lxml 5.2.2__cp312-cp312-manylinux_2_28_ppc64le.whl → 5.3.1__cp312-cp312-manylinux_2_28_ppc64le.whl

lxml/lxml.etree_api.h

@@ -1,4 +1,4 @@
-/* Generated by Cython 3.0.10 */
+/* Generated by Cython 3.0.11 */
 
 #ifndef __PYX_HAVE_API__lxml__etree
 #define __PYX_HAVE_API__lxml__etree
@@ -98,9 +98,9 @@ static void (*__pyx_api_f_4lxml_5etree_initTagMatch)(struct LxmlElementTagMatche
 #define initTagMatch __pyx_api_f_4lxml_5etree_initTagMatch
 static xmlNs *(*__pyx_api_f_4lxml_5etree_findOrBuildNodeNsPrefix)(struct LxmlDocument *, xmlNode *, const xmlChar *, const xmlChar *) = 0;
 #define findOrBuildNodeNsPrefix __pyx_api_f_4lxml_5etree_findOrBuildNodeNsPrefix
-#ifndef __PYX_HAVE_RT_ImportFunction_3_0_10
-#define __PYX_HAVE_RT_ImportFunction_3_0_10
-static int __Pyx_ImportFunction_3_0_10(PyObject *module, const char *funcname, void (**f)(void), const char *sig) {
+#ifndef __PYX_HAVE_RT_ImportFunction_3_0_11
+#define __PYX_HAVE_RT_ImportFunction_3_0_11
+static int __Pyx_ImportFunction_3_0_11(PyObject *module, const char *funcname, void (**f)(void), const char *sig) {
     PyObject *d = 0;
     PyObject *cobj = 0;
     union {
@@ -140,51 +140,51 @@ static int import_lxml__etree(void) {
   PyObject *module = 0;
   module = PyImport_ImportModule("lxml.etree");
   if (!module) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "deepcopyNodeToDocument", (void (**)(void))&__pyx_api_f_4lxml_5etree_deepcopyNodeToDocument, "struct LxmlElement *(struct LxmlDocument *, xmlNode *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "elementTreeFactory", (void (**)(void))&__pyx_api_f_4lxml_5etree_elementTreeFactory, "struct LxmlElementTree *(struct LxmlElement *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "newElementTree", (void (**)(void))&__pyx_api_f_4lxml_5etree_newElementTree, "struct LxmlElementTree *(struct LxmlElement *, PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "adoptExternalDocument", (void (**)(void))&__pyx_api_f_4lxml_5etree_adoptExternalDocument, "struct LxmlElementTree *(xmlDoc *, PyObject *, int)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "elementFactory", (void (**)(void))&__pyx_api_f_4lxml_5etree_elementFactory, "struct LxmlElement *(struct LxmlDocument *, xmlNode *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "makeElement", (void (**)(void))&__pyx_api_f_4lxml_5etree_makeElement, "struct LxmlElement *(PyObject *, struct LxmlDocument *, PyObject *, PyObject *, PyObject *, PyObject *, PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "makeSubElement", (void (**)(void))&__pyx_api_f_4lxml_5etree_makeSubElement, "struct LxmlElement *(struct LxmlElement *, PyObject *, PyObject *, PyObject *, PyObject *, PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "setElementClassLookupFunction", (void (**)(void))&__pyx_api_f_4lxml_5etree_setElementClassLookupFunction, "void (_element_class_lookup_function, PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "lookupDefaultElementClass", (void (**)(void))&__pyx_api_f_4lxml_5etree_lookupDefaultElementClass, "PyObject *(PyObject *, PyObject *, xmlNode *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "lookupNamespaceElementClass", (void (**)(void))&__pyx_api_f_4lxml_5etree_lookupNamespaceElementClass, "PyObject *(PyObject *, PyObject *, xmlNode *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "callLookupFallback", (void (**)(void))&__pyx_api_f_4lxml_5etree_callLookupFallback, "PyObject *(struct LxmlFallbackElementClassLookup *, struct LxmlDocument *, xmlNode *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "tagMatches", (void (**)(void))&__pyx_api_f_4lxml_5etree_tagMatches, "int (xmlNode *, const xmlChar *, const xmlChar *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "documentOrRaise", (void (**)(void))&__pyx_api_f_4lxml_5etree_documentOrRaise, "struct LxmlDocument *(PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "rootNodeOrRaise", (void (**)(void))&__pyx_api_f_4lxml_5etree_rootNodeOrRaise, "struct LxmlElement *(PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "hasText", (void (**)(void))&__pyx_api_f_4lxml_5etree_hasText, "int (xmlNode *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "hasTail", (void (**)(void))&__pyx_api_f_4lxml_5etree_hasTail, "int (xmlNode *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "textOf", (void (**)(void))&__pyx_api_f_4lxml_5etree_textOf, "PyObject *(xmlNode *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "tailOf", (void (**)(void))&__pyx_api_f_4lxml_5etree_tailOf, "PyObject *(xmlNode *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "setNodeText", (void (**)(void))&__pyx_api_f_4lxml_5etree_setNodeText, "int (xmlNode *, PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "setTailText", (void (**)(void))&__pyx_api_f_4lxml_5etree_setTailText, "int (xmlNode *, PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "attributeValue", (void (**)(void))&__pyx_api_f_4lxml_5etree_attributeValue, "PyObject *(xmlNode *, xmlAttr *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "attributeValueFromNsName", (void (**)(void))&__pyx_api_f_4lxml_5etree_attributeValueFromNsName, "PyObject *(xmlNode *, const xmlChar *, const xmlChar *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "getAttributeValue", (void (**)(void))&__pyx_api_f_4lxml_5etree_getAttributeValue, "PyObject *(struct LxmlElement *, PyObject *, PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "iterattributes", (void (**)(void))&__pyx_api_f_4lxml_5etree_iterattributes, "PyObject *(struct LxmlElement *, int)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "collectAttributes", (void (**)(void))&__pyx_api_f_4lxml_5etree_collectAttributes, "PyObject *(xmlNode *, int)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "setAttributeValue", (void (**)(void))&__pyx_api_f_4lxml_5etree_setAttributeValue, "int (struct LxmlElement *, PyObject *, PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "delAttribute", (void (**)(void))&__pyx_api_f_4lxml_5etree_delAttribute, "int (struct LxmlElement *, PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "delAttributeFromNsName", (void (**)(void))&__pyx_api_f_4lxml_5etree_delAttributeFromNsName, "int (xmlNode *, const xmlChar *, const xmlChar *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "hasChild", (void (**)(void))&__pyx_api_f_4lxml_5etree_hasChild, "int (xmlNode *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "findChild", (void (**)(void))&__pyx_api_f_4lxml_5etree_findChild, "xmlNode *(xmlNode *, Py_ssize_t)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "findChildForwards", (void (**)(void))&__pyx_api_f_4lxml_5etree_findChildForwards, "xmlNode *(xmlNode *, Py_ssize_t)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "findChildBackwards", (void (**)(void))&__pyx_api_f_4lxml_5etree_findChildBackwards, "xmlNode *(xmlNode *, Py_ssize_t)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "nextElement", (void (**)(void))&__pyx_api_f_4lxml_5etree_nextElement, "xmlNode *(xmlNode *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "previousElement", (void (**)(void))&__pyx_api_f_4lxml_5etree_previousElement, "xmlNode *(xmlNode *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "appendChild", (void (**)(void))&__pyx_api_f_4lxml_5etree_appendChild, "void (struct LxmlElement *, struct LxmlElement *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "appendChildToElement", (void (**)(void))&__pyx_api_f_4lxml_5etree_appendChildToElement, "int (struct LxmlElement *, struct LxmlElement *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "pyunicode", (void (**)(void))&__pyx_api_f_4lxml_5etree_pyunicode, "PyObject *(const xmlChar *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "utf8", (void (**)(void))&__pyx_api_f_4lxml_5etree_utf8, "PyObject *(PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "getNsTag", (void (**)(void))&__pyx_api_f_4lxml_5etree_getNsTag, "PyObject *(PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "getNsTagWithEmptyNs", (void (**)(void))&__pyx_api_f_4lxml_5etree_getNsTagWithEmptyNs, "PyObject *(PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "namespacedName", (void (**)(void))&__pyx_api_f_4lxml_5etree_namespacedName, "PyObject *(xmlNode *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "namespacedNameFromNsName", (void (**)(void))&__pyx_api_f_4lxml_5etree_namespacedNameFromNsName, "PyObject *(const xmlChar *, const xmlChar *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "iteratorStoreNext", (void (**)(void))&__pyx_api_f_4lxml_5etree_iteratorStoreNext, "void (struct LxmlElementIterator *, struct LxmlElement *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "initTagMatch", (void (**)(void))&__pyx_api_f_4lxml_5etree_initTagMatch, "void (struct LxmlElementTagMatcher *, PyObject *)") < 0) goto bad;
-  if (__Pyx_ImportFunction_3_0_10(module, "findOrBuildNodeNsPrefix", (void (**)(void))&__pyx_api_f_4lxml_5etree_findOrBuildNodeNsPrefix, "xmlNs *(struct LxmlDocument *, xmlNode *, const xmlChar *, const xmlChar *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "deepcopyNodeToDocument", (void (**)(void))&__pyx_api_f_4lxml_5etree_deepcopyNodeToDocument, "struct LxmlElement *(struct LxmlDocument *, xmlNode *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "elementTreeFactory", (void (**)(void))&__pyx_api_f_4lxml_5etree_elementTreeFactory, "struct LxmlElementTree *(struct LxmlElement *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "newElementTree", (void (**)(void))&__pyx_api_f_4lxml_5etree_newElementTree, "struct LxmlElementTree *(struct LxmlElement *, PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "adoptExternalDocument", (void (**)(void))&__pyx_api_f_4lxml_5etree_adoptExternalDocument, "struct LxmlElementTree *(xmlDoc *, PyObject *, int)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "elementFactory", (void (**)(void))&__pyx_api_f_4lxml_5etree_elementFactory, "struct LxmlElement *(struct LxmlDocument *, xmlNode *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "makeElement", (void (**)(void))&__pyx_api_f_4lxml_5etree_makeElement, "struct LxmlElement *(PyObject *, struct LxmlDocument *, PyObject *, PyObject *, PyObject *, PyObject *, PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "makeSubElement", (void (**)(void))&__pyx_api_f_4lxml_5etree_makeSubElement, "struct LxmlElement *(struct LxmlElement *, PyObject *, PyObject *, PyObject *, PyObject *, PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "setElementClassLookupFunction", (void (**)(void))&__pyx_api_f_4lxml_5etree_setElementClassLookupFunction, "void (_element_class_lookup_function, PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "lookupDefaultElementClass", (void (**)(void))&__pyx_api_f_4lxml_5etree_lookupDefaultElementClass, "PyObject *(PyObject *, PyObject *, xmlNode *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "lookupNamespaceElementClass", (void (**)(void))&__pyx_api_f_4lxml_5etree_lookupNamespaceElementClass, "PyObject *(PyObject *, PyObject *, xmlNode *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "callLookupFallback", (void (**)(void))&__pyx_api_f_4lxml_5etree_callLookupFallback, "PyObject *(struct LxmlFallbackElementClassLookup *, struct LxmlDocument *, xmlNode *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "tagMatches", (void (**)(void))&__pyx_api_f_4lxml_5etree_tagMatches, "int (xmlNode *, const xmlChar *, const xmlChar *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "documentOrRaise", (void (**)(void))&__pyx_api_f_4lxml_5etree_documentOrRaise, "struct LxmlDocument *(PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "rootNodeOrRaise", (void (**)(void))&__pyx_api_f_4lxml_5etree_rootNodeOrRaise, "struct LxmlElement *(PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "hasText", (void (**)(void))&__pyx_api_f_4lxml_5etree_hasText, "int (xmlNode *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "hasTail", (void (**)(void))&__pyx_api_f_4lxml_5etree_hasTail, "int (xmlNode *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "textOf", (void (**)(void))&__pyx_api_f_4lxml_5etree_textOf, "PyObject *(xmlNode *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "tailOf", (void (**)(void))&__pyx_api_f_4lxml_5etree_tailOf, "PyObject *(xmlNode *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "setNodeText", (void (**)(void))&__pyx_api_f_4lxml_5etree_setNodeText, "int (xmlNode *, PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "setTailText", (void (**)(void))&__pyx_api_f_4lxml_5etree_setTailText, "int (xmlNode *, PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "attributeValue", (void (**)(void))&__pyx_api_f_4lxml_5etree_attributeValue, "PyObject *(xmlNode *, xmlAttr *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "attributeValueFromNsName", (void (**)(void))&__pyx_api_f_4lxml_5etree_attributeValueFromNsName, "PyObject *(xmlNode *, const xmlChar *, const xmlChar *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "getAttributeValue", (void (**)(void))&__pyx_api_f_4lxml_5etree_getAttributeValue, "PyObject *(struct LxmlElement *, PyObject *, PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "iterattributes", (void (**)(void))&__pyx_api_f_4lxml_5etree_iterattributes, "PyObject *(struct LxmlElement *, int)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "collectAttributes", (void (**)(void))&__pyx_api_f_4lxml_5etree_collectAttributes, "PyObject *(xmlNode *, int)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "setAttributeValue", (void (**)(void))&__pyx_api_f_4lxml_5etree_setAttributeValue, "int (struct LxmlElement *, PyObject *, PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "delAttribute", (void (**)(void))&__pyx_api_f_4lxml_5etree_delAttribute, "int (struct LxmlElement *, PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "delAttributeFromNsName", (void (**)(void))&__pyx_api_f_4lxml_5etree_delAttributeFromNsName, "int (xmlNode *, const xmlChar *, const xmlChar *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "hasChild", (void (**)(void))&__pyx_api_f_4lxml_5etree_hasChild, "int (xmlNode *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "findChild", (void (**)(void))&__pyx_api_f_4lxml_5etree_findChild, "xmlNode *(xmlNode *, Py_ssize_t)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "findChildForwards", (void (**)(void))&__pyx_api_f_4lxml_5etree_findChildForwards, "xmlNode *(xmlNode *, Py_ssize_t)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "findChildBackwards", (void (**)(void))&__pyx_api_f_4lxml_5etree_findChildBackwards, "xmlNode *(xmlNode *, Py_ssize_t)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "nextElement", (void (**)(void))&__pyx_api_f_4lxml_5etree_nextElement, "xmlNode *(xmlNode *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "previousElement", (void (**)(void))&__pyx_api_f_4lxml_5etree_previousElement, "xmlNode *(xmlNode *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "appendChild", (void (**)(void))&__pyx_api_f_4lxml_5etree_appendChild, "void (struct LxmlElement *, struct LxmlElement *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "appendChildToElement", (void (**)(void))&__pyx_api_f_4lxml_5etree_appendChildToElement, "int (struct LxmlElement *, struct LxmlElement *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "pyunicode", (void (**)(void))&__pyx_api_f_4lxml_5etree_pyunicode, "PyObject *(const xmlChar *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "utf8", (void (**)(void))&__pyx_api_f_4lxml_5etree_utf8, "PyObject *(PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "getNsTag", (void (**)(void))&__pyx_api_f_4lxml_5etree_getNsTag, "PyObject *(PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "getNsTagWithEmptyNs", (void (**)(void))&__pyx_api_f_4lxml_5etree_getNsTagWithEmptyNs, "PyObject *(PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "namespacedName", (void (**)(void))&__pyx_api_f_4lxml_5etree_namespacedName, "PyObject *(xmlNode *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "namespacedNameFromNsName", (void (**)(void))&__pyx_api_f_4lxml_5etree_namespacedNameFromNsName, "PyObject *(const xmlChar *, const xmlChar *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "iteratorStoreNext", (void (**)(void))&__pyx_api_f_4lxml_5etree_iteratorStoreNext, "void (struct LxmlElementIterator *, struct LxmlElement *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "initTagMatch", (void (**)(void))&__pyx_api_f_4lxml_5etree_initTagMatch, "void (struct LxmlElementTagMatcher *, PyObject *)") < 0) goto bad;
+  if (__Pyx_ImportFunction_3_0_11(module, "findOrBuildNodeNsPrefix", (void (**)(void))&__pyx_api_f_4lxml_5etree_findOrBuildNodeNsPrefix, "xmlNs *(struct LxmlDocument *, xmlNode *, const xmlChar *, const xmlChar *)") < 0) goto bad;
   Py_DECREF(module); module = 0;
   return 0;
   bad:

lxml/parser.pxi

@@ -1747,9 +1747,11 @@ _HTML_DEFAULT_PARSE_OPTIONS = (
     htmlparser.HTML_PARSE_COMPACT
     )
 
+cdef object _UNUSED = object()
+
 cdef class HTMLParser(_FeedParser):
     """HTMLParser(self, encoding=None, remove_blank_text=False, \
-                   remove_comments=False, remove_pis=False, strip_cdata=True, \
+                   remove_comments=False, remove_pis=False, \
                    no_network=True, target=None, schema: XMLSchema =None, \
                    recover=True, compact=True, collect_ids=True, huge_tree=False)
 
@@ -1767,7 +1769,6 @@ cdef class HTMLParser(_FeedParser):
     - remove_blank_text  - discard empty text nodes that are ignorable (i.e. not actual text content)
     - remove_comments    - discard comments
     - remove_pis         - discard processing instructions
-    - strip_cdata        - replace CDATA sections by normal text content (default: True)
     - compact            - save memory for short text content (default: True)
     - default_doctype    - add a default doctype even if it is not found in the HTML (default: True)
     - collect_ids        - use a hash table of XML IDs for fast access (default: True)
@@ -1784,7 +1785,7 @@ cdef class HTMLParser(_FeedParser):
     reasons.
     """
     def __init__(self, *, encoding=None, remove_blank_text=False,
-                 remove_comments=False, remove_pis=False, strip_cdata=True,
+                 remove_comments=False, remove_pis=False, strip_cdata=_UNUSED,
                  no_network=True, target=None, XMLSchema schema=None,
                  recover=True, compact=True, default_doctype=True,
                  collect_ids=True, huge_tree=False):
@@ -1803,6 +1804,11 @@ cdef class HTMLParser(_FeedParser):
         if huge_tree:
             parse_options = parse_options | xmlparser.XML_PARSE_HUGE
 
+        if strip_cdata is not _UNUSED:
+            import warnings
+            warnings.warn(
+                "The 'strip_cdata' option of HTMLParser() has never done anything and will eventually be removed.",
+                DeprecationWarning)
         _BaseParser.__init__(self, parse_options, True, schema,
                              remove_comments, remove_pis, strip_cdata,
                              collect_ids, target, encoding)

lxml/serializer.pxi

@@ -403,53 +403,10 @@ cdef void _writeNextSiblings(tree.xmlOutputBuffer* c_buffer, xmlNode* c_node,
         c_sibling = c_sibling.next
 
 
-# copied and adapted from libxml2
-cdef unsigned char *xmlSerializeHexCharRef(unsigned char *out, int val) noexcept:
-    cdef xmlChar *ptr
-    cdef const xmlChar* hexdigits = b"0123456789ABCDEF"
-
-    out[0] = b'&'
-    out += 1
-    out[0] = b'#'
-    out += 1
-    out[0] = b'x'
-    out += 1
-
-    if val < 0x10:
-        ptr = out
-    elif val < 0x100:
-        ptr = out + 1
-    elif val < 0x1000:
-        ptr = out + 2
-    elif val < 0x10000:
-        ptr = out + 3
-    elif val < 0x100000:
-        ptr = out + 4
-    else:
-        ptr = out + 5
-
-    out = ptr + 1
-    while val > 0:
-        ptr[0] = hexdigits[val & 0xF]
-        ptr -= 1
-        val >>= 4
-
-    out[0] = b';'
-    out += 1
-    out[0] = 0
-
-    return out
-
-
 # copied and adapted from libxml2 (xmlBufAttrSerializeTxtContent())
 cdef _write_attr_string(tree.xmlOutputBuffer* buf, const char *string):
     cdef const char *base
     cdef const char *cur
-    cdef const unsigned char *ucur
-
-    cdef unsigned char tmp[12]
-    cdef int val = 0
-    cdef int l
 
     if string == NULL:
         return
@@ -511,57 +468,8 @@ cdef _write_attr_string(tree.xmlOutputBuffer* buf, const char *string):
             cur += 1
             base = cur
 
-        elif (<const unsigned char>cur[0] >= 0x80) and (cur[1] != 0):
-
-            if base != cur:
-                tree.xmlOutputBufferWrite(buf, cur - base, base)
-
-            ucur = <const unsigned char *>cur
-
-            if ucur[0] < 0xC0:
-                # invalid UTF-8 sequence
-                val = ucur[0]
-                l = 1
-
-            elif ucur[0] < 0xE0:
-                val = (ucur[0]) & 0x1F
-                val <<= 6
-                val |= (ucur[1]) & 0x3F
-                l = 2
-
-            elif (ucur[0] < 0xF0) and (ucur[2] != 0):
-                val = (ucur[0]) & 0x0F
-                val <<= 6
-                val |= (ucur[1]) & 0x3F
-                val <<= 6
-                val |= (ucur[2]) & 0x3F
-                l = 3
-
-            elif (ucur[0] < 0xF8) and (ucur[2] != 0) and (ucur[3] != 0):
-                val = (ucur[0]) & 0x07
-                val <<= 6
-                val |= (ucur[1]) & 0x3F
-                val <<= 6
-                val |= (ucur[2]) & 0x3F
-                val <<= 6
-                val |= (ucur[3]) & 0x3F
-                l = 4
-            else:
-                # invalid UTF-8 sequence
-                val = ucur[0]
-                l = 1
-
-            if (l == 1) or (not tree.xmlIsCharQ(val)):
-                raise ValueError(f"Invalid character: {val:X}")
-
-            # We could do multiple things here. Just save
-            # as a char ref
-            xmlSerializeHexCharRef(tmp, val)
-            tree.xmlOutputBufferWrite(buf, len(tmp), <const char*> tmp)
-            cur += l
-            base = cur
-
         else:
+            # Leave further encoding and escaping to the buffer encoder.
             cur += 1
 
     if base != cur:
@@ -1412,9 +1320,11 @@ cdef class _IncrementalFileWriter:
         self._status = WRITER_STARTING
         self._element_stack = []
         if encoding is None:
+            # We always need a document encoding to make the attribute serialisation
+            # of libxml2 identical to ours.
             encoding = b'ASCII'
         self._encoding = encoding
-        self._c_encoding = _cstr(encoding) if encoding is not None else NULL
+        self._c_encoding = _cstr(encoding)
         self._buffered = buffered
         self._target = _create_output_buffer(
             outfile, self._c_encoding, compresslevel, &self._c_out, close)

{lxml-5.2.2.dist-info → lxml-5.3.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.2
 Name: lxml
-Version: 5.2.2
+Version: 5.3.1
 Summary: Powerful and Pythonic XML processing library combining libxml2/libxslt with the ElementTree API.
 Home-page: https://lxml.de/
 Author: lxml dev team
@@ -30,16 +30,28 @@ Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Requires-Python: >=3.6
 License-File: LICENSE.txt
 License-File: LICENSES.txt
+Provides-Extra: source
+Requires-Dist: Cython<3.1.0,>=3.0.11; extra == "source"
 Provides-Extra: cssselect
-Requires-Dist: cssselect >=0.7 ; extra == 'cssselect'
+Requires-Dist: cssselect>=0.7; extra == "cssselect"
 Provides-Extra: html5
-Requires-Dist: html5lib ; extra == 'html5'
-Provides-Extra: html_clean
-Requires-Dist: lxml-html-clean ; extra == 'html_clean'
+Requires-Dist: html5lib; extra == "html5"
 Provides-Extra: htmlsoup
-Requires-Dist: BeautifulSoup4 ; extra == 'htmlsoup'
-Provides-Extra: source
-Requires-Dist: Cython >=3.0.10 ; extra == 'source'
+Requires-Dist: BeautifulSoup4; extra == "htmlsoup"
+Provides-Extra: html-clean
+Requires-Dist: lxml_html_clean; extra == "html-clean"
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: home-page
+Dynamic: license
+Dynamic: maintainer
+Dynamic: maintainer-email
+Dynamic: project-url
+Dynamic: provides-extra
+Dynamic: requires-python
+Dynamic: summary
 
 lxml is a Pythonic, mature binding for the libxml2 and libxslt libraries.  It
 provides safe and convenient access to these libraries using the ElementTree
@@ -64,25 +76,29 @@ an appropriate version of Cython installed.
 
 After an official release of a new stable series, bug fixes may become
 available at
-https://github.com/lxml/lxml/tree/lxml-5.2 .
-Running ``easy_install lxml==5.2bugfix`` will install
+https://github.com/lxml/lxml/tree/lxml-5.3 .
+Running ``easy_install lxml==5.3bugfix`` will install
 the unreleased branch state from
-https://github.com/lxml/lxml/tarball/lxml-5.2#egg=lxml-5.2bugfix
+https://github.com/lxml/lxml/tarball/lxml-5.3#egg=lxml-5.3bugfix
 as soon as a maintenance branch has been established.  Note that this
 requires Cython to be installed at an appropriate version for the build.
 
-5.2.2 (2024-05-12)
+5.3.1 (2025-02-09)
 ==================
 
 Bugs fixed
 ----------
 
-* GH#417: The ``test_feed_parser`` test could fail if ``lxml_html_clean`` was not installed.
-  It is now skipped in that case.
+* GH#440: Some tests were adapted for libxml2 2.14.0.
+  Patch by Nick Wellnhofer.
+
+* LP#2097175: ``DTD(external_id="…")`` erroneously required a byte string as ID value.
+
+* GH#450: ``iterparse()`` internally triggered the `DeprecationWarning`` added in lxml 5.3.0 when parsing HTML.
 
-* LP#2059910: The minimum CPU architecture for the Linux x86 binary wheels was set back to
-  "core2", without SSE 4.2.
+Other changes
+-------------
 
-* If libxml2 uses iconv, the compile time version is available as `etree.ICONV_COMPILED_VERSION`.
+* GH#442: Binary wheels for macOS no longer use the linker flag ``-flat_namespace``.