.2.5.tgz
/lxml-4.4.0.tgz
/lxml-4.4.1.tgz
/lxml-4.5.1.tgz

277
0001-Avoid-globally-overriding-the-libxml2-external-entit.patch Normal file
View File

@ -0,0 +1,277 @@
From fa1d856cad369d0ac64323ddec14b02281491706 Mon Sep 17 00:00:00 2001
From: Stefan Behnel <stefan_ml@behnel.de>
Date: Sat, 23 May 2020 09:34:22 +0200
Subject: [PATCH] Avoid globally overriding the libxml2 external entity
resolver and instead set it for each parser run. This improves the
interoperability with other users of libxml2 in the system, such as
libxmlsec.
---
CHANGES.txt | 11 +++++++++++
src/lxml/dtd.pxi | 6 ++++++
src/lxml/parser.pxi | 42 ++++++++++++++++++++++++++++++-----------
src/lxml/relaxng.pxi | 2 ++
src/lxml/schematron.pxi | 4 ++++
src/lxml/xinclude.pxi | 2 ++
src/lxml/xmlschema.pxi | 2 ++
src/lxml/xslt.pxi | 4 ++++
8 files changed, 62 insertions(+), 11 deletions(-)
diff --git a/CHANGES.txt b/CHANGES.txt
index 30e805997..07afb641b 100644
--- a/CHANGES.txt
+++ b/CHANGES.txt
@@ -2,6 +2,17 @@
lxml changelog
==============
+4.5.2 (2020-0?-??)
+==================
+
+Bugs fixed
+----------
+
+* LP#1880251: Instead of globally overwriting the document loader in libxml2, lxml now
+ sets it per parser run, which improves the interoperability with other users of libxml2
+ such as libxmlsec.
+
+
4.5.1 (2020-05-19)
==================
diff --git a/src/lxml/dtd.pxi b/src/lxml/dtd.pxi
index 595296546..5dcb80c46 100644
--- a/src/lxml/dtd.pxi
+++ b/src/lxml/dtd.pxi
@@ -277,14 +277,20 @@ cdef class DTD(_Validator):
if _isString(file):
file = _encodeFilename(file)
with self._error_log:
+ orig_loader = _register_document_loader()
self._c_dtd = xmlparser.xmlParseDTD(NULL, _xcstr(file))
+ _reset_document_loader(orig_loader)
elif hasattr(file, 'read'):
+ orig_loader = _register_document_loader()
self._c_dtd = _parseDtdFromFilelike(file)
+ _reset_document_loader(orig_loader)
else:
raise DTDParseError, u"file must be a filename or file-like object"
elif external_id is not None:
with self._error_log:
+ orig_loader = _register_document_loader()
self._c_dtd = xmlparser.xmlParseDTD(<const_xmlChar*>external_id, NULL)
+ _reset_document_loader(orig_loader)
else:
raise DTDParseError, u"either filename or external ID required"
diff --git a/src/lxml/parser.pxi b/src/lxml/parser.pxi
index 22620373c..3ed223bd5 100644
--- a/src/lxml/parser.pxi
+++ b/src/lxml/parser.pxi
@@ -502,7 +502,15 @@ cdef xmlparser.xmlParserInput* _local_resolver(const_char* c_url, const_char* c_
cdef xmlparser.xmlExternalEntityLoader __DEFAULT_ENTITY_LOADER
__DEFAULT_ENTITY_LOADER = xmlparser.xmlGetExternalEntityLoader()
-xmlparser.xmlSetExternalEntityLoader(<xmlparser.xmlExternalEntityLoader>_local_resolver)
+
+cdef xmlparser.xmlExternalEntityLoader _register_document_loader() nogil:
+ cdef xmlparser.xmlExternalEntityLoader old = xmlparser.xmlGetExternalEntityLoader()
+ xmlparser.xmlSetExternalEntityLoader(<xmlparser.xmlExternalEntityLoader>_local_resolver)
+ return old
+
+cdef void _reset_document_loader(xmlparser.xmlExternalEntityLoader old) nogil:
+ xmlparser.xmlSetExternalEntityLoader(old)
+
############################################################
## Parsers
@@ -514,6 +522,7 @@ cdef class _ParserContext(_ResolverContext):
cdef _ErrorLog _error_log
cdef _ParserSchemaValidationContext _validator
cdef xmlparser.xmlParserCtxt* _c_ctxt
+ cdef xmlparser.xmlExternalEntityLoader _orig_loader
cdef python.PyThread_type_lock _lock
cdef _Document _doc
cdef bint _collect_ids
@@ -561,7 +570,7 @@ cdef class _ParserContext(_ResolverContext):
else:
xmlparser.xmlClearParserCtxt(self._c_ctxt)
- cdef int prepare(self) except -1:
+ cdef int prepare(self, bint set_document_loader=True) except -1:
cdef int result
if config.ENABLE_THREADING and self._lock is not NULL:
with nogil:
@@ -572,19 +581,24 @@ cdef class _ParserContext(_ResolverContext):
self._error_log.clear()
self._doc = None
self._c_ctxt.sax.serror = _receiveParserError
+ self._orig_loader = _register_document_loader() if set_document_loader else NULL
if self._validator is not None:
self._validator.connect(self._c_ctxt, self._error_log)
return 0
cdef int cleanup(self) except -1:
- if self._validator is not None:
- self._validator.disconnect()
- self._resetParserContext()
- self.clear()
- self._doc = None
- self._c_ctxt.sax.serror = NULL
- if config.ENABLE_THREADING and self._lock is not NULL:
- python.PyThread_release_lock(self._lock)
+ if self._orig_loader is not NULL:
+ _reset_document_loader(self._orig_loader)
+ try:
+ if self._validator is not None:
+ self._validator.disconnect()
+ self._resetParserContext()
+ self.clear()
+ self._doc = None
+ self._c_ctxt.sax.serror = NULL
+ finally:
+ if config.ENABLE_THREADING and self._lock is not NULL:
+ python.PyThread_release_lock(self._lock)
return 0
cdef object _handleParseResult(self, _BaseParser parser,
@@ -1286,7 +1300,7 @@ cdef class _FeedParser(_BaseParser):
pctxt = context._c_ctxt
error = 0
if not self._feed_parser_running:
- context.prepare()
+ context.prepare(set_document_loader=False)
self._feed_parser_running = 1
c_filename = (_cstr(self._filename)
if self._filename is not None else NULL)
@@ -1296,6 +1310,7 @@ cdef class _FeedParser(_BaseParser):
# however if we give it all we got, we'll have nothing for
# *mlParseChunk() and things go wrong.
buffer_len = 4 if py_buffer_len > 4 else <int>py_buffer_len
+ orig_loader = _register_document_loader()
if self._for_html:
error = _htmlCtxtResetPush(
pctxt, c_data, buffer_len, c_filename, c_encoding,
@@ -1304,6 +1319,7 @@ cdef class _FeedParser(_BaseParser):
xmlparser.xmlCtxtUseOptions(pctxt, self._parse_options)
error = xmlparser.xmlCtxtResetPush(
pctxt, c_data, buffer_len, c_filename, c_encoding)
+ _reset_document_loader(orig_loader)
py_buffer_len -= buffer_len
c_data += buffer_len
if error:
@@ -1321,7 +1337,9 @@ cdef class _FeedParser(_BaseParser):
buffer_len = <int>py_buffer_len
if self._for_html:
c_node = pctxt.node # last node where the parser stopped
+ orig_loader = _register_document_loader()
error = htmlparser.htmlParseChunk(pctxt, c_data, buffer_len, 0)
+ _reset_document_loader(orig_loader)
# and now for the fun part: move node names to the dict
if pctxt.myDoc:
fixup_error = _fixHtmlDictSubtreeNames(
@@ -1331,7 +1349,9 @@ cdef class _FeedParser(_BaseParser):
pctxt.myDoc.dict = pctxt.dict
xmlparser.xmlDictReference(pctxt.dict)
else:
+ orig_loader = _register_document_loader()
error = xmlparser.xmlParseChunk(pctxt, c_data, buffer_len, 0)
+ _reset_document_loader(orig_loader)
py_buffer_len -= buffer_len
c_data += buffer_len
diff --git a/src/lxml/relaxng.pxi b/src/lxml/relaxng.pxi
index d161ce46e..6a82a295f 100644
--- a/src/lxml/relaxng.pxi
+++ b/src/lxml/relaxng.pxi
@@ -64,7 +64,9 @@ cdef class RelaxNG(_Validator):
doc = None
filename = _encodeFilename(file)
with self._error_log:
+ orig_loader = _register_document_loader()
parser_ctxt = relaxng.xmlRelaxNGNewParserCtxt(_cstr(filename))
+ _reset_document_loader(orig_loader)
elif (_getFilenameForFile(file) or '')[-4:].lower() == '.rnc':
_require_rnc2rng()
rng_data_utf8 = _utf8(_rnc2rng.dumps(_rnc2rng.load(file)))
diff --git a/src/lxml/schematron.pxi b/src/lxml/schematron.pxi
index af4ba7f01..dfd2cc05f 100644
--- a/src/lxml/schematron.pxi
+++ b/src/lxml/schematron.pxi
@@ -95,7 +95,9 @@ cdef class Schematron(_Validator):
filename = file
filename = _encodeFilename(filename)
with self._error_log:
+ orig_loader = _register_document_loader()
parser_ctxt = schematron.xmlSchematronNewParserCtxt(_cstr(filename))
+ _reset_document_loader(orig_loader)
else:
raise SchematronParseError, u"No tree or file given"
@@ -107,7 +109,9 @@ cdef class Schematron(_Validator):
try:
with self._error_log:
+ orig_loader = _register_document_loader()
self._c_schema = schematron.xmlSchematronParse(parser_ctxt)
+ _reset_document_loader(orig_loader)
finally:
schematron.xmlSchematronFreeParserCtxt(parser_ctxt)
diff --git a/src/lxml/xinclude.pxi b/src/lxml/xinclude.pxi
index f73afee61..6bac82923 100644
--- a/src/lxml/xinclude.pxi
+++ b/src/lxml/xinclude.pxi
@@ -49,11 +49,13 @@ cdef class XInclude:
if tree.LIBXML_VERSION < 20704 or not c_context:
__GLOBAL_PARSER_CONTEXT.pushImpliedContext(context)
with nogil:
+ orig_loader = _register_document_loader()
if c_context:
result = xinclude.xmlXIncludeProcessTreeFlagsData(
node._c_node, parse_options, c_context)
else:
result = xinclude.xmlXIncludeProcessTree(node._c_node)
+ _reset_document_loader(orig_loader)
if tree.LIBXML_VERSION < 20704 or not c_context:
__GLOBAL_PARSER_CONTEXT.popImpliedContext()
self._error_log.disconnect()
diff --git a/src/lxml/xmlschema.pxi b/src/lxml/xmlschema.pxi
index cc2c1928d..ab26d935e 100644
--- a/src/lxml/xmlschema.pxi
+++ b/src/lxml/xmlschema.pxi
@@ -77,7 +77,9 @@ cdef class XMLSchema(_Validator):
# resolve requests to the document's parser
__GLOBAL_PARSER_CONTEXT.pushImpliedContextFromParser(self._doc._parser)
with nogil:
+ orig_loader = _register_document_loader()
self._c_schema = xmlschema.xmlSchemaParse(parser_ctxt)
+ _reset_document_loader(orig_loader)
if self._doc is not None:
__GLOBAL_PARSER_CONTEXT.popImpliedContext()
xmlschema.xmlSchemaFreeParserCtxt(parser_ctxt)
diff --git a/src/lxml/xslt.pxi b/src/lxml/xslt.pxi
index e7b49600c..d483cfa30 100644
--- a/src/lxml/xslt.pxi
+++ b/src/lxml/xslt.pxi
@@ -397,7 +397,9 @@ cdef class XSLT:
c_doc._private = <python.PyObject*>self._xslt_resolver_context
with self._error_log:
+ orig_loader = _register_document_loader()
c_style = xslt.xsltParseStylesheetDoc(c_doc)
+ _reset_document_loader(orig_loader)
if c_style is NULL or c_style.errors:
tree.xmlFreeDoc(c_doc)
@@ -633,8 +635,10 @@ cdef class XSLT:
if self._access_control is not None:
self._access_control._register_in_context(transform_ctxt)
with self._error_log, nogil:
+ orig_loader = _register_document_loader()
c_result = xslt.xsltApplyStylesheetUser(
self._c_style, c_input_doc, params, NULL, NULL, transform_ctxt)
+ _reset_document_loader(orig_loader)
return c_result

10
python-lxml.spec
View File

@ -1,13 +1,14 @@
%global modname lxml
Name: python-%{modname}
Version: 4.4.1
Release: 5%{?dist}
Version: 4.5.1
Release: 1%{?dist}
Summary: XML processing library combining libxml2/libxslt with the ElementTree API
License: BSD
URL: https://github.com/lxml/lxml
Source0: https://lxml.de/files/%{modname}-%{version}.tgz
Patch0001: https://github.com/lxml/lxml/commit/fa1d856cad369d0ac64323ddec14b02281491706.patch#/0001-Avoid-globally-overriding-the-libxml2-external-entit.patch
BuildRequires: gcc
BuildRequires: libxml2-devel
@ -37,7 +38,7 @@ Suggests: python%{python3_version}dist(beautifulsoup4)
Python 3 version.
%prep
%autosetup -n %{modname}-%{version}
%autosetup -n %{modname}-%{version} -p1
# Remove pregenerated Cython C sources
find -type f -name '*.c' -print -delete
@ -57,6 +58,9 @@ env WITH_CYTHON=true %py3_build
%{python3_sitearch}/%{modname}-*.egg-info/
%changelog
* Mon Jun 01 2020 Igor Raits <ignatenkobrain@fedoraproject.org> - 4.5.1-1
- Update to 4.5.1
* Fri May 22 2020 Miro Hrončok <mhroncok@redhat.com> - 4.4.1-5
- Rebuilt for Python 3.9

2
sources
View File

@ -1 +1 @@
SHA512 (lxml-4.4.1.tgz) = 3f11469290868f5bd30631020ac170c40da7348853609edf6fc6b00437b053fd774e0dfc6e711703ac5d05398dfa1f31e59a185935c3dc8ef0e1914a518bd049
SHA512 (lxml-4.5.1.tgz) = 5332d2b691b3d0def42b907a013fa814054dcd150ab6d7adf34891b70667b3e907f80c956a58671d7a31c63caa6110b2e007d2ed82b5fc706e6ad7b428742883