[RFC] DOM HTML5 parsing and serialization support (#12111)

NEWS

@@ -10,6 +10,7 @@ DOM:
   . Implement #53655 (Improve speed of DOMNode::C14N() on large XML documents).
     (nielsdos)
   . Fix cloning attribute with namespace disappearing namespace. (nielsdos)
+  . Implement DOM HTML5 parsing and serialization RFC. (nielsdos)
 
 FTP:
   . Removed the deprecated inet_ntoa call support. (David Carlier)

UPGRADING

@@ -80,6 +80,14 @@ PHP 8.4 UPGRADE NOTES
   . Added constant DOMNode::DOCUMENT_POSITION_CONTAINS.
   . Added constant DOMNode::DOCUMENT_POSITION_CONTAINED_BY.
   . Added constant DOMNode::DOCUMENT_POSITION_IMPLEMENTATION_SPECIFIC.
+  . Implemented DOM HTML5 parsing and serialization.
+    RFC: https://wiki.php.net/rfc/domdocument_html5_parser.
+    This RFC adds the new DOM namespace along with class and constant aliases.
+    There are two new classes to handle HTML and XML documents:
+    DOM\HTMLDocument and DOM\XMLDocument.
+    These classes provide a cleaner API to handle HTML and XML documents.
+    Furthermore, the DOM\HTMLDocument class implements spec-compliant HTML5
+    parsing and serialization.
 
 - Phar:
   . Added support for the unix timestamp extension for zip archives.

UPGRADING.INTERNALS

@@ -52,6 +52,10 @@ PHP 8.4 INTERNALS UPGRADE NOTES
    - The function php_xsl_create_object() was removed as it was not used
      nor exported.
 
+ d. ext/libxml
+   - Added php_libxml_pretend_ctx_error_ex() to emit errors as if they had come
+     from libxml.
+
 ========================
 4. OpCode changes
 ========================

ext/dom/config.m4

@@ -12,7 +12,21 @@ if test "$PHP_DOM" != "no"; then
 
   PHP_SETUP_LIBXML(DOM_SHARED_LIBADD, [
     AC_DEFINE(HAVE_DOM,1,[ ])
+    PHP_LEXBOR_CFLAGS="-I@ext_srcdir@/lexbor -DLEXBOR_STATIC"
+    LEXBOR_DIR="lexbor/lexbor"
+    LEXBOR_SOURCES="$LEXBOR_DIR/ports/posix/lexbor/core/memory.c \
+      $LEXBOR_DIR/core/array_obj.c $LEXBOR_DIR/core/array.c $LEXBOR_DIR/core/avl.c $LEXBOR_DIR/core/bst.c $LEXBOR_DIR/core/diyfp.c $LEXBOR_DIR/core/conv.c $LEXBOR_DIR/core/dobject.c $LEXBOR_DIR/core/dtoa.c $LEXBOR_DIR/core/hash.c $LEXBOR_DIR/core/mem.c $LEXBOR_DIR/core/mraw.c $LEXBOR_DIR/core/print.c $LEXBOR_DIR/core/serialize.c $LEXBOR_DIR/core/shs.c $LEXBOR_DIR/core/str.c $LEXBOR_DIR/core/strtod.c \
+      $LEXBOR_DIR/dom/interface.c $LEXBOR_DIR/dom/interfaces/attr.c $LEXBOR_DIR/dom/interfaces/cdata_section.c $LEXBOR_DIR/dom/interfaces/character_data.c $LEXBOR_DIR/dom/interfaces/comment.c $LEXBOR_DIR/dom/interfaces/document.c $LEXBOR_DIR/dom/interfaces/document_fragment.c $LEXBOR_DIR/dom/interfaces/document_type.c $LEXBOR_DIR/dom/interfaces/element.c $LEXBOR_DIR/dom/interfaces/node.c $LEXBOR_DIR/dom/interfaces/processing_instruction.c $LEXBOR_DIR/dom/interfaces/shadow_root.c $LEXBOR_DIR/dom/interfaces/text.c \
+      $LEXBOR_DIR/html/tokenizer/error.c $LEXBOR_DIR/html/tokenizer/state_comment.c $LEXBOR_DIR/html/tokenizer/state_doctype.c $LEXBOR_DIR/html/tokenizer/state_rawtext.c $LEXBOR_DIR/html/tokenizer/state_rcdata.c $LEXBOR_DIR/html/tokenizer/state_script.c $LEXBOR_DIR/html/tokenizer/state.c \
+      $LEXBOR_DIR/html/tree/active_formatting.c $LEXBOR_DIR/html/tree/error.c $LEXBOR_DIR/html/tree/insertion_mode/after_after_body.c $LEXBOR_DIR/html/tree/insertion_mode/after_after_frameset.c $LEXBOR_DIR/html/tree/insertion_mode/after_body.c $LEXBOR_DIR/html/tree/insertion_mode/after_frameset.c $LEXBOR_DIR/html/tree/insertion_mode/after_head.c $LEXBOR_DIR/html/tree/insertion_mode/before_head.c $LEXBOR_DIR/html/tree/insertion_mode/before_html.c $LEXBOR_DIR/html/tree/insertion_mode/foreign_content.c $LEXBOR_DIR/html/tree/insertion_mode/in_body.c $LEXBOR_DIR/html/tree/insertion_mode/in_caption.c $LEXBOR_DIR/html/tree/insertion_mode/in_cell.c $LEXBOR_DIR/html/tree/insertion_mode/in_column_group.c $LEXBOR_DIR/html/tree/insertion_mode/in_frameset.c $LEXBOR_DIR/html/tree/insertion_mode/in_head.c $LEXBOR_DIR/html/tree/insertion_mode/in_head_noscript.c $LEXBOR_DIR/html/tree/insertion_mode/initial.c $LEXBOR_DIR/html/tree/insertion_mode/in_row.c $LEXBOR_DIR/html/tree/insertion_mode/in_select.c $LEXBOR_DIR/html/tree/insertion_mode/in_select_in_table.c $LEXBOR_DIR/html/tree/insertion_mode/in_table_body.c $LEXBOR_DIR/html/tree/insertion_mode/in_table.c $LEXBOR_DIR/html/tree/insertion_mode/in_table_text.c $LEXBOR_DIR/html/tree/insertion_mode/in_template.c $LEXBOR_DIR/html/tree/insertion_mode/text.c $LEXBOR_DIR/html/tree/open_elements.c \
+      $LEXBOR_DIR/encoding/big5.c $LEXBOR_DIR/encoding/decode.c $LEXBOR_DIR/encoding/encode.c $LEXBOR_DIR/encoding/encoding.c $LEXBOR_DIR/encoding/euc_kr.c $LEXBOR_DIR/encoding/gb18030.c $LEXBOR_DIR/encoding/iso_2022_jp_katakana.c $LEXBOR_DIR/encoding/jis0208.c $LEXBOR_DIR/encoding/jis0212.c $LEXBOR_DIR/encoding/range.c $LEXBOR_DIR/encoding/res.c $LEXBOR_DIR/encoding/single.c \
+      $LEXBOR_DIR/html/encoding.c $LEXBOR_DIR/html/interface.c $LEXBOR_DIR/html/parser.c $LEXBOR_DIR/html/token.c $LEXBOR_DIR/html/token_attr.c $LEXBOR_DIR/html/tokenizer.c $LEXBOR_DIR/html/tree.c \
+      $LEXBOR_DIR/html/interfaces/anchor_element.c $LEXBOR_DIR/html/interfaces/area_element.c $LEXBOR_DIR/html/interfaces/audio_element.c $LEXBOR_DIR/html/interfaces/base_element.c $LEXBOR_DIR/html/interfaces/body_element.c $LEXBOR_DIR/html/interfaces/br_element.c $LEXBOR_DIR/html/interfaces/button_element.c $LEXBOR_DIR/html/interfaces/canvas_element.c $LEXBOR_DIR/html/interfaces/data_element.c $LEXBOR_DIR/html/interfaces/data_list_element.c $LEXBOR_DIR/html/interfaces/details_element.c $LEXBOR_DIR/html/interfaces/dialog_element.c $LEXBOR_DIR/html/interfaces/directory_element.c $LEXBOR_DIR/html/interfaces/div_element.c $LEXBOR_DIR/html/interfaces/d_list_element.c $LEXBOR_DIR/html/interfaces/document.c $LEXBOR_DIR/html/interfaces/element.c $LEXBOR_DIR/html/interfaces/embed_element.c $LEXBOR_DIR/html/interfaces/field_set_element.c $LEXBOR_DIR/html/interfaces/font_element.c $LEXBOR_DIR/html/interfaces/form_element.c $LEXBOR_DIR/html/interfaces/frame_element.c $LEXBOR_DIR/html/interfaces/frame_set_element.c $LEXBOR_DIR/html/interfaces/head_element.c $LEXBOR_DIR/html/interfaces/heading_element.c $LEXBOR_DIR/html/interfaces/hr_element.c $LEXBOR_DIR/html/interfaces/html_element.c $LEXBOR_DIR/html/interfaces/iframe_element.c $LEXBOR_DIR/html/interfaces/image_element.c $LEXBOR_DIR/html/interfaces/input_element.c $LEXBOR_DIR/html/interfaces/label_element.c $LEXBOR_DIR/html/interfaces/legend_element.c $LEXBOR_DIR/html/interfaces/li_element.c $LEXBOR_DIR/html/interfaces/link_element.c $LEXBOR_DIR/html/interfaces/map_element.c $LEXBOR_DIR/html/interfaces/marquee_element.c $LEXBOR_DIR/html/interfaces/media_element.c $LEXBOR_DIR/html/interfaces/menu_element.c $LEXBOR_DIR/html/interfaces/meta_element.c $LEXBOR_DIR/html/interfaces/meter_element.c $LEXBOR_DIR/html/interfaces/mod_element.c $LEXBOR_DIR/html/interfaces/object_element.c $LEXBOR_DIR/html/interfaces/o_list_element.c $LEXBOR_DIR/html/interfaces/opt_group_element.c $LEXBOR_DIR/html/interfaces/option_element.c $LEXBOR_DIR/html/interfaces/output_element.c $LEXBOR_DIR/html/interfaces/paragraph_element.c $LEXBOR_DIR/html/interfaces/param_element.c $LEXBOR_DIR/html/interfaces/picture_element.c $LEXBOR_DIR/html/interfaces/pre_element.c $LEXBOR_DIR/html/interfaces/progress_element.c $LEXBOR_DIR/html/interfaces/quote_element.c $LEXBOR_DIR/html/interfaces/script_element.c $LEXBOR_DIR/html/interfaces/select_element.c $LEXBOR_DIR/html/interfaces/slot_element.c $LEXBOR_DIR/html/interfaces/source_element.c $LEXBOR_DIR/html/interfaces/span_element.c $LEXBOR_DIR/html/interfaces/style_element.c $LEXBOR_DIR/html/interfaces/table_caption_element.c $LEXBOR_DIR/html/interfaces/table_cell_element.c $LEXBOR_DIR/html/interfaces/table_col_element.c $LEXBOR_DIR/html/interfaces/table_element.c $LEXBOR_DIR/html/interfaces/table_row_element.c $LEXBOR_DIR/html/interfaces/table_section_element.c $LEXBOR_DIR/html/interfaces/template_element.c $LEXBOR_DIR/html/interfaces/text_area_element.c $LEXBOR_DIR/html/interfaces/time_element.c $LEXBOR_DIR/html/interfaces/title_element.c $LEXBOR_DIR/html/interfaces/track_element.c $LEXBOR_DIR/html/interfaces/u_list_element.c $LEXBOR_DIR/html/interfaces/unknown_element.c $LEXBOR_DIR/html/interfaces/video_element.c $LEXBOR_DIR/html/interfaces/window.c \
+      $LEXBOR_DIR/selectors/selectors.c \
+      $LEXBOR_DIR/ns/ns.c \
+      $LEXBOR_DIR/tag/tag.c"
     PHP_NEW_EXTENSION(dom, [php_dom.c attr.c document.c \
+                            xml_document.c html_document.c html5_serializer.c html5_parser.c namespace_compat.c \
                             domexception.c parentnode.c \
                             processinginstruction.c cdatasection.c \
                             documentfragment.c domimplementation.c \
@@ -21,8 +35,9 @@ if test "$PHP_DOM" != "no"; then
                             nodelist.c text.c comment.c \
                             entityreference.c \
                             notation.c xpath.c dom_iterators.c \
-                            namednodemap.c],
+                            namednodemap.c \
-                            $ext_shared)
+                            $LEXBOR_SOURCES],
+                            $ext_shared,,$PHP_LEXBOR_CFLAGS)
     PHP_SUBST(DOM_SHARED_LIBADD)
     PHP_INSTALL_HEADERS([ext/dom/xml_common.h])
     PHP_ADD_EXTENSION_DEP(dom, libxml)

ext/dom/config.w32

@@ -8,13 +8,29 @@ if (PHP_DOM == "yes") {
 		CHECK_HEADER_ADD_INCLUDE("libxml/parser.h", "CFLAGS_DOM", PHP_PHP_BUILD + "\\include\\libxml2")
 	) {
 		EXTENSION("dom", "php_dom.c attr.c document.c \
+			xml_document.c html_document.c html5_serializer.c html5_parser.c namespace_compat.c \
 			domexception.c parentnode.c processinginstruction.c \
 			cdatasection.c documentfragment.c domimplementation.c element.c \
 			node.c characterdata.c documenttype.c \
 			entity.c nodelist.c text.c comment.c \
 			entityreference.c \
 			notation.c xpath.c dom_iterators.c \
-			namednodemap.c");
+			namednodemap.c", null, "-Iext/dom/lexbor");
+
+		ADD_SOURCES("ext/dom/lexbor/lexbor/ports/windows_nt/lexbor/core", "memory.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/core", "array_obj.c array.c avl.c bst.c diyfp.c conv.c dobject.c dtoa.c hash.c mem.c mraw.c print.c serialize.c shs.c str.c strtod.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/dom", "interface.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/dom/interfaces", "attr.c cdata_section.c character_data.c comment.c document.c document_fragment.c document_type.c element.c node.c processing_instruction.c shadow_root.c text.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/html/tokenizer", "error.c state_comment.c state_doctype.c state_rawtext.c state_rcdata.c state_script.c state.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/html/tree", "active_formatting.c open_elements.c error.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/html/tree/insertion_mode", "after_after_body.c after_after_frameset.c after_body.c after_frameset.c after_head.c before_head.c before_html.c foreign_content.c in_body.c in_caption.c in_cell.c in_column_group.c in_frameset.c in_head.c in_head_noscript.c initial.c in_row.c in_select.c in_select_in_table.c in_table_body.c in_table.c in_table_text.c in_template.c text.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/html", "encoding.c interface.c parser.c token.c token_attr.c tokenizer.c tree.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/encoding", "big5.c decode.c encode.c encoding.c euc_kr.c gb18030.c iso_2022_jp_katakana.c jis0208.c jis0212.c range.c res.c single.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/html/interfaces", "anchor_element.c area_element.c audio_element.c base_element.c body_element.c br_element.c button_element.c canvas_element.c data_element.c data_list_element.c details_element.c dialog_element.c directory_element.c div_element.c d_list_element.c document.c element.c embed_element.c field_set_element.c font_element.c form_element.c frame_element.c frame_set_element.c head_element.c heading_element.c hr_element.c html_element.c iframe_element.c image_element.c input_element.c label_element.c legend_element.c li_element.c link_element.c map_element.c marquee_element.c media_element.c menu_element.c meta_element.c meter_element.c mod_element.c object_element.c o_list_element.c opt_group_element.c option_element.c output_element.c paragraph_element.c param_element.c picture_element.c pre_element.c progress_element.c quote_element.c script_element.c select_element.c slot_element.c source_element.c span_element.c style_element.c table_caption_element.c table_cell_element.c table_col_element.c table_element.c table_row_element.c table_section_element.c template_element.c text_area_element.c time_element.c title_element.c track_element.c u_list_element.c unknown_element.c video_element.c window.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/selectors", "selectors.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/ns", "ns.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/tag", "tag.c", "dom");
+		ADD_FLAG("CFLAGS_DOM", "/D LEXBOR_STATIC ");
 
 		AC_DEFINE("HAVE_DOM", 1, "DOM support");
 

ext/dom/document.c

@@ -35,9 +35,6 @@ struct _idsIterator {
 	xmlNode *element;
 };
 
-#define DOM_LOAD_STRING 0
-#define DOM_LOAD_FILE 1
-
 /*
 * class DOMDocument extends DOMNode
 *
@@ -486,7 +483,7 @@ zend_result dom_document_config_read(dom_object *obj, zval *retval)
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#core-ID-2141741547
 Since:
 */
-PHP_METHOD(DOMDocument, createElement)
+PHP_METHOD(DOM_Document, createElement)
 {
 	zval *id;
 	xmlNode *node;
@@ -521,7 +518,7 @@ PHP_METHOD(DOMDocument, createElement)
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#core-ID-35CB04B5
 Since:
 */
-PHP_METHOD(DOMDocument, createDocumentFragment)
+PHP_METHOD(DOM_Document, createDocumentFragment)
 {
 	zval *id;
 	xmlNode *node;
@@ -549,7 +546,7 @@ PHP_METHOD(DOMDocument, createDocumentFragment)
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#core-ID-1975348127
 Since:
 */
-PHP_METHOD(DOMDocument, createTextNode)
+PHP_METHOD(DOM_Document, createTextNode)
 {
 	zval *id;
 	xmlNode *node;
@@ -579,7 +576,7 @@ PHP_METHOD(DOMDocument, createTextNode)
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#core-ID-1334481328
 Since:
 */
-PHP_METHOD(DOMDocument, createComment)
+PHP_METHOD(DOM_Document, createComment)
 {
 	zval *id;
 	xmlNode *node;
@@ -609,7 +606,7 @@ PHP_METHOD(DOMDocument, createComment)
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#core-ID-D26C0AF8
 Since:
 */
-PHP_METHOD(DOMDocument, createCDATASection)
+PHP_METHOD(DOM_Document, createCDATASection)
 {
 	zval *id;
 	xmlNode *node;
@@ -639,7 +636,7 @@ PHP_METHOD(DOMDocument, createCDATASection)
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#core-ID-135944439
 Since:
 */
-PHP_METHOD(DOMDocument, createProcessingInstruction)
+PHP_METHOD(DOM_Document, createProcessingInstruction)
 {
 	zval *id;
 	xmlNode *node;
@@ -676,7 +673,7 @@ PHP_METHOD(DOMDocument, createProcessingInstruction)
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#core-ID-1084891198
 Since:
 */
-PHP_METHOD(DOMDocument, createAttribute)
+PHP_METHOD(DOM_Document, createAttribute)
 {
 	zval *id;
 	xmlAttrPtr node;
@@ -747,7 +744,7 @@ PHP_METHOD(DOMDocument, createEntityReference)
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#core-ID-A6C9094
 Since:
 */
-PHP_METHOD(DOMDocument, getElementsByTagName)
+PHP_METHOD(DOM_Document, getElementsByTagName)
 {
 	size_t name_len;
 	dom_object *intern, *namednode;
@@ -768,7 +765,7 @@ PHP_METHOD(DOMDocument, getElementsByTagName)
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#Core-Document-importNode
 Since: DOM Level 2
 */
-PHP_METHOD(DOMDocument, importNode)
+PHP_METHOD(DOM_Document, importNode)
 {
 	zval *node;
 	xmlDocPtr docp;
@@ -776,8 +773,6 @@ PHP_METHOD(DOMDocument, importNode)
 	dom_object *intern, *nodeobj;
 	int ret;
 	bool recursive = 0;
-	/* See http://www.xmlsoft.org/html/libxml-tree.html#xmlDocCopyNode for meaning of values */
-	int extended_recursive;
 
 	if (zend_parse_parameters(ZEND_NUM_ARGS(), "O|b", &node, dom_node_class_entry, &recursive) == FAILURE) {
 		RETURN_THROWS();
@@ -796,11 +791,7 @@ PHP_METHOD(DOMDocument, importNode)
 	if (nodep->doc == docp) {
 		retnodep = nodep;
 	} else {
-		extended_recursive = recursive;
+		retnodep = dom_clone_node(nodep, docp, intern, recursive);
-		if ((recursive == 0) && (nodep->type == XML_ELEMENT_NODE)) {
-			extended_recursive = 2;
-		}
-		retnodep = xmlDocCopyNode(nodep, docp, extended_recursive);
 		if (!retnodep) {
 			RETURN_FALSE;
 		}
@@ -827,7 +818,7 @@ PHP_METHOD(DOMDocument, importNode)
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#core-ID-DocCrElNS
 Since: DOM Level 2
 */
-PHP_METHOD(DOMDocument, createElementNS)
+PHP_METHOD(DOM_Document, createElementNS)
 {
 	zval *id;
 	xmlDocPtr docp;
@@ -887,7 +878,7 @@ PHP_METHOD(DOMDocument, createElementNS)
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#core-ID-DocCrAttrNS
 Since: DOM Level 2
 */
-PHP_METHOD(DOMDocument, createAttributeNS)
+PHP_METHOD(DOM_Document, createAttributeNS)
 {
 	zval *id;
 	xmlDocPtr docp;
@@ -953,7 +944,7 @@ PHP_METHOD(DOMDocument, createAttributeNS)
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#core-ID-getElBTNNS
 Since: DOM Level 2
 */
-PHP_METHOD(DOMDocument, getElementsByTagNameNS)
+PHP_METHOD(DOM_Document, getElementsByTagNameNS)
 {
 	size_t uri_len, name_len;
 	dom_object *intern, *namednode;
@@ -974,7 +965,7 @@ PHP_METHOD(DOMDocument, getElementsByTagNameNS)
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#core-ID-getElBId
 Since: DOM Level 2
 */
-PHP_METHOD(DOMDocument, getElementById)
+PHP_METHOD(DOM_Document, getElementById)
 {
 	zval *id;
 	xmlDocPtr docp;
@@ -1051,7 +1042,7 @@ bool php_dom_adopt_node(xmlNodePtr nodep, dom_object *dom_object_new_document, x
 Since: DOM Level 3
 Modern spec URL: https://dom.spec.whatwg.org/#dom-document-adoptnode
 */
-PHP_METHOD(DOMDocument, adoptNode)
+PHP_METHOD(DOM_Document, adoptNode)
 {
 	zval *node_zval;
 	if (zend_parse_parameters(ZEND_NUM_ARGS(), "O", &node_zval, dom_node_class_entry) == FAILURE) {
@@ -1088,7 +1079,7 @@ PHP_METHOD(DOMDocument, adoptNode)
 /* {{{ URL: http://www.w3.org/TR/2003/WD-DOM-Level-3-Core-20030226/DOM3-Core.html#core-Document3-normalizeDocument
 Since: DOM Level 3
 */
-PHP_METHOD(DOMDocument, normalizeDocument)
+PHP_METHOD(DOM_Document, normalizeDocument)
 {
 	zval *id;
 	xmlDocPtr docp;
@@ -1149,11 +1140,11 @@ PHP_METHOD(DOMDocument, __construct)
 }
 /* }}} end DOMDocument::__construct */
 
-char *_dom_get_valid_file_path(char *source, char *resolved_path, int resolved_path_len ) /* {{{ */
+const char *_dom_get_valid_file_path(const char *source, char *resolved_path, int resolved_path_len ) /* {{{ */
 {
 	xmlURI *uri;
 	xmlChar *escsource;
-	char *file_dest;
+	const char *file_dest;
 	int isFileUri = 0;
 
 	uri = xmlCreateURI();
@@ -1206,7 +1197,7 @@ char *_dom_get_valid_file_path(char *source, char *resolved_path, int resolved_p
 }
 /* }}} */
 
-static xmlDocPtr dom_document_parser(zval *id, int mode, char *source, size_t source_len, size_t options) /* {{{ */
+xmlDocPtr dom_document_parser(zval *id, dom_load_mode mode, const char *source, size_t source_len, size_t options, xmlCharEncodingHandlerPtr encoding) /* {{{ */
 {
 	xmlDocPtr ret;
 	xmlParserCtxtPtr ctxt = NULL;
@@ -1215,10 +1206,14 @@ static xmlDocPtr dom_document_parser(zval *id, int mode, char *source, size_t so
 	int old_error_reporting = 0;
 	char *directory=NULL, resolved_path[MAXPATHLEN + 1];
 
+	libxml_doc_props const* doc_props;
+	if (id == NULL) {
+		doc_props = dom_get_doc_props_read_only(NULL);
+	} else {
 		dom_object *intern = Z_DOMOBJ_P(id);
 		php_libxml_ref_obj *document = intern->document;
-
+		doc_props = dom_get_doc_props_read_only(document);
-	libxml_doc_props const* doc_props = dom_get_doc_props_read_only(document);
+	}
 	validate = doc_props->validateonparse;
 	resolve_externals = doc_props->resolveexternals;
 	keep_blanks = doc_props->preservewhitespace;
@@ -1228,12 +1223,11 @@ static xmlDocPtr dom_document_parser(zval *id, int mode, char *source, size_t so
 	xmlInitParser();
 
 	if (mode == DOM_LOAD_FILE) {
-		char *file_dest;
 		if (CHECK_NULL_PATH(source, source_len)) {
-			zend_value_error("Path to document must not contain any null bytes");
+			zend_argument_value_error(1, "must not contain any null bytes");
 			return NULL;
 		}
-		file_dest = _dom_get_valid_file_path(source, resolved_path, MAXPATHLEN);
+		const char *file_dest = _dom_get_valid_file_path(source, resolved_path, MAXPATHLEN);
 		if (file_dest) {
 			ctxt = xmlCreateFileParserCtxt(file_dest);
 		}
@@ -1246,6 +1240,8 @@ static xmlDocPtr dom_document_parser(zval *id, int mode, char *source, size_t so
 		return(NULL);
 	}
 
+	(void) xmlSwitchToEncoding(ctxt, encoding);
+
 	/* If loading from memory, we need to set the base directory for the document */
 	if (mode != DOM_LOAD_FILE) {
 #ifdef HAVE_GETCWD
@@ -1319,7 +1315,7 @@ static xmlDocPtr dom_document_parser(zval *id, int mode, char *source, size_t so
 }
 /* }}} */
 
-static void dom_finish_loading_document(zval *this, zval *return_value, xmlDocPtr newdoc)
+static void php_dom_finish_loading_document(zval *this, zval *return_value, xmlDocPtr newdoc)
 {
 	if (!newdoc)
 		RETURN_FALSE;
@@ -1327,11 +1323,13 @@ static void dom_finish_loading_document(zval *this, zval *return_value, xmlDocPt
 	dom_object *intern = Z_DOMOBJ_P(this);
 	size_t old_modification_nr = 0;
 	if (intern != NULL) {
+		bool is_modern_api_class = false;
 		xmlDocPtr docp = (xmlDocPtr) dom_object_get_node(intern);
 		dom_doc_propsptr doc_prop = NULL;
 		if (docp != NULL) {
 			const php_libxml_ref_obj *doc_ptr = intern->document;
 			ZEND_ASSERT(doc_ptr != NULL); /* Must exist, we have a document */
+			is_modern_api_class = doc_ptr->is_modern_api_class;
 			old_modification_nr = doc_ptr->cache_tag.modification_nr;
 			php_libxml_decrement_node_ptr((php_libxml_node_object *) intern);
 			doc_prop = intern->document->doc_props;
@@ -1346,6 +1344,7 @@ static void dom_finish_loading_document(zval *this, zval *return_value, xmlDocPt
 			RETURN_FALSE;
 		}
 		intern->document->doc_props = doc_prop;
+		intern->document->is_modern_api_class = is_modern_api_class;
 	}
 
 	php_libxml_increment_node_ptr((php_libxml_node_object *)intern, (xmlNodePtr)newdoc, (void *)intern);
@@ -1358,8 +1357,8 @@ static void dom_finish_loading_document(zval *this, zval *return_value, xmlDocPt
 	RETURN_TRUE;
 }
 
-/* {{{ static void dom_parse_document(INTERNAL_FUNCTION_PARAMETERS, int mode) */
+static void dom_parse_document(INTERNAL_FUNCTION_PARAMETERS, int mode)
-static void dom_parse_document(INTERNAL_FUNCTION_PARAMETERS, int mode) {
+{
 	char *source;
 	size_t source_len;
 	zend_long options = 0;
@@ -1381,11 +1380,9 @@ static void dom_parse_document(INTERNAL_FUNCTION_PARAMETERS, int mode) {
 		RETURN_FALSE;
 	}
 
-	xmlDocPtr newdoc = dom_document_parser(ZEND_THIS, mode, source, source_len, options);
+	xmlDocPtr newdoc = dom_document_parser(ZEND_THIS, mode, source, source_len, options, NULL);
-
+	php_dom_finish_loading_document(ZEND_THIS, return_value, newdoc);
-	dom_finish_loading_document(ZEND_THIS, return_value, newdoc);
 }
-/* }}} end dom_parser_document */
 
 /* {{{ URL: http://www.w3.org/TR/DOM-Level-3-LS/load-save.html#LS-DocumentLS-load
 Since: DOM Level 3
@@ -1664,7 +1661,8 @@ static void _dom_document_schema_validate(INTERNAL_FUNCTION_PARAMETERS, int type
 	zval *id;
 	xmlDoc *docp;
 	dom_object *intern;
-	char *source = NULL, *valid_file = NULL;
+	char *source = NULL;
+	const char *valid_file = NULL;
 	size_t source_len = 0;
 	int valid_opts = 0;
 	zend_long flags = 0;
@@ -1756,14 +1754,14 @@ static void _dom_document_schema_validate(INTERNAL_FUNCTION_PARAMETERS, int type
 /* }}} */
 
 /* {{{ */
-PHP_METHOD(DOMDocument, schemaValidate)
+PHP_METHOD(DOM_Document, schemaValidate)
 {
 	_dom_document_schema_validate(INTERNAL_FUNCTION_PARAM_PASSTHRU, DOM_LOAD_FILE);
 }
 /* }}} end dom_document_schema_validate_file */
 
 /* {{{ */
-PHP_METHOD(DOMDocument, schemaValidateSource)
+PHP_METHOD(DOM_Document, schemaValidateSource)
 {
 	_dom_document_schema_validate(INTERNAL_FUNCTION_PARAM_PASSTHRU, DOM_LOAD_STRING);
 }
@@ -1774,7 +1772,8 @@ static void _dom_document_relaxNG_validate(INTERNAL_FUNCTION_PARAMETERS, int typ
 	zval *id;
 	xmlDoc *docp;
 	dom_object *intern;
-	char *source = NULL, *valid_file = NULL;
+	char *source = NULL;
+	const char *valid_file = NULL;
 	size_t source_len = 0;
 	xmlRelaxNGParserCtxtPtr parser;
 	xmlRelaxNGPtr           sptr;
@@ -1852,14 +1851,14 @@ static void _dom_document_relaxNG_validate(INTERNAL_FUNCTION_PARAMETERS, int typ
 /* }}} */
 
 /* {{{ */
-PHP_METHOD(DOMDocument, relaxNGValidate)
+PHP_METHOD(DOM_Document, relaxNGValidate)
 {
 	_dom_document_relaxNG_validate(INTERNAL_FUNCTION_PARAM_PASSTHRU, DOM_LOAD_FILE);
 }
 /* }}} end dom_document_relaxNG_validate_file */
 
 /* {{{ */
-PHP_METHOD(DOMDocument, relaxNGValidateSource)
+PHP_METHOD(DOM_Document, relaxNGValidateSource)
 {
 	_dom_document_relaxNG_validate(INTERNAL_FUNCTION_PARAM_PASSTHRU, DOM_LOAD_STRING);
 }
@@ -1923,7 +1922,7 @@ static void dom_load_html(INTERNAL_FUNCTION_PARAMETERS, int mode) /* {{{ */
 	xmlDocPtr newdoc = ctxt->myDoc;
 	htmlFreeParserCtxt(ctxt);
 
-	dom_finish_loading_document(ZEND_THIS, return_value, newdoc);
+	php_dom_finish_loading_document(ZEND_THIS, return_value, newdoc);
 }
 /* }}} */
 
@@ -2065,7 +2064,7 @@ PHP_METHOD(DOMDocument, saveHTML)
 #endif  /* defined(LIBXML_HTML_ENABLED) */
 
 /* {{{ Register extended class used to create base node type */
-PHP_METHOD(DOMDocument, registerNodeClass)
+PHP_METHOD(DOM_Document, registerNodeClass)
 {
 	zend_class_entry *basece = dom_node_class_entry, *ce = NULL;
 	dom_object *intern;
@@ -2074,6 +2073,11 @@ PHP_METHOD(DOMDocument, registerNodeClass)
 		RETURN_THROWS();
 	}
 
+	if (basece->ce_flags & ZEND_ACC_ABSTRACT) {
+		zend_argument_value_error(1, "must not be an abstract class");
+		RETURN_THROWS();
+	}
+
 	if (ce == NULL || instanceof_function(ce, basece)) {
 		if (UNEXPECTED(ce != NULL && (ce->ce_flags & ZEND_ACC_ABSTRACT))) {
 			zend_argument_value_error(2, "must not be an abstract class");
@@ -2091,7 +2095,7 @@ PHP_METHOD(DOMDocument, registerNodeClass)
 /* {{{ URL: https://dom.spec.whatwg.org/#dom-parentnode-replacechildren
 Since:
 */
-PHP_METHOD(DOMDocument, replaceChildren)
+PHP_METHOD(DOM_Document, replaceChildren)
 {
 	uint32_t argc = 0;
 	zval *args;

ext/dom/dom_ce.h

@@ -23,6 +23,8 @@ extern PHP_DOM_EXPORT zend_class_entry *dom_domexception_class_entry;
 extern PHP_DOM_EXPORT zend_class_entry *dom_domimplementation_class_entry;
 extern PHP_DOM_EXPORT zend_class_entry *dom_documentfragment_class_entry;
 extern PHP_DOM_EXPORT zend_class_entry *dom_document_class_entry;
+extern PHP_DOM_EXPORT zend_class_entry *dom_html_document_class_entry;
+extern PHP_DOM_EXPORT zend_class_entry *dom_xml_document_class_entry;
 extern PHP_DOM_EXPORT zend_class_entry *dom_nodelist_class_entry;
 extern PHP_DOM_EXPORT zend_class_entry *dom_namednodemap_class_entry;
 extern PHP_DOM_EXPORT zend_class_entry *dom_characterdata_class_entry;
@@ -36,6 +38,7 @@ extern PHP_DOM_EXPORT zend_class_entry *dom_notation_class_entry;
 extern PHP_DOM_EXPORT zend_class_entry *dom_entity_class_entry;
 extern PHP_DOM_EXPORT zend_class_entry *dom_entityreference_class_entry;
 extern PHP_DOM_EXPORT zend_class_entry *dom_processinginstruction_class_entry;
+extern PHP_DOM_EXPORT zend_class_entry *dom_abstract_base_document_class_entry;
 #ifdef LIBXML_XPATH_ENABLED
 extern PHP_DOM_EXPORT zend_class_entry *dom_xpath_class_entry;
 #endif

ext/dom/dom_properties.h

@@ -61,6 +61,9 @@ zend_result dom_document_recover_write(dom_object *obj, zval *newval);
 zend_result dom_document_substitue_entities_read(dom_object *obj, zval *retval);
 zend_result dom_document_substitue_entities_write(dom_object *obj, zval *newval);
 
+/* html5 document properties */
+zend_result dom_html_document_encoding_write(dom_object *obj, zval *retval);
+
 /* documenttype properties */
 zend_result dom_documenttype_name_read(dom_object *obj, zval *retval);
 zend_result dom_documenttype_entities_read(dom_object *obj, zval *retval);

ext/dom/html5_parser.c

@@ -0,0 +1,328 @@
+/*
+   +----------------------------------------------------------------------+
+   | Copyright (c) The PHP Group                                          |
+   +----------------------------------------------------------------------+
+   | This source file is subject to version 3.01 of the PHP license,      |
+   | that is bundled with this package in the file LICENSE, and is        |
+   | available through the world-wide-web at the following url:           |
+   | https://www.php.net/license/3_01.txt                                 |
+   | If you did not receive a copy of the PHP license and are unable to   |
+   | obtain it through the world-wide-web, please send a note to          |
+   | license@php.net so we can mail you a copy immediately.               |
+   +----------------------------------------------------------------------+
+   | Authors: Niels Dossche <nielsdos@php.net>                            |
+   +----------------------------------------------------------------------+
+*/
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php.h"
+#if defined(HAVE_LIBXML) && defined(HAVE_DOM)
+#include "html5_parser.h"
+#include "namespace_compat.h"
+#include <lexbor/html/parser.h>
+#include <lexbor/html/interfaces/element.h>
+#include <libxml/tree.h>
+#include <libxml/parserInternals.h>
+#include <libxml/HTMLtree.h>
+#include <Zend/zend.h>
+
+#define WORK_LIST_INIT_SIZE 128
+/* libxml2 reserves 2 pointer-sized words for interned strings */
+#define LXML_INTERNED_STRINGS_SIZE (sizeof(void *) * 2)
+
+typedef struct {
+    lxb_dom_node_t *node;
+    uintptr_t current_active_namespace;
+    xmlNodePtr lxml_parent;
+    xmlNsPtr lxml_ns;
+} work_list_item;
+
+static void lexbor_libxml2_bridge_work_list_item_push(
+    lexbor_array_obj_t *array,
+    lxb_dom_node_t *node,
+    uintptr_t current_active_namespace,
+    xmlNodePtr lxml_parent,
+    xmlNsPtr lxml_ns
+)
+{
+    work_list_item *item = (work_list_item *) lexbor_array_obj_push_wo_cls(array);
+    item->node = node;
+    item->current_active_namespace = current_active_namespace;
+    item->lxml_parent = lxml_parent;
+    item->lxml_ns = lxml_ns;
+}
+
+static unsigned short sanitize_line_nr(size_t line)
+{
+    if (line > USHRT_MAX) {
+        return USHRT_MAX;
+    }
+    return (unsigned short) line;
+}
+
+static const xmlChar *get_libxml_namespace_href(uintptr_t lexbor_namespace)
+{
+    if (lexbor_namespace == LXB_NS_SVG) {
+        return (const xmlChar *) DOM_SVG_NS_URI;
+    } else if (lexbor_namespace == LXB_NS_MATH) {
+        return (const xmlChar *) DOM_MATHML_NS_URI;
+    } else {
+        return (const xmlChar *) DOM_XHTML_NS_URI;
+    }
+}
+
+static lexbor_libxml2_bridge_status lexbor_libxml2_bridge_convert(
+    lxb_dom_node_t *start_node,
+    xmlDocPtr lxml_doc,
+    bool compact_text_nodes,
+    bool create_default_ns
+)
+{
+    lexbor_libxml2_bridge_status retval = LEXBOR_LIBXML2_BRIDGE_STATUS_OK;
+
+    lexbor_array_obj_t work_list;
+    lexbor_array_obj_init(&work_list, WORK_LIST_INIT_SIZE, sizeof(work_list_item));
+
+    for (lxb_dom_node_t *node = start_node; node != NULL; node = node->prev) {
+        lexbor_libxml2_bridge_work_list_item_push(&work_list, node, LXB_NS__UNDEF, (xmlNodePtr) lxml_doc, NULL);
+    }
+
+    work_list_item *current_stack_item;
+    while ((current_stack_item = lexbor_array_obj_pop(&work_list)) != NULL) {
+        lxb_dom_node_t *node = current_stack_item->node;
+        xmlNodePtr lxml_parent = current_stack_item->lxml_parent;
+
+        /* CDATA section and processing instructions don't occur in parsed HTML documents.
+         * The historical types are not emitted by the parser either. */
+        if (node->type == LXB_DOM_NODE_TYPE_ELEMENT) {
+            /* Note: HTML isn't exactly XML-namespace-aware; as this is an HTML parser we only care about the local name.
+             * If a prefix:name format is used, then the local name will be "prefix:name" and the prefix will be empty.
+             * There is however still somewhat of a concept of namespaces. There are three: HTML (the default), SVG, and MATHML. */
+            lxb_dom_element_t *element = lxb_dom_interface_element(node);
+            const lxb_char_t *name = lxb_dom_element_local_name(element, NULL);
+            xmlNodePtr lxml_element = xmlNewDocNode(lxml_doc, NULL, name, NULL);
+            if (UNEXPECTED(lxml_element == NULL)) {
+                retval = LEXBOR_LIBXML2_BRIDGE_STATUS_OOM;
+                goto out;
+            }
+            xmlAddChild(lxml_parent, lxml_element);
+            lxml_element->line = sanitize_line_nr(node->line);
+
+            /* Namespaces, note: namespace switches are uncommon */
+            uintptr_t entering_namespace = element->node.ns;
+            xmlNsPtr current_lxml_ns = current_stack_item->lxml_ns;
+            if (create_default_ns && UNEXPECTED(entering_namespace != current_stack_item->current_active_namespace)) {
+                current_lxml_ns = xmlNewNs(lxml_element, get_libxml_namespace_href(entering_namespace), NULL);
+            }
+            /* Instead of xmlSetNs() because we know the arguments are valid. Prevents overhead. */
+            lxml_element->ns = current_lxml_ns;
+
+            for (lxb_dom_node_t *child_node = element->node.last_child; child_node != NULL; child_node = child_node->prev) {
+                lexbor_libxml2_bridge_work_list_item_push(
+                    &work_list,
+                    child_node,
+                    entering_namespace,
+                    lxml_element,
+                    current_lxml_ns
+                );
+            }
+
+            for (lxb_dom_attr_t *attr = element->last_attr; attr != NULL; attr = attr->prev) {
+                lexbor_libxml2_bridge_work_list_item_push(
+                    &work_list,
+                    (lxb_dom_node_t *) attr,
+                    entering_namespace,
+                    lxml_element,
+                    current_lxml_ns
+                );
+            }
+        } else if (node->type == LXB_DOM_NODE_TYPE_TEXT) {
+            lxb_dom_text_t *text = lxb_dom_interface_text(node);
+            const lxb_char_t *data = text->char_data.data.data;
+            size_t data_length = text->char_data.data.length;
+            if (UNEXPECTED(data_length >= INT_MAX)) {
+                retval = LEXBOR_LIBXML2_BRIDGE_STATUS_OVERFLOW;
+                goto out;
+            }
+            xmlNodePtr lxml_text;
+            if (compact_text_nodes && data_length < LXML_INTERNED_STRINGS_SIZE) {
+                /* See xmlSAX2TextNode() in libxml2 */
+                lxml_text = xmlMalloc(sizeof(*lxml_text));
+                if (UNEXPECTED(lxml_text == NULL)) {
+                    retval = LEXBOR_LIBXML2_BRIDGE_STATUS_OOM;
+                    goto out;
+                }
+                memset(lxml_text, 0, sizeof(*lxml_text));
+                lxml_text->name = xmlStringText;
+                lxml_text->type = XML_TEXT_NODE;
+                lxml_text->doc = lxml_doc;
+                lxml_text->content = (xmlChar *) &lxml_text->properties;
+                memcpy(lxml_text->content, data, data_length + 1 /* include '\0' */);
+            } else {
+                lxml_text = xmlNewDocTextLen(lxml_doc, (const xmlChar *) data, data_length);
+                if (UNEXPECTED(lxml_text == NULL)) {
+                    retval = LEXBOR_LIBXML2_BRIDGE_STATUS_OOM;
+                    goto out;
+                }
+            }
+            xmlAddChild(lxml_parent, lxml_text);
+            if (node->line >= USHRT_MAX) {
+                lxml_text->line = USHRT_MAX;
+                lxml_text->psvi = (void *) (ptrdiff_t) node->line;
+            } else {
+                lxml_text->line = (unsigned short) node->line;
+            }
+        } else if (node->type == LXB_DOM_NODE_TYPE_DOCUMENT_TYPE) {
+            lxb_dom_document_type_t *doctype = lxb_dom_interface_document_type(node);
+            const lxb_char_t *name = lxb_dom_document_type_name(doctype, NULL);
+            size_t public_id_len, system_id_len;
+            const lxb_char_t *public_id = lxb_dom_document_type_public_id(doctype, &public_id_len);
+            const lxb_char_t *system_id = lxb_dom_document_type_system_id(doctype, &system_id_len);
+            xmlDtdPtr lxml_dtd = xmlCreateIntSubset(
+                lxml_doc,
+                name,
+                public_id_len ? public_id : NULL,
+                system_id_len ? system_id : NULL
+            );
+            if (UNEXPECTED(lxml_dtd == NULL)) {
+                retval = LEXBOR_LIBXML2_BRIDGE_STATUS_OOM;
+                goto out;
+            }
+            /* libxml2 doesn't support line numbers on this anyway, it returns -1 instead, so don't bother */
+        } else if (node->type == LXB_DOM_NODE_TYPE_ATTRIBUTE) {
+            lxb_dom_attr_t *attr = lxb_dom_interface_attr(node);
+            do {
+                /* Same namespace remark as for elements */
+                const lxb_char_t *local_name = lxb_dom_attr_local_name(attr, NULL);
+                const lxb_char_t *value = lxb_dom_attr_value(attr, NULL);
+                xmlAttrPtr lxml_attr = xmlSetNsProp(lxml_parent, NULL, local_name, value);
+                if (UNEXPECTED(lxml_attr == NULL)) {
+                    retval = LEXBOR_LIBXML2_BRIDGE_STATUS_OOM;
+                    goto out;
+                }
+                attr = attr->next;
+                /* libxml2 doesn't support line numbers on this anyway, it derives them instead, so don't bother */
+            } while (attr);
+        } else if (node->type == LXB_DOM_NODE_TYPE_COMMENT) {
+            lxb_dom_comment_t *comment = lxb_dom_interface_comment(node);
+            xmlNodePtr lxml_comment = xmlNewDocComment(lxml_doc, comment->char_data.data.data);
+            if (UNEXPECTED(lxml_comment == NULL)) {
+                retval = LEXBOR_LIBXML2_BRIDGE_STATUS_OOM;
+                goto out;
+            }
+            xmlAddChild(lxml_parent, lxml_comment);
+            lxml_comment->line = sanitize_line_nr(node->line);
+        }
+    }
+
+out:
+    lexbor_array_obj_destroy(&work_list, false);
+    return retval;
+}
+
+void lexbor_libxml2_bridge_parse_context_init(lexbor_libxml2_bridge_parse_context *ctx)
+{
+    memset(ctx, 0, sizeof(*ctx));
+}
+
+void lexbor_libxml2_bridge_parse_set_error_callbacks(
+    lexbor_libxml2_bridge_parse_context *ctx,
+    lexbor_libxml2_bridge_tokenizer_error_reporter tokenizer_error_reporter,
+    lexbor_libxml2_bridge_tree_error_reporter tree_error_reporter
+)
+{
+    ctx->tokenizer_error_reporter = tokenizer_error_reporter;
+    ctx->tree_error_reporter = tree_error_reporter;
+}
+
+lexbor_libxml2_bridge_status lexbor_libxml2_bridge_convert_document(
+    lxb_html_document_t *document,
+    xmlDocPtr *doc_out,
+    bool compact_text_nodes,
+    bool create_default_ns
+)
+{
+#ifdef LIBXML_HTML_ENABLED
+    xmlDocPtr lxml_doc = htmlNewDocNoDtD(NULL, NULL);
+#else
+    /* If HTML support is not enabled, then htmlNewDocNoDtD() is not available.
+     * This code mimics the behaviour. */
+    xmlDocPtr lxml_doc = xmlNewDoc((const xmlChar *) "1.0");
+    lxml_doc->type = XML_HTML_DOCUMENT_NODE;
+#endif
+    if (!lxml_doc) {
+        return LEXBOR_LIBXML2_BRIDGE_STATUS_OOM;
+    }
+    lexbor_libxml2_bridge_status status = lexbor_libxml2_bridge_convert(
+        lxb_dom_interface_node(document)->last_child,
+        lxml_doc,
+        compact_text_nodes,
+        create_default_ns
+    );
+    if (status != LEXBOR_LIBXML2_BRIDGE_STATUS_OK) {
+        xmlFreeDoc(lxml_doc);
+        return status;
+    }
+    *doc_out = lxml_doc;
+    return LEXBOR_LIBXML2_BRIDGE_STATUS_OK;
+}
+
+void lexbor_libxml2_bridge_report_errors(
+    const lexbor_libxml2_bridge_parse_context *ctx,
+    lxb_html_parser_t *parser,
+    const lxb_char_t *input_html,
+    size_t chunk_offset,
+    size_t *error_index_offset_tokenizer,
+    size_t *error_index_offset_tree
+)
+{
+    void *error;
+
+    /* Tokenizer errors */
+    lexbor_array_obj_t *parse_errors = lxb_html_parser_tokenizer(parser)->parse_errors;
+    size_t index = *error_index_offset_tokenizer;
+    while ((error = lexbor_array_obj_get(parse_errors, index)) != NULL) {
+        /* See https://github.com/lexbor/lexbor/blob/master/source/lexbor/html/tokenizer/error.h */
+        lxb_html_tokenizer_error_t *token_error = error;
+        if (ctx->tokenizer_error_reporter) {
+            ctx->tokenizer_error_reporter(
+                ctx->application_data,
+                token_error,
+                token_error->pos - input_html + chunk_offset
+            );
+        }
+        index++;
+    }
+    *error_index_offset_tokenizer = index;
+
+    /* Tree parser errors */
+    parse_errors = lxb_html_parser_tree(parser)->parse_errors;
+    index = *error_index_offset_tree;
+    while ((error = lexbor_array_obj_get(parse_errors, index)) != NULL) {
+        /* See https://github.com/lexbor/lexbor/blob/master/source/lexbor/html/tree/error.h */
+        lxb_html_tree_error_t *tree_error = error;
+        if (ctx->tree_error_reporter) {
+            ctx->tree_error_reporter(
+                ctx->application_data,
+                tree_error,
+                tree_error->line + 1,
+                tree_error->column + 1,
+                tree_error->length
+            );
+        }
+        index++;
+    }
+    *error_index_offset_tree = index;
+}
+
+void lexbor_libxml2_bridge_copy_observations(lxb_html_tree_t *tree, lexbor_libxml2_bridge_extracted_observations *observations)
+{
+    observations->has_explicit_html_tag = tree->has_explicit_html_tag;
+    observations->has_explicit_head_tag = tree->has_explicit_head_tag;
+    observations->has_explicit_body_tag = tree->has_explicit_body_tag;
+}
+
+#endif  /* HAVE_LIBXML && HAVE_DOM */

ext/dom/html5_parser.h

@@ -0,0 +1,86 @@
+/*
+   +----------------------------------------------------------------------+
+   | Copyright (c) The PHP Group                                          |
+   +----------------------------------------------------------------------+
+   | This source file is subject to version 3.01 of the PHP license,      |
+   | that is bundled with this package in the file LICENSE, and is        |
+   | available through the world-wide-web at the following url:           |
+   | https://www.php.net/license/3_01.txt                                 |
+   | If you did not receive a copy of the PHP license and are unable to   |
+   | obtain it through the world-wide-web, please send a note to          |
+   | license@php.net so we can mail you a copy immediately.               |
+   +----------------------------------------------------------------------+
+   | Authors: Niels Dossche <nielsdos@php.net>                            |
+   +----------------------------------------------------------------------+
+*/
+
+#ifndef HTML5_PARSER_H
+#define HTML5_PARSER_H
+
+#include <lexbor/html/parser.h>
+#include <libxml/tree.h>
+#include <Zend/zend_portability.h>
+
+typedef enum {
+    LEXBOR_LIBXML2_BRIDGE_STATUS_OK = 0,
+    LEXBOR_LIBXML2_BRIDGE_STATUS_CANNOT_INIT,
+    LEXBOR_LIBXML2_BRIDGE_STATUS_FATAL_PARSE,
+    LEXBOR_LIBXML2_BRIDGE_STATUS_OVERFLOW,
+    LEXBOR_LIBXML2_BRIDGE_STATUS_OOM,
+} lexbor_libxml2_bridge_status;
+
+typedef void (*lexbor_libxml2_bridge_tokenizer_error_reporter)(
+    void *application_data,
+    lxb_html_tokenizer_error_t *error,
+    size_t offset
+);
+typedef void (*lexbor_libxml2_bridge_tree_error_reporter)(
+    void *application_data,
+    lxb_html_tree_error_t *error,
+    size_t line,
+    size_t column,
+    size_t len
+);
+
+typedef struct {
+    bool has_explicit_html_tag;
+    bool has_explicit_head_tag;
+    bool has_explicit_body_tag;
+} lexbor_libxml2_bridge_extracted_observations;
+
+typedef struct {
+    /* Private fields */
+    lexbor_libxml2_bridge_tokenizer_error_reporter tokenizer_error_reporter;
+    lexbor_libxml2_bridge_tree_error_reporter tree_error_reporter;
+    /* Public fields */
+    lexbor_libxml2_bridge_extracted_observations observations;
+    /* Application data, do what you want with this */
+    void *application_data;
+} lexbor_libxml2_bridge_parse_context;
+
+void lexbor_libxml2_bridge_parse_context_init(lexbor_libxml2_bridge_parse_context *ctx);
+void lexbor_libxml2_bridge_parse_set_error_callbacks(
+    lexbor_libxml2_bridge_parse_context *ctx,
+    lexbor_libxml2_bridge_tokenizer_error_reporter tokenizer_error_reporter,
+    lexbor_libxml2_bridge_tree_error_reporter tree_error_reporter
+);
+lexbor_libxml2_bridge_status lexbor_libxml2_bridge_convert_document(
+    lxb_html_document_t *document,
+    xmlDocPtr *doc_out,
+    bool compact_text_nodes,
+    bool create_default_ns
+);
+void lexbor_libxml2_bridge_report_errors(
+    const lexbor_libxml2_bridge_parse_context *ctx,
+    lxb_html_parser_t *parser,
+    const lxb_char_t *input_html,
+    size_t chunk_offset,
+    size_t *error_index_offset_tokenizer,
+    size_t *error_index_offset_tree
+);
+void lexbor_libxml2_bridge_copy_observations(
+    lxb_html_tree_t *tree,
+    lexbor_libxml2_bridge_extracted_observations *observations
+);
+
+#endif

ext/dom/html5_serializer.c

@@ -0,0 +1,356 @@
+/*
+   +----------------------------------------------------------------------+
+   | Copyright (c) The PHP Group                                          |
+   +----------------------------------------------------------------------+
+   | This source file is subject to version 3.01 of the PHP license,      |
+   | that is bundled with this package in the file LICENSE, and is        |
+   | available through the world-wide-web at the following url:           |
+   | https://www.php.net/license/3_01.txt                                 |
+   | If you did not receive a copy of the PHP license and are unable to   |
+   | obtain it through the world-wide-web, please send a note to          |
+   | license@php.net so we can mail you a copy immediately.               |
+   +----------------------------------------------------------------------+
+   | Authors: Niels Dossche <nielsdos@php.net>                            |
+   +----------------------------------------------------------------------+
+*/
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php.h"
+#if defined(HAVE_LIBXML) && defined(HAVE_DOM)
+#include "php_dom.h"
+#include "html5_serializer.h"
+#include "namespace_compat.h"
+#include <lexbor/encoding/encoding.h>
+
+#define TRY(x) do { if (UNEXPECTED((x) != SUCCESS)) { return FAILURE; } } while (0)
+
+static bool dom_is_ns(const xmlNode *node, const char *uri)
+{
+	return node->ns != NULL && strcmp((const char *) node->ns->href, uri) == 0;
+}
+
+static bool dom_is_html_ns(const xmlNode *node)
+{
+	return node->ns == NULL || dom_is_ns(node, DOM_XHTML_NS_URI);
+}
+
+static bool dom_local_name_compare_ex(const xmlNode *node, const char *tag, size_t tag_length, size_t name_length)
+{
+	return name_length == tag_length && zend_binary_strcmp((const char *) node->name, name_length, tag, tag_length) == 0;
+}
+
+static zend_result dom_html5_serialize_doctype(dom_html5_serialize_context *ctx, const xmlDtd *dtd)
+{
+	TRY(ctx->write_string_len(ctx->application_data, "<!DOCTYPE ", strlen("<!DOCTYPE ")));
+	TRY(ctx->write_string(ctx->application_data, (const char *) dtd->name));
+	return ctx->write_string_len(ctx->application_data, ">", strlen(">"));
+}
+
+static zend_result dom_html5_serialize_comment(dom_html5_serialize_context *ctx, const xmlNode *node)
+{
+	TRY(ctx->write_string_len(ctx->application_data, "<!--", strlen("<!--")));
+	TRY(ctx->write_string(ctx->application_data, (const char *) node->content));
+	return ctx->write_string_len(ctx->application_data, "-->", strlen("-->"));
+}
+
+static zend_result dom_html5_serialize_processing_instruction(dom_html5_serialize_context *ctx, const xmlNode *node)
+{
+	TRY(ctx->write_string_len(ctx->application_data, "<?", strlen("<?")));
+	TRY(ctx->write_string(ctx->application_data, (const char *) node->name));
+	TRY(ctx->write_string_len(ctx->application_data, " ", strlen(" ")));
+	TRY(ctx->write_string(ctx->application_data, (const char *) node->content));
+	return ctx->write_string_len(ctx->application_data, ">", strlen(">"));
+}
+
+/* https://html.spec.whatwg.org/multipage/parsing.html#escapingString */
+static zend_result dom_html5_escape_string(dom_html5_serialize_context *ctx, const char *content, bool attribute_mode)
+{
+	const char *last_output = content;
+
+	while (*content != '\0') {
+		switch (*content) {
+			/* Step 1 */
+			case '&': {
+				TRY(ctx->write_string_len(ctx->application_data, last_output, content - last_output));
+				TRY(ctx->write_string_len(ctx->application_data, "&amp;", strlen("&amp;")));
+				last_output = content + 1;
+				break;
+			}
+
+			/* Step 2 (non-breaking space) (note: uses UTF-8 internally) */
+			case '\xC2': {
+				if (content[1] == '\xA0') {
+					TRY(ctx->write_string_len(ctx->application_data, last_output, content - last_output));
+					TRY(ctx->write_string_len(ctx->application_data, "&nbsp;", strlen("&nbsp;")));
+					content++; /* Consume A0 too */
+					last_output = content + 1;
+				}
+				break;
+			}
+
+			/* Step 3 */
+			case '"': {
+				if (attribute_mode) {
+					TRY(ctx->write_string_len(ctx->application_data, last_output, content - last_output));
+					TRY(ctx->write_string_len(ctx->application_data, "&quot;", strlen("&quot;")));
+					last_output = content + 1;
+				}
+				break;
+			}
+
+			/* Step 4 */
+			case '<': {
+				if (!attribute_mode) {
+					TRY(ctx->write_string_len(ctx->application_data, last_output, content - last_output));
+					TRY(ctx->write_string_len(ctx->application_data, "&lt;", strlen("&lt;")));
+					last_output = content + 1;
+				}
+				break;
+			}
+			case '>': {
+				if (!attribute_mode) {
+					TRY(ctx->write_string_len(ctx->application_data, last_output, content - last_output));
+					TRY(ctx->write_string_len(ctx->application_data, "&gt;", strlen("&gt;")));
+					last_output = content + 1;
+				}
+				break;
+			}
+		}
+
+		content++;
+	}
+
+	return ctx->write_string_len(ctx->application_data, last_output, content - last_output);
+}
+
+static zend_result dom_html5_serialize_text_node(dom_html5_serialize_context *ctx, const xmlNode *node)
+{
+	if (node->parent->type == XML_ELEMENT_NODE && dom_is_html_ns(node->parent)) {
+		const xmlNode *parent = node->parent;
+		size_t name_length = strlen((const char *) parent->name);
+		/* Spec tells us to only emit noscript content as-is if scripting is enabled.
+		 * However, the user agent (PHP) does not support (JS) scripting.
+		 * Furthermore, if actually consumed by a browser then we should err on the safe side and not emit the content as-is. */
+		if (dom_local_name_compare_ex(parent, "style", strlen("style"), name_length)
+			|| dom_local_name_compare_ex(parent, "script", strlen("script"), name_length)
+			|| dom_local_name_compare_ex(parent, "xmp", strlen("xmp"), name_length)
+			|| dom_local_name_compare_ex(parent, "iframe", strlen("iframe"), name_length)
+			|| dom_local_name_compare_ex(parent, "noembed", strlen("noembed"), name_length)
+			|| dom_local_name_compare_ex(parent, "noframes", strlen("noframes"), name_length)
+			|| dom_local_name_compare_ex(parent, "plaintext", strlen("plaintext"), name_length)) {
+			return ctx->write_string(ctx->application_data, (const char *) node->content);
+		}
+	}
+
+	return dom_html5_escape_string(ctx, (const char *) node->content, false);
+}
+
+static zend_result dom_html5_serialize_element_tag_name(dom_html5_serialize_context *ctx, const xmlNode *node)
+{
+	/* Note: it is not the serializer's responsibility to care about uppercase/lowercase (see createElement() note) */
+	if (node->ns != NULL && node->ns->prefix != NULL
+		&& !(dom_is_html_ns(node) || dom_is_ns(node, DOM_MATHML_NS_URI) || dom_is_ns(node, DOM_SVG_NS_URI))) {
+		TRY(ctx->write_string(ctx->application_data, (const char *) node->ns->prefix));
+		TRY(ctx->write_string_len(ctx->application_data, ":", strlen(":")));
+	}
+	return ctx->write_string(ctx->application_data, (const char *) node->name);
+}
+
+static zend_result dom_html5_serialize_element_start(dom_html5_serialize_context *ctx, const xmlNode *node)
+{
+	TRY(ctx->write_string_len(ctx->application_data, "<", strlen("<")));
+	TRY(dom_html5_serialize_element_tag_name(ctx, node));
+
+	/* We don't support the "is" value during element creation, so no handling here. */
+
+	/* Some namespace declarations are also attributes (see https://html.spec.whatwg.org/multipage/parsing.html#create-an-element-for-the-token) */
+	for (const xmlNs *ns = node->nsDef; ns != NULL; ns = ns->next) {
+		if (!dom_ns_is_also_an_attribute(ns)) {
+			continue;
+		}
+
+		if (ns->prefix != NULL) {
+			TRY(ctx->write_string_len(ctx->application_data, " xmlns:", strlen(" xmlns:")));
+			TRY(ctx->write_string(ctx->application_data, (const char *) ns->prefix));
+			TRY(ctx->write_string_len(ctx->application_data, "=\"", strlen("=\"")));
+		} else {
+			TRY(ctx->write_string_len(ctx->application_data, " xmlns=\"", strlen(" xmlns=\"")));
+		}
+		TRY(ctx->write_string(ctx->application_data, (const char *) ns->href));
+		TRY(ctx->write_string_len(ctx->application_data, "\"", strlen("\"")));
+	}
+
+	for (const xmlAttr *attr = node->properties; attr; attr = attr->next) {
+		TRY(ctx->write_string_len(ctx->application_data, " ", strlen(" ")));
+		if (attr->ns == NULL) {
+			TRY(ctx->write_string(ctx->application_data, (const char *) attr->name));
+		} else {
+			if (dom_is_ns((const xmlNode *) attr, DOM_XML_NS_URI)) {
+				TRY(ctx->write_string_len(ctx->application_data, "xml:", strlen("xml:")));
+				TRY(ctx->write_string(ctx->application_data, (const char *) attr->name));
+			} else if (dom_is_ns((const xmlNode *) attr, DOM_XMLNS_NS_URI)) {
+				/* Compatibility for real attributes */
+				if (strcmp((const char *) attr->name, "xmlns") == 0) {
+					TRY(ctx->write_string_len(ctx->application_data, "xmlns", strlen("xmlns")));
+				} else {
+					TRY(ctx->write_string_len(ctx->application_data, "xmlns:", strlen("xmlns:")));
+					TRY(ctx->write_string(ctx->application_data, (const char *) attr->name));
+				}
+			} else if (dom_is_ns((const xmlNode *) attr, DOM_XLINK_NS_URI)) {
+				TRY(ctx->write_string_len(ctx->application_data, "xlink:", strlen("xlink:")));
+				TRY(ctx->write_string(ctx->application_data, (const char *) attr->name));
+			} else if (attr->ns->prefix == NULL) {
+				TRY(ctx->write_string(ctx->application_data, (const char *) attr->name));
+			} else {
+				TRY(ctx->write_string(ctx->application_data, (const char *) attr->ns->prefix));
+				TRY(ctx->write_string_len(ctx->application_data, ":", strlen(":")));
+				TRY(ctx->write_string(ctx->application_data, (const char *) attr->name));
+			}
+		}
+		TRY(ctx->write_string_len(ctx->application_data, "=\"", strlen("=\"")));
+		xmlChar *content = xmlNodeGetContent((const xmlNode *) attr);
+		if (content != NULL) {
+			zend_result result = dom_html5_escape_string(ctx, (const char *) content, true);
+			xmlFree(content);
+			TRY(result);
+		}
+		TRY(ctx->write_string_len(ctx->application_data, "\"", strlen("\"")));
+	}
+
+	return ctx->write_string_len(ctx->application_data, ">", strlen(">"));
+
+	/* Note: "continue on to the next child if the element is void" is handled in the iteration and dom_html5_serialize_element_end() */
+}
+
+/* https://html.spec.whatwg.org/multipage/syntax.html#void-elements
+ * https://html.spec.whatwg.org/multipage/parsing.html#serializes-as-void */
+static bool dom_html5_serializes_as_void(const xmlNode *node)
+{
+	if (dom_is_html_ns(node)) {
+		size_t name_length = strlen((const char *) node->name);
+		if (/* These are the void elements from https://html.spec.whatwg.org/multipage/syntax.html#void-elements */
+			dom_local_name_compare_ex(node, "area", strlen("area"), name_length)
+			|| dom_local_name_compare_ex(node, "base", strlen("base"), name_length)
+			|| dom_local_name_compare_ex(node, "br", strlen("br"), name_length)
+			|| dom_local_name_compare_ex(node, "col", strlen("col"), name_length)
+			|| dom_local_name_compare_ex(node, "embed", strlen("embed"), name_length)
+			|| dom_local_name_compare_ex(node, "hr", strlen("hr"), name_length)
+			|| dom_local_name_compare_ex(node, "img", strlen("img"), name_length)
+			|| dom_local_name_compare_ex(node, "input", strlen("input"), name_length)
+			|| dom_local_name_compare_ex(node, "link", strlen("link"), name_length)
+			|| dom_local_name_compare_ex(node, "meta", strlen("meta"), name_length)
+			|| dom_local_name_compare_ex(node, "source", strlen("source"), name_length)
+			|| dom_local_name_compare_ex(node, "track", strlen("track"), name_length)
+			|| dom_local_name_compare_ex(node, "wbr", strlen("wbr"), name_length)
+			/* These are the additional names from https://html.spec.whatwg.org/multipage/parsing.html#serializes-as-void */
+			|| dom_local_name_compare_ex(node, "basefont", strlen("basefont"), name_length)
+			|| dom_local_name_compare_ex(node, "bgsound", strlen("bgsound"), name_length)
+			|| dom_local_name_compare_ex(node, "frame", strlen("frame"), name_length)
+			|| dom_local_name_compare_ex(node, "keygen", strlen("keygen"), name_length)
+			|| dom_local_name_compare_ex(node, "param", strlen("param"), name_length)) {
+			return true;
+		}
+	}
+	return false;
+}
+
+static zend_result dom_html5_serialize_element_end(dom_html5_serialize_context *ctx, const xmlNode *node)
+{
+	if (!dom_html5_serializes_as_void(node)) {
+		TRY(ctx->write_string_len(ctx->application_data, "</", strlen("</")));
+		TRY(dom_html5_serialize_element_tag_name(ctx, node));
+		return ctx->write_string_len(ctx->application_data, ">", strlen(">"));
+	}
+	return SUCCESS;
+}
+
+/* https://html.spec.whatwg.org/multipage/parsing.html#html-fragment-serialisation-algorithm */
+static zend_result dom_html5_serialize_node(dom_html5_serialize_context *ctx, const xmlNode *node, const xmlNode *bound)
+{
+	while (node != NULL) {
+		switch (node->type) {
+			case XML_DTD_NODE: {
+				TRY(dom_html5_serialize_doctype(ctx, (const xmlDtd *) node));
+				break;
+			}
+
+			case XML_CDATA_SECTION_NODE:
+			case XML_TEXT_NODE: {
+				TRY(dom_html5_serialize_text_node(ctx, node));
+				break;
+			}
+
+			case XML_PI_NODE: {
+				TRY(dom_html5_serialize_processing_instruction(ctx, node));
+				break;
+			}
+
+			case XML_COMMENT_NODE: {
+				TRY(dom_html5_serialize_comment(ctx, node));
+				break;
+			}
+
+			case XML_ELEMENT_NODE: {
+				TRY(dom_html5_serialize_element_start(ctx, node));
+				if (node->children) {
+					if (!dom_html5_serializes_as_void(node)) {
+						node = node->children;
+						continue;
+					}
+				} else {
+					/* Not descended, so wouldn't put the closing tag as it's normally only done when going back upwards. */
+					TRY(dom_html5_serialize_element_end(ctx, node));
+				}
+				break;
+			}
+
+			default:
+				break;
+		}
+
+		if (node->next) {
+			node = node->next;
+		} else {
+			/* Go upwards, until we find a parent node with a next sibling, or until we hit the bound. */
+			do {
+				node = node->parent;
+				if (node == bound) {
+					return SUCCESS;
+				}
+				if (node->type == XML_ELEMENT_NODE) {
+					TRY(dom_html5_serialize_element_end(ctx, node));
+				}
+			} while (node->next == NULL);
+			node = node->next;
+		}
+	}
+
+	return SUCCESS;
+}
+
+/* https://html.spec.whatwg.org/multipage/parsing.html#serialising-html-fragments (Date 2023-10-18)
+ * Note: this serializes the _children_, excluding the node itself! */
+zend_result dom_html5_serialize(dom_html5_serialize_context *ctx, const xmlNode *node)
+{
+	/* Step 1. Note that this algorithm serializes children. Only elements, documents, and fragments can have children. */
+	if (node->type != XML_ELEMENT_NODE
+		&& node->type != XML_DOCUMENT_FRAG_NODE
+		&& node->type != XML_DOCUMENT_NODE
+		&& node->type != XML_HTML_DOCUMENT_NODE) {
+		return SUCCESS;
+	}
+	if (node->type == XML_ELEMENT_NODE && dom_html5_serializes_as_void(node)) {
+		return SUCCESS;
+	}
+
+	/* Step 2 not needed because we're not using a string to store the serialized data */
+	/* Step 3 not needed because we don't support template contents yet */
+
+	/* Step 4 */
+	return dom_html5_serialize_node(ctx, node->children, node);
+}
+
+#endif  /* HAVE_LIBXML && HAVE_DOM */

ext/dom/html5_serializer.h

@@ -0,0 +1,31 @@
+/*
+   +----------------------------------------------------------------------+
+   | Copyright (c) The PHP Group                                          |
+   +----------------------------------------------------------------------+
+   | This source file is subject to version 3.01 of the PHP license,      |
+   | that is bundled with this package in the file LICENSE, and is        |
+   | available through the world-wide-web at the following url:           |
+   | https://www.php.net/license/3_01.txt                                 |
+   | If you did not receive a copy of the PHP license and are unable to   |
+   | obtain it through the world-wide-web, please send a note to          |
+   | license@php.net so we can mail you a copy immediately.               |
+   +----------------------------------------------------------------------+
+   | Authors: Niels Dossche <nielsdos@php.net>                            |
+   +----------------------------------------------------------------------+
+*/
+
+#ifndef HTML5_SERIALIZER_H
+#define HTML5_SERIALIZER_H
+
+#include <Zend/zend_types.h>
+#include <libxml/tree.h>
+
+typedef struct {
+	zend_result (*write_string)(void *application_data, const char *buf);
+	zend_result (*write_string_len)(void *application_data, const char *buf, size_t len);
+	void *application_data;
+} dom_html5_serialize_context;
+
+zend_result dom_html5_serialize(dom_html5_serialize_context *ctx, const xmlNode *node);
+
+#endif

ext/dom/namespace_compat.c

@@ -0,0 +1,54 @@
+/*
+   +----------------------------------------------------------------------+
+   | Copyright (c) The PHP Group                                          |
+   +----------------------------------------------------------------------+
+   | This source file is subject to version 3.01 of the PHP license,      |
+   | that is bundled with this package in the file LICENSE, and is        |
+   | available through the world-wide-web at the following url:           |
+   | https://www.php.net/license/3_01.txt                                 |
+   | If you did not receive a copy of the PHP license and are unable to   |
+   | obtain it through the world-wide-web, please send a note to          |
+   | license@php.net so we can mail you a copy immediately.               |
+   +----------------------------------------------------------------------+
+   | Authors: Niels Dossche <nielsdos@php.net>                            |
+   +----------------------------------------------------------------------+
+*/
+
+#ifdef HAVE_CONFIG_H
+#include "config.h"
+#endif
+
+#include "php.h"
+#if defined(HAVE_LIBXML) && defined(HAVE_DOM)
+#include "php_dom.h"
+#include "namespace_compat.h"
+
+bool dom_ns_is_also_an_attribute(const xmlNs *ns) {
+    return ns->_private != NULL;
+}
+
+void dom_ns_compat_mark_attribute(xmlNsPtr ns) {
+    ns->_private = (void *) 1;
+}
+
+void dom_ns_compat_mark_attribute_list(xmlNsPtr ns) {
+    while (ns != NULL) {
+        dom_ns_compat_mark_attribute(ns);
+        ns = ns->next;
+    }
+}
+
+void dom_ns_compat_copy_attribute_list_mark(xmlNsPtr copy, const xmlNs *original) {
+    /* It's possible that the original list is shorter than the copy list
+     * because of additional namespace copies from within a fragment. */
+    while (original != NULL) {
+        ZEND_ASSERT(copy != NULL);
+        if (dom_ns_is_also_an_attribute(original)) {
+            dom_ns_compat_mark_attribute(copy);
+        }
+        copy = copy->next;
+        original = original->next;
+    }
+}
+
+#endif  /* HAVE_LIBXML && HAVE_DOM */

ext/dom/namespace_compat.h

@@ -0,0 +1,39 @@
+/*
+   +----------------------------------------------------------------------+
+   | Copyright (c) The PHP Group                                          |
+   +----------------------------------------------------------------------+
+   | This source file is subject to version 3.01 of the PHP license,      |
+   | that is bundled with this package in the file LICENSE, and is        |
+   | available through the world-wide-web at the following url:           |
+   | https://www.php.net/license/3_01.txt                                 |
+   | If you did not receive a copy of the PHP license and are unable to   |
+   | obtain it through the world-wide-web, please send a note to          |
+   | license@php.net so we can mail you a copy immediately.               |
+   +----------------------------------------------------------------------+
+   | Authors: Niels Dossche <nielsdos@php.net>                            |
+   +----------------------------------------------------------------------+
+*/
+
+#ifndef NAMESPACE_COMPAT_H
+#define NAMESPACE_COMPAT_H
+
+#include <libxml/tree.h>
+
+/* https://infra.spec.whatwg.org/#namespaces */
+#define DOM_XHTML_NS_URI "http://www.w3.org/1999/xhtml"
+#define DOM_MATHML_NS_URI "http://www.w3.org/1998/Math/MathML"
+#define DOM_SVG_NS_URI "http://www.w3.org/2000/svg"
+#define DOM_XLINK_NS_URI "http://www.w3.org/1999/xlink"
+#define DOM_XML_NS_URI "http://www.w3.org/XML/1998/namespace"
+#define DOM_XMLNS_NS_URI "http://www.w3.org/2000/xmlns/"
+
+/* These functions make it possible to make a namespace declaration also visible as an attribute by
+ * setting a flag that can be checked with dom_ns_is_also_an_attribute().
+ * This is used in the serializer for example. */
+
+bool dom_ns_is_also_an_attribute(const xmlNs *ns);
+void dom_ns_compat_mark_attribute(xmlNsPtr ns);
+void dom_ns_compat_mark_attribute_list(xmlNsPtr ns);
+void dom_ns_compat_copy_attribute_list_mark(xmlNsPtr copy, const xmlNs *original);
+
+#endif

ext/dom/node.c

@@ -1329,40 +1329,12 @@ PHP_METHOD(DOMNode, cloneNode)
 
 	DOM_GET_OBJ(n, id, xmlNodePtr, intern);
 
-	node = xmlDocCopyNode(n, n->doc, recursive);
+	node = dom_clone_node(n, n->doc, intern, recursive);
 
 	if (!node) {
 		RETURN_FALSE;
 	}
 
-	/* When deep is false Element nodes still require the attributes
-	Following taken from libxml as xmlDocCopyNode doesn't do this */
-	if (n->type == XML_ELEMENT_NODE && recursive == 0) {
-		if (n->nsDef != NULL) {
-			node->nsDef = xmlCopyNamespaceList(n->nsDef);
-		}
-		if (n->ns != NULL) {
-			xmlNsPtr ns;
-			ns = xmlSearchNs(n->doc, node, n->ns->prefix);
-			if (ns == NULL) {
-				ns = xmlSearchNs(n->doc, n, n->ns->prefix);
-				if (ns != NULL) {
-					xmlNodePtr root = node;
-
-					while (root->parent != NULL) {
-						root = root->parent;
-					}
-					node->ns = xmlNewNs(root, ns->href, ns->prefix);
-				}
-			} else {
-				node->ns = ns;
-			}
-		}
-		if (n->properties != NULL) {
-			node->properties = xmlCopyPropList(node, n->properties);
-		}
-	}
-
 	if (node->type == XML_ATTRIBUTE_NODE && n->ns != NULL && node->ns == NULL) {
 		/* Let reconciliation deal with this. The lifetime of the namespace poses no problem
 		 * because we're increasing the refcount of the document proxy at the return.

ext/dom/php_dom.c

@@ -27,6 +27,8 @@
 #include "php_dom_arginfo.h"
 #include "dom_properties.h"
 #include "zend_interfaces.h"
+#include "lexbor/lexbor/core/types.h"
+#include "lexbor/lexbor/core/lexbor.h"
 
 #include "ext/standard/info.h"
 #define PHP_XPATH 1
@@ -40,6 +42,8 @@ PHP_DOM_EXPORT zend_class_entry *dom_childnode_class_entry;
 PHP_DOM_EXPORT zend_class_entry *dom_domimplementation_class_entry;
 PHP_DOM_EXPORT zend_class_entry *dom_documentfragment_class_entry;
 PHP_DOM_EXPORT zend_class_entry *dom_document_class_entry;
+PHP_DOM_EXPORT zend_class_entry *dom_html_document_class_entry;
+PHP_DOM_EXPORT zend_class_entry *dom_xml_document_class_entry;
 PHP_DOM_EXPORT zend_class_entry *dom_nodelist_class_entry;
 PHP_DOM_EXPORT zend_class_entry *dom_namednodemap_class_entry;
 PHP_DOM_EXPORT zend_class_entry *dom_characterdata_class_entry;
@@ -53,6 +57,7 @@ PHP_DOM_EXPORT zend_class_entry *dom_notation_class_entry;
 PHP_DOM_EXPORT zend_class_entry *dom_entity_class_entry;
 PHP_DOM_EXPORT zend_class_entry *dom_entityreference_class_entry;
 PHP_DOM_EXPORT zend_class_entry *dom_processinginstruction_class_entry;
+PHP_DOM_EXPORT zend_class_entry *dom_abstract_base_document_class_entry;
 #ifdef LIBXML_XPATH_ENABLED
 PHP_DOM_EXPORT zend_class_entry *dom_xpath_class_entry;
 #endif
@@ -70,6 +75,8 @@ zend_object_handlers dom_xpath_object_handlers;
 static HashTable classes;
 /* {{{ prop handler tables */
 static HashTable dom_document_prop_handlers;
+static HashTable dom_xml_document_prop_handlers;
+static HashTable dom_html_document_prop_handlers;
 static HashTable dom_documentfragment_prop_handlers;
 static HashTable dom_node_prop_handlers;
 static HashTable dom_nodelist_prop_handlers;
@@ -206,6 +213,7 @@ static void dom_copy_doc_props(php_libxml_ref_obj *source_doc, php_libxml_ref_ob
 			zend_hash_copy(dest->classmap, source->classmap, NULL);
 		}
 
+		dest_doc->is_modern_api_class = source_doc->is_modern_api_class;
 	}
 }
 
@@ -490,6 +498,12 @@ static void dom_update_refcount_after_clone(dom_object *original, xmlNodePtr ori
 	php_libxml_increment_node_ptr((php_libxml_node_object *)clone, cloned_node, (void *)clone);
 	if (original->document != clone->document) {
 		dom_copy_doc_props(original->document, clone->document);
+		/* Workaround libxml2 bug, see https://gitlab.gnome.org/GNOME/libxml2/-/commit/07920b4381873187c02df53fa9b5d44aff3a7041 */
+#if LIBXML_VERSION < 20911
+		if (original_node->type == XML_HTML_DOCUMENT_NODE) {
+			cloned_node->type = XML_HTML_DOCUMENT_NODE;
+		}
+#endif
 	}
 }
 
@@ -586,6 +600,38 @@ static zend_object *dom_objects_store_clone_obj(zend_object *zobject);
 void dom_xpath_objects_free_storage(zend_object *object);
 #endif
 
+static void *dom_malloc(size_t size) {
+	return emalloc(size);
+}
+
+static void *dom_realloc(void *dst, size_t size) {
+	return erealloc(dst, size);
+}
+
+static void *dom_calloc(size_t num, size_t size) {
+	return ecalloc(num, size);
+}
+
+static void dom_free(void *ptr) {
+	efree(ptr);
+}
+
+static void register_nondeprecated_xml_props(HashTable *table)
+{
+	dom_register_prop_handler(table, "encoding", sizeof("encoding")-1, dom_document_encoding_read, dom_document_encoding_write);
+	dom_register_prop_handler(table, "xmlEncoding", sizeof("xmlEncoding")-1, dom_document_encoding_read, NULL);
+	dom_register_prop_handler(table, "standalone", sizeof("standalone")-1, dom_document_standalone_read, dom_document_standalone_write);
+	dom_register_prop_handler(table, "xmlStandalone", sizeof("xmlStandalone")-1, dom_document_standalone_read, dom_document_standalone_write);
+	dom_register_prop_handler(table, "version", sizeof("version")-1, dom_document_version_read, dom_document_version_write);
+	dom_register_prop_handler(table, "xmlVersion", sizeof("xmlVersion")-1, dom_document_version_read, dom_document_version_write);
+	dom_register_prop_handler(table, "formatOutput", sizeof("formatOutput")-1, dom_document_format_output_read, dom_document_format_output_write);
+	dom_register_prop_handler(table, "validateOnParse", sizeof("validateOnParse")-1, dom_document_validate_on_parse_read, dom_document_validate_on_parse_write);
+	dom_register_prop_handler(table, "resolveExternals", sizeof("resolveExternals")-1, dom_document_resolve_externals_read, dom_document_resolve_externals_write);
+	dom_register_prop_handler(table, "preserveWhiteSpace", sizeof("preserveWhitespace")-1, dom_document_preserve_whitespace_read, dom_document_preserve_whitespace_write);
+	dom_register_prop_handler(table, "recover", sizeof("recover")-1, dom_document_recover_read, dom_document_recover_write);
+	dom_register_prop_handler(table, "substituteEntities", sizeof("substituteEntities")-1, dom_document_substitue_entities_read, dom_document_substitue_entities_write);
+}
+
 /* {{{ PHP_MINIT_FUNCTION(dom) */
 PHP_MINIT_FUNCTION(dom)
 {
@@ -675,36 +721,49 @@ PHP_MINIT_FUNCTION(dom)
 	zend_hash_merge(&dom_documentfragment_prop_handlers, &dom_node_prop_handlers, dom_copy_prop_handler, 0);
 	zend_hash_add_ptr(&classes, dom_documentfragment_class_entry->name, &dom_documentfragment_prop_handlers);
 
-	dom_document_class_entry = register_class_DOMDocument(dom_node_class_entry, dom_parentnode_class_entry);
+	dom_abstract_base_document_class_entry = register_class_DOM_Document(dom_node_class_entry, dom_parentnode_class_entry);
+	/* No need to set create_object as it's abstract. */
+	HashTable dom_abstract_base_document_prop_handlers;
+	zend_hash_init(&dom_abstract_base_document_prop_handlers, 0, NULL, dom_dtor_prop_handler, 1);
+	dom_register_prop_handler(&dom_abstract_base_document_prop_handlers, "doctype", sizeof("doctype")-1, dom_document_doctype_read, NULL);
+	dom_register_prop_handler(&dom_abstract_base_document_prop_handlers, "documentElement", sizeof("documentElement")-1, dom_document_document_element_read, NULL);
+	dom_register_prop_handler(&dom_abstract_base_document_prop_handlers, "strictErrorChecking", sizeof("strictErrorChecking")-1, dom_document_strict_error_checking_read, dom_document_strict_error_checking_write);
+	dom_register_prop_handler(&dom_abstract_base_document_prop_handlers, "documentURI", sizeof("documentURI")-1, dom_document_document_uri_read, dom_document_document_uri_write);
+	dom_register_prop_handler(&dom_abstract_base_document_prop_handlers, "firstElementChild", sizeof("firstElementChild")-1, dom_parent_node_first_element_child_read, NULL);
+	dom_register_prop_handler(&dom_abstract_base_document_prop_handlers, "lastElementChild", sizeof("lastElementChild")-1, dom_parent_node_last_element_child_read, NULL);
+	dom_register_prop_handler(&dom_abstract_base_document_prop_handlers, "childElementCount", sizeof("childElementCount")-1, dom_parent_node_child_element_count, NULL);
+	zend_hash_merge(&dom_abstract_base_document_prop_handlers, &dom_node_prop_handlers, dom_copy_prop_handler, 0);
+	/* No need to register in &classes, because this is only used for merging. This is destroyed down below. */
+
+	dom_document_class_entry = register_class_DOMDocument(dom_abstract_base_document_class_entry);
 	dom_document_class_entry->create_object = dom_objects_new;
 	zend_hash_init(&dom_document_prop_handlers, 0, NULL, dom_dtor_prop_handler, 1);
-	dom_register_prop_handler(&dom_document_prop_handlers, "doctype", sizeof("doctype")-1, dom_document_doctype_read, NULL);
 	dom_register_prop_handler(&dom_document_prop_handlers, "implementation", sizeof("implementation")-1, dom_document_implementation_read, NULL);
-	dom_register_prop_handler(&dom_document_prop_handlers, "documentElement", sizeof("documentElement")-1, dom_document_document_element_read, NULL);
 	dom_register_prop_handler(&dom_document_prop_handlers, "actualEncoding", sizeof("actualEncoding")-1, dom_document_encoding_read, NULL);
-	dom_register_prop_handler(&dom_document_prop_handlers, "encoding", sizeof("encoding")-1, dom_document_encoding_read, dom_document_encoding_write);
-	dom_register_prop_handler(&dom_document_prop_handlers, "xmlEncoding", sizeof("xmlEncoding")-1, dom_document_encoding_read, NULL);
-	dom_register_prop_handler(&dom_document_prop_handlers, "standalone", sizeof("standalone")-1, dom_document_standalone_read, dom_document_standalone_write);
-	dom_register_prop_handler(&dom_document_prop_handlers, "xmlStandalone", sizeof("xmlStandalone")-1, dom_document_standalone_read, dom_document_standalone_write);
-	dom_register_prop_handler(&dom_document_prop_handlers, "version", sizeof("version")-1, dom_document_version_read, dom_document_version_write);
-	dom_register_prop_handler(&dom_document_prop_handlers, "xmlVersion", sizeof("xmlVersion")-1, dom_document_version_read, dom_document_version_write);
-	dom_register_prop_handler(&dom_document_prop_handlers, "strictErrorChecking", sizeof("strictErrorChecking")-1, dom_document_strict_error_checking_read, dom_document_strict_error_checking_write);
-	dom_register_prop_handler(&dom_document_prop_handlers, "documentURI", sizeof("documentURI")-1, dom_document_document_uri_read, dom_document_document_uri_write);
 	dom_register_prop_handler(&dom_document_prop_handlers, "config", sizeof("config")-1, dom_document_config_read, NULL);
-	dom_register_prop_handler(&dom_document_prop_handlers, "formatOutput", sizeof("formatOutput")-1, dom_document_format_output_read, dom_document_format_output_write);
+	register_nondeprecated_xml_props(&dom_document_prop_handlers);
-	dom_register_prop_handler(&dom_document_prop_handlers, "validateOnParse", sizeof("validateOnParse")-1, dom_document_validate_on_parse_read, dom_document_validate_on_parse_write);
-	dom_register_prop_handler(&dom_document_prop_handlers, "resolveExternals", sizeof("resolveExternals")-1, dom_document_resolve_externals_read, dom_document_resolve_externals_write);
-	dom_register_prop_handler(&dom_document_prop_handlers, "preserveWhiteSpace", sizeof("preserveWhitespace")-1, dom_document_preserve_whitespace_read, dom_document_preserve_whitespace_write);
-	dom_register_prop_handler(&dom_document_prop_handlers, "recover", sizeof("recover")-1, dom_document_recover_read, dom_document_recover_write);
-	dom_register_prop_handler(&dom_document_prop_handlers, "substituteEntities", sizeof("substituteEntities")-1, dom_document_substitue_entities_read, dom_document_substitue_entities_write);
 
-	dom_register_prop_handler(&dom_document_prop_handlers, "firstElementChild", sizeof("firstElementChild")-1, dom_parent_node_first_element_child_read, NULL);
+	zend_hash_merge(&dom_document_prop_handlers, &dom_abstract_base_document_prop_handlers, dom_copy_prop_handler, 0);
-	dom_register_prop_handler(&dom_document_prop_handlers, "lastElementChild", sizeof("lastElementChild")-1, dom_parent_node_last_element_child_read, NULL);
-	dom_register_prop_handler(&dom_document_prop_handlers, "childElementCount", sizeof("childElementCount")-1, dom_parent_node_child_element_count, NULL);
-
-	zend_hash_merge(&dom_document_prop_handlers, &dom_node_prop_handlers, dom_copy_prop_handler, 0);
 	zend_hash_add_ptr(&classes, dom_document_class_entry->name, &dom_document_prop_handlers);
 
+	dom_html_document_class_entry = register_class_DOM_HTMLDocument(dom_abstract_base_document_class_entry);
+	dom_document_class_entry->create_object = dom_objects_new;
+	zend_hash_init(&dom_html_document_prop_handlers, 0, NULL, dom_dtor_prop_handler, 1);
+	dom_register_prop_handler(&dom_html_document_prop_handlers, "encoding", sizeof("encoding")-1, dom_document_encoding_read, dom_html_document_encoding_write);
+
+	zend_hash_merge(&dom_html_document_prop_handlers, &dom_abstract_base_document_prop_handlers, dom_copy_prop_handler, 0);
+	zend_hash_add_ptr(&classes, dom_html_document_class_entry->name, &dom_html_document_prop_handlers);
+
+	dom_xml_document_class_entry = register_class_DOM_XMLDocument(dom_abstract_base_document_class_entry);
+	dom_xml_document_class_entry->create_object = dom_objects_new;
+	zend_hash_init(&dom_xml_document_prop_handlers, 0, NULL, dom_dtor_prop_handler, 1);
+	register_nondeprecated_xml_props(&dom_xml_document_prop_handlers);
+
+	zend_hash_merge(&dom_xml_document_prop_handlers, &dom_abstract_base_document_prop_handlers, dom_copy_prop_handler, 0);
+	zend_hash_add_ptr(&classes, dom_xml_document_class_entry->name, &dom_xml_document_prop_handlers);
+
+	zend_hash_destroy(&dom_abstract_base_document_prop_handlers);
+
 	dom_nodelist_class_entry = register_class_DOMNodeList(zend_ce_aggregate, zend_ce_countable);
 	dom_nodelist_class_entry->create_object = dom_nnodemap_objects_new;
 	dom_nodelist_class_entry->default_object_handlers = &dom_nodelist_object_handlers;
@@ -845,6 +904,8 @@ PHP_MINIT_FUNCTION(dom)
 
 	php_libxml_register_export(dom_node_class_entry, php_dom_export_node);
 
+	lexbor_memory_setup(dom_malloc, dom_realloc, dom_calloc, dom_free);
+
 	return SUCCESS;
 }
 /* }}} */
@@ -876,6 +937,8 @@ PHP_MINFO_FUNCTION(dom)
 PHP_MSHUTDOWN_FUNCTION(dom) /* {{{ */
 {
 	zend_hash_destroy(&dom_document_prop_handlers);
+	zend_hash_destroy(&dom_html_document_prop_handlers);
+	zend_hash_destroy(&dom_xml_document_prop_handlers);
 	zend_hash_destroy(&dom_documentfragment_prop_handlers);
 	zend_hash_destroy(&dom_node_prop_handlers);
 	zend_hash_destroy(&dom_namespace_node_prop_handlers);
@@ -1172,9 +1235,21 @@ PHP_DOM_EXPORT bool php_dom_create_object(xmlNodePtr obj, zval *return_value, do
 
 	switch (obj->type) {
 		case XML_DOCUMENT_NODE:
+		{
+			if (domobj && domobj->document->is_modern_api_class) {
+				ce = dom_xml_document_class_entry;
+			} else {
+				ce = dom_document_class_entry;
+			}
+			break;
+		}
 		case XML_HTML_DOCUMENT_NODE:
 		{
+			if (domobj && domobj->document->is_modern_api_class) {
+				ce = dom_html_document_class_entry;
+			} else {
 				ce = dom_document_class_entry;
+			}
 			break;
 		}
 		case XML_DTD_NODE:
@@ -1249,20 +1324,27 @@ PHP_DOM_EXPORT bool php_dom_create_object(xmlNodePtr obj, zval *return_value, do
 	if (domobj && domobj->document) {
 		ce = dom_get_doc_classmap(domobj->document, ce);
 	}
+	php_dom_instantiate_object_helper(return_value, ce, obj, domobj);
+	return 0;
+}
+/* }}} end php_domobject_new */
+
+dom_object *php_dom_instantiate_object_helper(zval *return_value, zend_class_entry *ce, xmlNodePtr obj, dom_object *parent)
+{
 	object_init_ex(return_value, ce);
 
-	intern = Z_DOMOBJ_P(return_value);
+	dom_object *intern = Z_DOMOBJ_P(return_value);
 	if (obj->doc != NULL) {
-		if (domobj != NULL) {
+		if (parent != NULL) {
-			intern->document = domobj->document;
+			intern->document = parent->document;
 		}
 		php_libxml_increment_doc_ref((php_libxml_node_object *)intern, obj->doc);
 	}
 
 	php_libxml_increment_node_ptr((php_libxml_node_object *)intern, obj, (void *)intern);
-	return 0;
+
+	return intern;
 }
-/* }}} end php_domobject_new */
 
 void php_dom_create_implementation(zval *retval) {
 	object_init_ex(retval, dom_domimplementation_class_entry);
@@ -1819,4 +1901,23 @@ static int dom_nodemap_has_dimension(zend_object *object, zval *member, int chec
 	return offset >= 0 && offset < php_dom_get_namednodemap_length(php_dom_obj_from_obj(object));
 } /* }}} end dom_nodemap_has_dimension */
 
+xmlNodePtr dom_clone_node(xmlNodePtr node, xmlDocPtr doc, const dom_object *intern, bool recursive)
+{
+	/* See http://www.xmlsoft.org/html/libxml-tree.html#xmlDocCopyNode for meaning of values */
+	int extended_recursive = recursive;
+	if (!recursive && node->type == XML_ELEMENT_NODE) {
+		extended_recursive = 2;
+	}
+	xmlNodePtr copy = xmlDocCopyNode(node, doc, extended_recursive);
+	if (UNEXPECTED(!copy)) {
+		return NULL;
+	}
+
+	if (intern->document && intern->document->is_modern_api_class) {
+		dom_mark_namespaces_for_copy_based_on_copy(copy, node);
+	}
+
+	return copy;
+}
+
 #endif /* HAVE_DOM */

ext/dom/php_dom.h

@@ -114,6 +114,8 @@ static inline dom_object_namespace_node *php_dom_namespace_node_obj_from_obj(zen
 
 #include "domexception.h"
 
+#define DOM_HTML_NO_DEFAULT_NS (1U << 31)
+
 dom_object *dom_object_get_data(xmlNodePtr obj);
 dom_doc_propsptr dom_get_doc_props(php_libxml_ref_obj *document);
 libxml_doc_props const* dom_get_doc_props_read_only(const php_libxml_ref_obj *document);
@@ -154,6 +156,17 @@ bool php_dom_adopt_node(xmlNodePtr nodep, dom_object *dom_object_new_document, x
 xmlNsPtr dom_get_ns_resolve_prefix_conflict(xmlNodePtr tree, const char *uri);
 void php_dom_reconcile_attribute_namespace_after_insertion(xmlAttrPtr attrp);
 
+void php_dom_document_constructor(INTERNAL_FUNCTION_PARAMETERS);
+
+dom_object *php_dom_instantiate_object_helper(zval *return_value, zend_class_entry *ce, xmlNodePtr obj, dom_object *parent);
+
+typedef enum {
+	DOM_LOAD_STRING = 0,
+	DOM_LOAD_FILE = 1,
+} dom_load_mode;
+
+xmlDocPtr dom_document_parser(zval *id, dom_load_mode mode, const char *source, size_t source_len, size_t options, xmlCharEncodingHandlerPtr encoding);
+
 /* parentnode */
 void dom_parent_node_prepend(dom_object *context, zval *nodes, uint32_t nodesc);
 void dom_parent_node_append(dom_object *context, zval *nodes, uint32_t nodesc);
@@ -174,6 +187,9 @@ void php_dom_nodelist_get_item_into_zval(dom_nnodemap_object *objmap, zend_long
 int php_dom_get_namednodemap_length(dom_object *obj);
 int php_dom_get_nodelist_length(dom_object *obj);
 
+xmlNodePtr dom_clone_node(xmlNodePtr node, xmlDocPtr doc, const dom_object *intern, bool recursive);
+void dom_mark_namespaces_for_copy_based_on_copy(xmlNodePtr copy, const xmlNode *original);
+
 #define DOM_GET_INTERN(__id, __intern) { \
 	__intern = Z_DOMOBJ_P(__id); \
 	if (UNEXPECTED(__intern->ptr == NULL)) { \

ext/dom/tests/DOMDocument_adoptNode.phpt

@@ -141,7 +141,7 @@ string(27) "<?xml version="1.0"?>
 Not Supported Error
 -- Adopt a document (strict error off) --
 
-Warning: DOMDocument::adoptNode(): Not Supported Error in %s on line %d
+Warning: DOM\Document::adoptNode(): Not Supported Error in %s on line %d
 -- Adopt an attribute --
 bool(true)
 bool(true)

ext/dom/tests/DOMDocument_load_error6.phpt

@@ -23,5 +23,5 @@ var_dump($dom->load(str_repeat(" ", PHP_MAXPATHLEN + 1)));
 ?>
 --EXPECT--
 DOMDocument::load(): Argument #1 ($filename) must not be empty
-Path to document must not contain any null bytes
+DOMDocument::load(): Argument #1 ($filename) must not contain any null bytes
 bool(false)

ext/dom/tests/DOMDocument_relaxNGValidateSource_error1.phpt

@@ -35,5 +35,5 @@ var_dump($result);
 
 ?>
 --EXPECTF--
-Warning: DOMDocument::relaxNGValidateSource(): Did not expect element pear there in %s on line %d
+Warning: DOM\Document::relaxNGValidateSource(): Did not expect element pear there in %s on line %d
 bool(false)

ext/dom/tests/DOMDocument_relaxNGValidateSource_error2.phpt

@@ -31,7 +31,7 @@ var_dump($result);
 
 ?>
 --EXPECTF--
-Warning: DOMDocument::relaxNGValidateSource(): xmlRelaxNGParseElement: element has no content in %s on line %d
+Warning: DOM\Document::relaxNGValidateSource(): xmlRelaxNGParseElement: element has no content in %s on line %d
 
-Warning: DOMDocument::relaxNGValidateSource(): Invalid RelaxNG in %s on line %d
+Warning: DOM\Document::relaxNGValidateSource(): Invalid RelaxNG in %s on line %d
 bool(false)

ext/dom/tests/DOMDocument_relaxNGValidate_error1.phpt

@@ -20,5 +20,5 @@ $result = $doc->relaxNGValidate($rng);
 var_dump($result);
 ?>
 --EXPECTF--
-Warning: DOMDocument::relaxNGValidate(): Did not expect element pear there in %s on line %d
+Warning: DOM\Document::relaxNGValidate(): Did not expect element pear there in %s on line %d
 bool(false)

ext/dom/tests/DOMDocument_relaxNGValidate_error2.phpt

@@ -20,9 +20,9 @@ $result = $doc->relaxNGValidate($rng);
 var_dump($result);
 ?>
 --EXPECTF--
-Warning: DOMDocument::relaxNGValidate(): I/O warning : failed to load external entity "%s/foo.rng" in %s on line %d
+Warning: DOM\Document::relaxNGValidate(): I/O warning : failed to load external entity "%s/foo.rng" in %s on line %d
 
-Warning: DOMDocument::relaxNGValidate(): xmlRelaxNGParse: could not load %s/foo.rng in %s on line %d
+Warning: DOM\Document::relaxNGValidate(): xmlRelaxNGParse: could not load %s/foo.rng in %s on line %d
 
-Warning: DOMDocument::relaxNGValidate(): Invalid RelaxNG in %s on line %d
+Warning: DOM\Document::relaxNGValidate(): Invalid RelaxNG in %s on line %d
 bool(false)

ext/dom/tests/DOMDocument_schemaValidateSource_error1.phpt

@@ -17,13 +17,13 @@ var_dump($result);
 
 ?>
 --EXPECTF--
-Warning: DOMDocument::schemaValidateSource(): Entity: line 1: parser error : Start tag expected, '<' not found in %s.php on line %d
+Warning: DOM\Document::schemaValidateSource(): Entity: line 1: parser error : Start tag expected, '<' not found in %s on line %d
 
-Warning: DOMDocument::schemaValidateSource(): string that is not a schema in %s.php on line %d
+Warning: DOM\Document::schemaValidateSource(): string that is not a schema in %s on line %d
 
-Warning: DOMDocument::schemaValidateSource(): ^ in %s.php on line %d
+Warning: DOM\Document::schemaValidateSource(): ^ in %s on line %d
 
-Warning: DOMDocument::schemaValidateSource(): Failed to parse the XML resource 'in_memory_buffer'. in %s.php on line %d
+Warning: DOM\Document::schemaValidateSource(): Failed to parse the XML resource 'in_memory_buffer'. in %s on line %d
 
-Warning: DOMDocument::schemaValidateSource(): Invalid Schema in %s.php on line %d
+Warning: DOM\Document::schemaValidateSource(): Invalid Schema in %s on line %d
 bool(false)

ext/dom/tests/DOMDocument_schemaValidateSource_error2.phpt

@@ -19,5 +19,5 @@ var_dump($result);
 
 ?>
 --EXPECTF--
-Warning: DOMDocument::schemaValidateSource(): Element 'books': No matching global declaration available for the validation root. in %s.php on line %d
+Warning: DOM\Document::schemaValidateSource(): Element 'books': No matching global declaration available for the validation root. in %s on line %d
 bool(false)

ext/dom/tests/DOMDocument_schemaValidateSource_error3.phpt

@@ -20,4 +20,4 @@ try {
 
 ?>
 --EXPECT--
-DOMDocument::schemaValidateSource(): Argument #1 ($source) must not be empty
+DOM\Document::schemaValidateSource(): Argument #1 ($source) must not be empty

ext/dom/tests/DOMDocument_schemaValidate_error1.phpt

@@ -17,13 +17,13 @@ var_dump($result);
 
 ?>
 --EXPECTF--
-Warning: DOMDocument::schemaValidate(): %sbook-not-a-schema.xsd:1: parser error : Start tag expected, '<' not found in %s.php on line %d
+Warning: DOM\Document::schemaValidate(): %s/book-not-a-schema.xsd:1: parser error : Start tag expected, '<' not found in %s on line %d
 
-Warning: DOMDocument::schemaValidate(): Let's see what happens upon parsing a file that doesn't contain a schema. in %s.php on line %d
+Warning: DOM\Document::schemaValidate(): Let's see what happens upon parsing a file that doesn't contain a schema. in %s on line %d
 
-Warning: DOMDocument::schemaValidate(): ^ in %s.php on line %d
+Warning: DOM\Document::schemaValidate(): ^ in %s on line %d
 
-Warning: DOMDocument::schemaValidate(): Failed to parse the XML resource '%sbook-not-a-schema.xsd'. in %s.php on line %d
+Warning: DOM\Document::schemaValidate(): Failed to parse the XML resource '%s/book-not-a-schema.xsd'. in %s on line %d
 
-Warning: DOMDocument::schemaValidate(): Invalid Schema in %s.php on line %d
+Warning: DOM\Document::schemaValidate(): Invalid Schema in %s on line %d
 bool(false)

ext/dom/tests/DOMDocument_schemaValidate_error2.phpt

@@ -17,5 +17,5 @@ var_dump($result);
 
 ?>
 --EXPECTF--
-Warning: DOMDocument::schemaValidate(): Element 'books': No matching global declaration available for the validation root. in %s.php on line %d
+Warning: DOM\Document::schemaValidate(): Element 'books': No matching global declaration available for the validation root. in %s on line %d
 bool(false)

ext/dom/tests/DOMDocument_schemaValidate_error3.phpt

@@ -20,4 +20,4 @@ try {
 
 ?>
 --EXPECT--
-DOMDocument::schemaValidate(): Argument #1 ($filename) must not be empty
+DOM\Document::schemaValidate(): Argument #1 ($filename) must not be empty

ext/dom/tests/DOMDocument_schemaValidate_error5.phpt

@@ -17,9 +17,9 @@ var_dump($result);
 
 ?>
 --EXPECTF--
-Warning: DOMDocument::schemaValidate(): I/O warning : failed to load external entity "%snon-existent-file" in %s.php on line %d
+Warning: DOM\Document::schemaValidate(): I/O warning : failed to load external entity "%s/non-existent-file" in %s on line %d
 
-Warning: DOMDocument::schemaValidate(): Failed to locate the main schema resource at '%s/non-existent-file'. in %s.php on line %d
+Warning: DOM\Document::schemaValidate(): Failed to locate the main schema resource at '%s/non-existent-file'. in %s on line %d
 
-Warning: DOMDocument::schemaValidate(): Invalid Schema in %s.php on line %d
+Warning: DOM\Document::schemaValidate(): Invalid Schema in %s on line %d
 bool(false)

ext/dom/tests/DOMDocument_schemaValidate_error6.phpt

@@ -19,7 +19,7 @@ var_dump($doc->schemaValidate(str_repeat(" ", PHP_MAXPATHLEN + 1)));
 
 ?>
 --EXPECTF--
-DOMDocument::schemaValidate(): Argument #1 ($filename) must not contain any null bytes
+DOM\Document::schemaValidate(): Argument #1 ($filename) must not contain any null bytes
 
-Warning: DOMDocument::schemaValidate(): Invalid Schema file source in %s on line %d
+Warning: DOM\Document::schemaValidate(): Invalid Schema file source in %s on line %d
 bool(false)

ext/dom/tests/DOMDocument_strictErrorChecking_variation.phpt

@@ -56,4 +56,4 @@ See if strictErrorChecking is off
 bool(false)
 Should raise PHP error because strictErrorChecking is off
 
-Warning: DOMDocument::createAttribute(): Invalid Character Error in %sDOMDocument_strictErrorChecking_variation.php on line %d
+Warning: DOM\Document::createAttribute(): Invalid Character Error in %s on line %d

ext/dom/tests/domobject_debug_handler.phpt

@@ -20,14 +20,12 @@ object(DOMDocument)#1 (41) {
   ["dynamicProperty"]=>
   object(stdClass)#2 (0) {
   }
-  ["doctype"]=>
-  NULL
   ["implementation"]=>
   string(22) "(object value omitted)"
-  ["documentElement"]=>
-  string(22) "(object value omitted)"
   ["actualEncoding"]=>
   NULL
+  ["config"]=>
+  NULL
   ["encoding"]=>
   NULL
   ["xmlEncoding"]=>
@@ -40,12 +38,6 @@ object(DOMDocument)#1 (41) {
   string(3) "1.0"
   ["xmlVersion"]=>
   string(3) "1.0"
-  ["strictErrorChecking"]=>
-  bool(true)
-  ["documentURI"]=>
-  string(%d) %s
-  ["config"]=>
-  NULL
   ["formatOutput"]=>
   bool(false)
   ["validateOnParse"]=>
@@ -58,6 +50,14 @@ object(DOMDocument)#1 (41) {
   bool(false)
   ["substituteEntities"]=>
   bool(false)
+  ["doctype"]=>
+  NULL
+  ["documentElement"]=>
+  string(22) "(object value omitted)"
+  ["strictErrorChecking"]=>
+  bool(true)
+  ["documentURI"]=>
+  string(%d) "%s"
   ["firstElementChild"]=>
   string(22) "(object value omitted)"
   ["lastElementChild"]=>

ext/dom/tests/modern/html/encoding/HTMLDocument_GB18030.phpt

@@ -0,0 +1,36 @@
+--TEST--
+DOM\HTMLDocument GB18030 encoding test
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . "/gb18030.html");
+var_dump($dom->encoding);
+$dom->documentElement->firstChild->nextElementSibling->textContent = "é";
+$output = $dom->saveHTML();
+echo $output, "\n";
+$dom->saveHTMLFile(__DIR__ . "/gb18030_output.tmp");
+var_dump(file_get_contents(__DIR__ . "/gb18030_output.tmp") === $output);
+
+echo "--- After changing encoding to UTF-8 ---\n";
+$dom->encoding = "UTF-8";
+echo $dom->saveHTML(), "\n";
+
+?>
+--CLEAN--
+<?php
+@unlink(__DIR__ . "/gb18030_output.tmp");
+?>
+--EXPECT--
+string(7) "gb18030"
+<!DOCTYPE html><html><head>
+<meta charset="gb18030">
+</head>
+<body>¨¦</body></html>
+bool(true)
+--- After changing encoding to UTF-8 ---
+<!DOCTYPE html><html><head>
+<meta charset="gb18030">
+</head>
+<body>é</body></html>

ext/dom/tests/modern/html/encoding/HTMLDocument_Shift_JIS.phpt

@@ -0,0 +1,40 @@
+--TEST--
+DOM\HTMLDocument Shift JIS encoding test
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . "/shift_jis.html");
+var_dump($dom->encoding);
+$dom->documentElement->firstChild->nextElementSibling->textContent .= "é";
+$output = $dom->saveHTML();
+echo $output, "\n";
+$dom->saveHTMLFile(__DIR__ . "/shift_jis.tmp");
+var_dump(file_get_contents(__DIR__ . "/shift_jis.tmp") === $output);
+
+echo "--- After changing encoding to UTF-8 ---\n";
+$dom->encoding = "UTF-8";
+echo $dom->saveHTML(), "\n";
+
+?>
+--CLEAN--
+<?php
+@unlink(__DIR__ . "/shift_jis.tmp");
+?>
+--EXPECT--
+string(9) "Shift_JIS"
+<!DOCTYPE html><html><head>
+<meta charset="shift_jis">
+</head>
+<body>
+    ‚â‚ 
+?</body></html>
+bool(true)
+--- After changing encoding to UTF-8 ---
+<!DOCTYPE html><html><head>
+<meta charset="shift_jis">
+</head>
+<body>
+    ã‚„ã<E2809E>‚
+é</body></html>

ext/dom/tests/modern/html/encoding/HTMLDocument_UTF16BE_BOM.phpt

@@ -0,0 +1,38 @@
+--TEST--
+DOM\HTMLDocument UTF-16BE BOM encoding test
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . "/utf16be_bom.html");
+var_dump($dom->encoding);
+$dom->documentElement->firstChild->nextElementSibling->textContent = "é";
+$output = $dom->saveHTML();
+echo $output, "\n";
+$dom->saveHTMLFile(__DIR__ . "/utf16be_bom_output.tmp");
+var_dump(file_get_contents(__DIR__ . "/utf16be_bom_output.tmp") === $output);
+
+echo "--- After changing encoding to UTF-8 ---\n";
+$dom->encoding = "UTF-8";
+echo $dom->saveHTML(), "\n";
+
+?>
+--CLEAN--
+<?php
+@unlink(__DIR__ . "/utf16be_bom_output.tmp");
+?>
+--EXPECTF--
+string(8) "UTF-16BE"
+%0<%0!%0D%0O%0C%0T%0Y%0P%0E%0 %0h%0t%0m%0l%0>%0<%0h%0t%0m%0l%0>%0<%0h%0e%0a%0d%0>%0
+%0<%0!%0-%0-%0 %0i%0n%0t%0e%0n%0t%0i%0o%0n%0a%0l%0 %0l%0i%0e%0s%0 %0a%0n%0d%0 %0d%0e%0c%0e%0i%0t%0 %0-%0-%0>%0
+%0<%0m%0e%0t%0a%0 %0c%0h%0a%0r%0s%0e%0t%0=%0"%0u%0t%0f%0-%08%0"%0>%0
+%0<%0/%0h%0e%0a%0d%0>%0
+%0<%0b%0o%0d%0y%0>%0é%0<%0/%0b%0o%0d%0y%0>%0<%0/%0h%0t%0m%0l%0>
+bool(true)
+--- After changing encoding to UTF-8 ---
+<!DOCTYPE html><html><head>
+<!-- intentional lies and deceit -->
+<meta charset="utf-8">
+</head>
+<body>é</body></html>

ext/dom/tests/modern/html/encoding/HTMLDocument_UTF16LE_BOM.phpt

@@ -0,0 +1,38 @@
+--TEST--
+DOM\HTMLDocument UTF-16LE BOM encoding test
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . "/utf16le_bom.html");
+var_dump($dom->encoding);
+$dom->documentElement->firstChild->nextElementSibling->textContent = "é";
+$output = $dom->saveHTML();
+echo $output, "\n";
+$dom->saveHTMLFile(__DIR__ . "/utf16le_bom_output.tmp");
+var_dump(file_get_contents(__DIR__ . "/utf16le_bom_output.tmp") === $output);
+
+echo "--- After changing encoding to UTF-8 ---\n";
+$dom->encoding = "UTF-8";
+echo $dom->saveHTML(), "\n";
+
+?>
+--CLEAN--
+<?php
+@unlink(__DIR__ . "/utf16le_bom_output.tmp");
+?>
+--EXPECTF--
+string(8) "UTF-16LE"
+<%0!%0D%0O%0C%0T%0Y%0P%0E%0 %0h%0t%0m%0l%0>%0<%0h%0t%0m%0l%0>%0<%0h%0e%0a%0d%0>%0
+%0<%0!%0-%0-%0 %0i%0n%0t%0e%0n%0t%0i%0o%0n%0a%0l%0 %0l%0i%0e%0s%0 %0a%0n%0d%0 %0d%0e%0c%0e%0i%0t%0 %0-%0-%0>%0
+%0<%0m%0e%0t%0a%0 %0c%0h%0a%0r%0s%0e%0t%0=%0"%0u%0t%0f%0-%08%0"%0>%0
+%0<%0/%0h%0e%0a%0d%0>%0
+%0<%0b%0o%0d%0y%0>%0é%0<%0/%0b%0o%0d%0y%0>%0<%0/%0h%0t%0m%0l%0>%0
+bool(true)
+--- After changing encoding to UTF-8 ---
+<!DOCTYPE html><html><head>
+<!-- intentional lies and deceit -->
+<meta charset="utf-8">
+</head>
+<body>é</body></html>

ext/dom/tests/modern/html/encoding/HTMLDocument_UTF8_BOM.phpt

@@ -0,0 +1,38 @@
+--TEST--
+DOM\HTMLDocument UTF-8 BOM encoding test
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . "/utf8_bom.html");
+var_dump($dom->encoding);
+$dom->documentElement->firstChild->nextElementSibling->textContent = "é";
+$output = $dom->saveHTML();
+echo $output, "\n";
+$dom->saveHTMLFile(__DIR__ . "/utf8_bom_output.tmp");
+var_dump(file_get_contents(__DIR__ . "/utf8_bom_output.tmp") === $output);
+
+echo "--- After changing encoding to UTF-8 ---\n";
+$dom->encoding = "UTF-8";
+echo $dom->saveHTML(), "\n";
+
+?>
+--CLEAN--
+<?php
+@unlink(__DIR__ . "/utf8_bom_output.tmp");
+?>
+--EXPECT--
+string(5) "UTF-8"
+<!DOCTYPE html><html><head>
+<!-- intentional lies and deceit -->
+<meta charset="utf-16">
+</head>
+<body>é</body></html>
+bool(true)
+--- After changing encoding to UTF-8 ---
+<!DOCTYPE html><html><head>
+<!-- intentional lies and deceit -->
+<meta charset="utf-16">
+</head>
+<body>é</body></html>

ext/dom/tests/modern/html/encoding/HTMLDocument_Windows1251.phpt

@@ -0,0 +1,40 @@
+--TEST--
+DOM\HTMLDocument Windows-1251 encoding test
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . "/windows1251.html");
+var_dump($dom->encoding);
+$dom->documentElement->firstChild->nextElementSibling->textContent .= "é"; // Note: won't show up in Windows 1251 because it doesn't exist there
+$output = $dom->saveHTML();
+echo $output, "\n";
+$dom->saveHTMLFile(__DIR__ . "/windows1251_output.tmp");
+var_dump(file_get_contents(__DIR__ . "/windows1251_output.tmp") === $output);
+
+echo "--- After changing encoding to UTF-8 ---\n";
+$dom->encoding = "UTF-8";
+echo $dom->saveHTML(), "\n";
+
+?>
+--CLEAN--
+<?php
+@unlink(__DIR__ . "/windows1251_output.tmp");
+?>
+--EXPECT--
+string(12) "windows-1251"
+<!DOCTYPE html><html><head>
+<meta charset="windows-1251">
+</head>
+<body>
+    A ô B á C
+?</body></html>
+bool(true)
+--- After changing encoding to UTF-8 ---
+<!DOCTYPE html><html><head>
+<meta charset="windows-1251">
+</head>
+<body>
+    A ф B б C
+é</body></html>

ext/dom/tests/modern/html/encoding/HTMLDocument_createFromFile_http_header.phpt

@@ -0,0 +1,96 @@
+--TEST--
+DOM\HTMLDocument::createFromFile() HTTP header Content-Type
+--EXTENSIONS--
+dom
+--SKIPIF--
+<?php
+if (@!include "./ext/standard/tests/http/server.inc") die('skip server.inc not available');
+http_server_skipif();
+?>
+--FILE--
+<?php
+require "./ext/standard/tests/http/server.inc";
+
+$tests = [
+    "Invalid type/subtype" => [
+        "/html; Charset=\"ISO-8859-1\"",
+        "text/; Charset=\"ISO-8859-1\"",
+        "tex°t/html; Charset=\"ISO-8859-1\"",
+        "/; Charset=\"ISO-8859-1\"",
+        "$/€; Charset=\"ISO-8859-1\"",
+        "; Charset=\"ISO-8859-1\"",
+        ";",
+        "",
+        "    \t",
+    ],
+    "Valid type/subtype without charset" => [
+        "text/html; x=ISO-8859-1",
+        "text/html; x=\"ISO-8859-1\"",
+        "text/html; charet=\"ISO-8859-1\"",
+        "text/html; chars et=\"ISO-8859-1\"",
+    ],
+    "All valid inputs" => [
+        "text/html; charset=ISO-8859-1",
+        "\t\r text/html; charset=ISO-8859-1   \t",
+        "text/html; foo=bar;charset=ISO-8859-1",
+        "text/html; foo=bar;charset=ISO-8859-1;bar=\"foooooo\"",
+        "text/html;;;; charset=ISO-8859-1",
+        "text/html; Charset=\"ISO-8859-1\"",
+        "text/html; Charset=\"ISO\\-8859-1\"",
+        "text/html; ;;   ; ;; Charset=\"ISO-8859-1\"",
+        "text/html;Charset=\"ISO-8859-1",
+        "tex.t/h#\$%!&'*%2B-.^_`|~tml;Charset=\"ISO-8859-1\"", // Note: have to encode + as 2B because of implementation details of http_server()
+    ],
+    "Valid input, but invalid encoding name" => [
+        "text/html;Charset=\"ISO-8859-1\\",
+        "text/html;Charset=\"ISO-8859-1\\\"",
+        "text/html;Charset=\"foobar\\\"",
+        "text/html;Charset=\"\\\"",
+        "text/html;Charset=",
+    ],
+];
+
+foreach ($tests as $name => $headers) {
+    echo "--- $name ---\n";
+    $responses = array_map(fn ($header) => "data://text/plain,HTTP/1.1 200 OK\r\nContent-Type: " . $header . "\r\n\r\n" . "<p>\xE4\xF6\xFC</p>\n", $headers);
+    ['pid' => $pid, 'uri' => $uri] = http_server($responses);
+    for ($i = 0; $i < count($responses); $i++) {
+        $result = DOM\HTMLDocument::createFromFile($uri, LIBXML_NOERROR);
+        echo $result->textContent;
+    }
+    http_server_kill($pid);
+}
+?>
+--EXPECT--
+--- Invalid type/subtype ---
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>
+--- Valid type/subtype without charset ---
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>
+--- All valid inputs ---
+äöü
+äöü
+äöü
+äöü
+äöü
+äöü
+äöü
+äöü
+äöü
+äöü
+--- Valid input, but invalid encoding name ---
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>
+<EFBFBD><EFBFBD><EFBFBD>

ext/dom/tests/modern/html/encoding/HTMLDocument_createFromFile_override_encoding.phpt

@@ -0,0 +1,33 @@
+--TEST--
+DOM\HTMLDocument::createFromFile() with overrideEncoding
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+try {
+    DOM\HTMLDocument::createFromFile(__DIR__ . '/gb18030_without_charset.html', overrideEncoding: 'nonexistent');
+} catch (ValueError $e) {
+    echo $e->getMessage(), "\n";
+}
+
+// The override encoding matches with the document encoding attribute
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . '/gb18030_without_charset.html', overrideEncoding: 'GB18030');
+var_dump($dom->documentElement->lastChild->textContent);
+var_dump($dom->encoding);
+
+// The override encoding mismatches with the document encoding attribute
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . '/fallback_encoding.html', overrideEncoding: 'Windows-1252');
+var_dump($dom->documentElement->lastChild->textContent);
+var_dump($dom->encoding);
+
+?>
+--EXPECT--
+DOM\HTMLDocument::createFromFile(): Argument #3 ($overrideEncoding) must be a valid document encoding
+string(20) "
+    Héllo, world!
+"
+string(7) "gb18030"
+string(1) "
+"
+string(12) "windows-1252"

ext/dom/tests/modern/html/encoding/HTMLDocument_createFromString_override_encoding.phpt

@@ -0,0 +1,33 @@
+--TEST--
+DOM\HTMLDocument::createFromString() with overrideEncoding
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+try {
+    DOM\HTMLDocument::createFromString(file_get_contents(__DIR__ . '/gb18030_without_charset.html'), overrideEncoding: 'nonexistent');
+} catch (ValueError $e) {
+    echo $e->getMessage(), "\n";
+}
+
+// The override encoding matches with the document encoding attribute
+$dom = DOM\HTMLDocument::createFromString(file_get_contents(__DIR__ . '/gb18030_without_charset.html'), overrideEncoding: 'GB18030');
+var_dump($dom->documentElement->lastChild->textContent);
+var_dump($dom->encoding);
+
+// The override encoding mismatches with the document encoding attribute
+$dom = DOM\HTMLDocument::createFromString(file_get_contents(__DIR__ . '/fallback_encoding.html'), overrideEncoding: 'Windows-1252');
+var_dump($dom->documentElement->lastChild->textContent);
+var_dump($dom->encoding);
+
+?>
+--EXPECT--
+DOM\HTMLDocument::createFromString(): Argument #3 ($overrideEncoding) must be a valid document encoding
+string(20) "
+    Héllo, world!
+"
+string(7) "gb18030"
+string(1) "
+"
+string(12) "windows-1252"

ext/dom/tests/modern/html/encoding/HTMLDocument_encoding_edge_case_01.phpt

@@ -0,0 +1,16 @@
+--TEST--
+DOM\HTMLDocument edge case encoding 01
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+// UTF-8 -> UTF-8
+// Create a UTF-8 string where a UTF-8 byte sequence falls over the boundary of the 4096 byte buffer
+$dom = DOM\HTMLDocument::createEmpty();
+$dom->append(str_repeat("A", 4096 - 2) . "\xf0\x90\x8d\x88AA");
+var_dump($dom->saveHTML());
+
+?>
+--EXPECT--
+string(4100) "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA𐍈AA"

ext/dom/tests/modern/html/encoding/HTMLDocument_encoding_edge_case_02.phpt

@@ -0,0 +1,27 @@
+--TEST--
+DOM\HTMLDocument edge case encoding 02
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+// UTF-8 -> GB18030
+$dom = DOM\HTMLDocument::createEmpty("GB18030");
+// Create a UTF-8 string where a UTF-8 byte sequence falls over the boundary of the 4096 byte buffer
+// *and* the sequence also falls over the boundary for the result
+$dom->append(str_repeat("A", 4096 - 2) . "\xf0\x90\x8d\x88AA");
+var_dump($output = $dom->saveHTML());
+
+// GB18030 encoding of the above UTF-8 symbol
+var_dump($output[4094] == "\x90");
+var_dump($output[4095] == "\x30");
+var_dump($output[4096] == "\xd5");
+var_dump($output[4097] == "\x30");
+
+?>
+--EXPECT--
+string(4100) "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA𐍈AA"
+bool(true)
+bool(true)
+bool(true)
+bool(true)

ext/dom/tests/modern/html/encoding/HTMLDocument_encoding_edge_case_03.phpt

@@ -0,0 +1,17 @@
+--TEST--
+DOM\HTMLDocument edge case encoding 03
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+// UTF-8 -> GB18030
+$dom = DOM\HTMLDocument::createEmpty("GB18030");
+// Create a UTF-8 string where an invalid UTF-8 byte sequence falls over the boundary of the 4096 byte buffer
+// Note: the strange ?1?7 sequence is the GB18030 encoding for the unicode replacement character
+$dom->append(str_repeat("A", 4096 - 2) . "\xff\xff\xff");
+var_dump($dom->saveHTML());
+
+?>
+--EXPECT--
+string(4106) "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA<41><41><EFBFBD>"

ext/dom/tests/modern/html/encoding/HTMLDocument_encoding_edge_case_04.phpt

@@ -0,0 +1,16 @@
+--TEST--
+DOM\HTMLDocument edge case encoding 04
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+// UTF-8 -> UTF-8
+$dom = DOM\HTMLDocument::createEmpty();
+// Create a UTF-8 string where an invalid UTF-8 byte sequence falls over the boundary of the 4096 byte buffer
+$dom->append(str_repeat("A", 4096 - 2) . "\xff\xff\xff");
+var_dump($dom->saveHTML());
+
+?>
+--EXPECT--
+string(4103) "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA<41><41><EFBFBD>"

ext/dom/tests/modern/html/encoding/HTMLDocument_encoding_edge_case_05.phpt

@@ -0,0 +1,21 @@
+--TEST--
+DOM\HTMLDocument edge case encoding 05
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+// GB18030 -> UTF-8
+$header = "<!doctype html><html><head><meta charset=\"gb18030\"></head><body>";
+$padding_required_until_4094 = 4094 - strlen($header);
+$trailer = "\x90\x30\xd5\x30";
+$dom = DOM\HTMLDocument::createFromString($header . str_repeat("A", $padding_required_until_4094) . $trailer);
+// GB18030 byte sequence crossing the 4096 boundary
+var_dump($dom->encoding);
+$dom->encoding = "UTF-8";
+var_dump($dom->saveHTML());
+
+?>
+--EXPECT--
+string(7) "gb18030"
+string(4112) "<!DOCTYPE html><html><head><meta charset="gb18030"></head><body>AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA𐍈</body></html>"

ext/dom/tests/modern/html/encoding/HTMLDocument_encoding_edge_case_06.phpt

@@ -0,0 +1,16 @@
+--TEST--
+DOM\HTMLDocument edge case encoding 06
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+// UTF-8 -> UTF-8
+$dom = DOM\HTMLDocument::createEmpty();
+// Create a UTF-8 string where a *broken* UTF-8 byte sequence falls over the boundary of the 4096 byte buffer
+$dom->append(str_repeat("A", 4096 - 1) . "\xf0\x90");
+var_dump($dom->saveHTML());
+
+?>
+--EXPECT--
+string(4101) "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA<41><41>"

ext/dom/tests/modern/html/encoding/HTMLDocument_encoding_edge_case_07.phpt

@@ -0,0 +1,21 @@
+--TEST--
+DOM\HTMLDocument edge case encoding 07
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+// GB18030 -> UTF-8
+$header = "<!doctype html><html><head><meta charset=\"gb18030\"></head><body>";
+$padding_required_until_4095 = 4095 - strlen($header);
+$trailer = "\x90\x30";
+$dom = DOM\HTMLDocument::createFromString($header . str_repeat("A", $padding_required_until_4095) . $trailer);
+// GB18030 *broken* byte sequence crossing the 4096 boundary
+var_dump($dom->encoding);
+$dom->encoding = "UTF-8";
+var_dump($dom->saveHTML());
+
+?>
+--EXPECT--
+string(7) "gb18030"
+string(4115) "<!DOCTYPE html><html><head><meta charset="gb18030"></head><body>AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA<41><41></body></html>"

ext/dom/tests/modern/html/encoding/HTMLDocument_encoding_field_test.phpt

@@ -0,0 +1,43 @@
+--TEST--
+DOM\HTMLDocument test values for encoding field
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createEmpty();
+var_dump($dom->encoding);
+$dom->encoding = "CSeuckr";
+var_dump($dom->encoding);
+try {
+    $dom->encoding = "nope";
+} catch (ValueError $e) {
+    echo $e->getMessage(), "\n";
+}
+var_dump($dom->encoding);
+$dom->encoding = "Windows-1251";
+var_dump($dom->encoding);
+try {
+    $dom->encoding = NULL;
+} catch (ValueError $e) {
+    echo $e->getMessage(), "\n";
+}
+var_dump($dom->encoding);
+echo $dom->saveHTML();
+
+try {
+    $dom = DOM\HTMLDocument::createEmpty("bogus");
+} catch (ValueError $e) {
+    echo $e->getMessage(), "\n";
+}
+
+?>
+--EXPECT--
+string(5) "UTF-8"
+string(6) "EUC-KR"
+Invalid document encoding
+string(6) "EUC-KR"
+string(12) "windows-1251"
+Invalid document encoding
+string(12) "windows-1251"
+DOM\HTMLDocument::createEmpty(): Argument #1 ($encoding) must be a valid document encoding

ext/dom/tests/modern/html/encoding/HTMLDocument_encoding_unicode_error.phpt

@@ -0,0 +1,26 @@
+--TEST--
+DOM\HTMLDocument loading with unicode codepoints resulting in an error
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+echo "--- createFromFile ---\n";
+DOM\HTMLDocument::createFromFile(__DIR__."/utf16le_error.html");
+echo "--- createFromString ---\n";
+DOM\HTMLDocument::createFromString(file_get_contents(__DIR__."/utf16le_error.html"));
+?>
+--EXPECTF--
+--- createFromFile ---
+
+Warning: DOM\HTMLDocument::createFromFile(): tokenizer error missing-end-tag-name in %s on line %d
+
+Warning: DOM\HTMLDocument::createFromFile(): tree error unexpected-token in %s on line %d
+
+Warning: DOM\HTMLDocument::createFromFile(): tree error unexpected-token in %s on line %d
+--- createFromString ---
+
+Warning: DOM\HTMLDocument::createFromString(): tokenizer error missing-end-tag-name in Entity, line: 7, column: 29 in %s on line %d
+
+Warning: DOM\HTMLDocument::createFromString(): tree error unexpected-token in Entity, line: 7, column: 14-17 in %s on line %d
+
+Warning: DOM\HTMLDocument::createFromString(): tree error unexpected-token in Entity, line: 8, column: 7-10 in %s on line %d

ext/dom/tests/modern/html/encoding/HTMLDocument_fallback_encoding.phpt

@@ -0,0 +1,23 @@
+--TEST--
+DOM\HTMLDocument fallback encoding test
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . "/fallback_encoding.html");
+var_dump($dom->encoding);
+echo $dom->saveHTML();
+
+?>
+--CLEAN--
+<?php
+@unlink(__DIR__ . "/windows1251_output.tmp");
+?>
+--EXPECT--
+string(5) "UTF-8"
+<!DOCTYPE html><html><head>
+<meta charset="I don't exist">
+</head>
+<body>
+</body></html>

ext/dom/tests/modern/html/encoding/HTMLDocument_override_encoding_incompatible_charset.phpt

@@ -0,0 +1,36 @@
+--TEST--
+DOM\HTMLDocument: overrideEncoding with incompatible charset
+--EXTENSIONS--
+iconv
+dom
+--FILE--
+<?php
+$doc = DOM\HTMLDocument::createFromString(
+	iconv(
+		'ISO-8859-1',
+		'UTF-8',
+		<<<'DOC'
+			<!DOCTYPE html>
+			<html>
+				<head>
+					<meta charset="iso-8859-1">
+					<title>äöü</title>
+				</head>
+				<body>
+					äöü
+				</body>
+			</html>
+			DOC,
+	),
+	overrideEncoding: 'utf-8'
+);
+
+var_dump(iconv('UTF-8', 'ISO-8859-1', $doc->getElementsByTagName('title')->item(0)->textContent));
+var_dump(iconv('UTF-8', 'ISO-8859-1', $doc->getElementsByTagName('body')->item(0)->textContent));
+?>
+--EXPECT--
+string(3) "äöü"
+string(9) "
+		äöü
+	
+"

ext/dom/tests/modern/html/encoding/fallback_encoding.html

@@ -0,0 +1,6 @@
+<!doctype html>
+<head>
+<meta charset="I don't exist">
+</head>
+<body>
+</body>

ext/dom/tests/modern/html/encoding/gb18030.html

@@ -0,0 +1,7 @@
+<!doctype html>
+<head>
+<meta charset="gb18030">
+</head>
+<body>
+    H¨¦llo, world!
+</body>

ext/dom/tests/modern/html/encoding/gb18030_without_charset.html

@@ -0,0 +1,7 @@
+<!doctype html>
+<head>
+<title>No charset!</title>
+</head>
+<body>
+    H¨¦llo, world!
+</body>

ext/dom/tests/modern/html/encoding/shift_jis.html

@@ -0,0 +1,7 @@
+<!doctype html>
+<head>
+<meta charset="shift_jis">
+</head>
+<body>
+    <p>やあ</p>
+</body>

ext/dom/tests/modern/html/encoding/utf8_bom.html

@@ -0,0 +1,7 @@
+<!doctype html>
+<head>
+<!-- intentional lies and deceit -->
+<meta charset="utf-16">
+</head>
+<body>
+</body>

ext/dom/tests/modern/html/encoding/windows1251.html

@@ -0,0 +1,7 @@
+<!doctype html>
+<head>
+<meta charset="windows-1251">
+</head>
+<body>
+    A ô B á C
+</body>

ext/dom/tests/modern/html/interactions/HTMLDocument_adopt_DOMDocument.phpt

@@ -0,0 +1,30 @@
+--TEST--
+DOM\HTMLDocument adopts a DOMDocument
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = new DOMDocument();
+$dom->loadHTML(<<<HTML
+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<html>
+<head>
+</head>
+<body>
+</body>
+</html>
+HTML);
+
+$dom2 = DOM\HTMLDocument::createEmpty();
+$dom2->appendChild($dom2->adoptNode($dom->documentElement));
+echo $dom2->saveHTML();
+
+?>
+--EXPECT--
+<html>
+<head>
+</head>
+<body>
+</body>
+</html>

ext/dom/tests/modern/html/interactions/HTMLDocument_clone.phpt

@@ -0,0 +1,23 @@
+--TEST--
+Cloning a DOM\HTMLDocument
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromString("<p>foo</p>");
+
+$dom2 = clone $dom;
+var_dump($dom2->firstChild->tagName);
+var_dump($dom2->firstChild->textContent);
+
+$element = $dom2->firstChild;
+unset($dom2);
+var_dump(get_class($element->ownerDocument));
+
+?>
+--EXPECTF--
+Warning: DOM\HTMLDocument::createFromString(): tree error unexpected-token-in-initial-mode in Entity, line: 1, column: 2 in %s on line %d
+string(4) "html"
+string(3) "foo"
+string(16) "DOM\HTMLDocument"

ext/dom/tests/modern/html/interactions/HTMLDocument_documentURI.phpt

@@ -0,0 +1,47 @@
+--TEST--
+DOM\HTMLDocument::documentURI
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . "/test foo.html", LIBXML_NOERROR);
+var_dump($dom->documentURI);
+
+$memory = fopen("php://memory", "w+");
+fwrite($memory, "foobar");
+rewind($memory);
+$dom = DOM\HTMLDocument::createFromFile("php://memory");
+var_dump($dom->documentURI);
+fclose($memory);
+
+class DummyWrapper {
+    public $context;
+
+    public function stream_open($path, $mode, $options, &$opened_path) {
+        return true;
+    }
+
+    public function stream_read($count) {
+        return "";
+    }
+
+    public function stream_eof() {
+        return true;
+    }
+
+    public function stream_close() {
+        return true;
+    }
+}
+
+stream_wrapper_register("dummy", DummyWrapper::class);
+
+$dom = DOM\HTMLDocument::createFromFile("dummy://foo/ bar");
+var_dump($dom->documentURI);
+
+?>
+--EXPECTF--
+string(%d) "file:/%stest%20foo.html"
+string(12) "php://memory"
+string(16) "dummy://foo/ bar"

ext/dom/tests/modern/html/interactions/HTMLDocument_getElementsByTagName.phpt

@@ -0,0 +1,98 @@
+--TEST--
+Test DOM\HTMLDocument::getElementsByTagName(NS)
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromString(<<<HTML
+<!DOCTYPE html>
+<html>
+<head>
+    <title>Test</title>
+</head>
+<body>
+    <h1>Test</h1>
+    <p>Test</p>
+    <svg width="100" height="100">
+        <circle cx="0" cy="0" r="10"/>
+    </svg>
+    <math>
+        <mtable id="table"></mtable>
+    </math>
+</body>
+</html>
+HTML);
+
+echo "--- getElementsByTagName ---\n";
+
+var_dump($dom->getElementsByTagName("p")[0]?->nodeName);
+var_dump($dom->getElementsByTagName("math")[0]?->nodeName);
+var_dump($dom->getElementsByTagName("mtable")[0]?->nodeName);
+var_dump($dom->getElementsByTagName("svg")[0]?->nodeName);
+var_dump($dom->getElementsByTagName("circle")[0]?->nodeName);
+
+echo "--- getElementsByTagNameNS (*) ---\n";
+
+var_dump($dom->getElementsByTagNameNS("*", "p")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("*", "math")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("*", "mtable")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("*", "svg")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("*", "circle")[0]?->nodeName);
+
+echo "--- getElementsByTagNameNS (xhtml) ---\n";
+
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/1999/xhtml", "p")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/1999/xhtml", "math")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/1999/xhtml", "mtable")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/1999/xhtml", "svg")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/1999/xhtml", "circle")[0]?->nodeName);
+
+echo "--- getElementsByTagNameNS (svg) ---\n";
+
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/2000/svg", "p")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/2000/svg", "math")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/2000/svg", "mtable")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/2000/svg", "svg")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/2000/svg", "circle")[0]?->nodeName);
+
+echo "--- getElementsByTagNameNS (math) ---\n";
+
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/1998/Math/MathML", "p")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/1998/Math/MathML", "math")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/1998/Math/MathML", "mtable")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/1998/Math/MathML", "svg")[0]?->nodeName);
+var_dump($dom->getElementsByTagNameNS("http://www.w3.org/1998/Math/MathML", "circle")[0]?->nodeName);
+
+?>
+--EXPECT--
+--- getElementsByTagName ---
+string(1) "p"
+string(4) "math"
+string(6) "mtable"
+string(3) "svg"
+string(6) "circle"
+--- getElementsByTagNameNS (*) ---
+string(1) "p"
+string(4) "math"
+string(6) "mtable"
+string(3) "svg"
+string(6) "circle"
+--- getElementsByTagNameNS (xhtml) ---
+string(1) "p"
+NULL
+NULL
+NULL
+NULL
+--- getElementsByTagNameNS (svg) ---
+NULL
+NULL
+NULL
+string(3) "svg"
+string(6) "circle"
+--- getElementsByTagNameNS (math) ---
+NULL
+string(4) "math"
+string(6) "mtable"
+NULL
+NULL

ext/dom/tests/modern/html/interactions/HTMLDocument_registerNodeClass_01.phpt

@@ -0,0 +1,17 @@
+--TEST--
+DOM\HTMLDocument::registerNodeClass 01
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = new DOMDocument();
+$dom->registerNodeClass("DOM\\HTMLDocument", "DOMDocument");
+
+?>
+--EXPECTF--
+Fatal error: Uncaught Error: DOM\Document::registerNodeClass(): Argument #2 ($extendedClass) must be a class name derived from DOM\HTMLDocument or null, DOMDocument given in %s:%d
+Stack trace:
+#0 %s(%d): DOM\Document->registerNodeClass('DOM\\HTMLDocumen...', 'DOMDocument')
+#1 {main}
+  thrown in %s on line %d

ext/dom/tests/modern/html/interactions/HTMLDocument_registerNodeClass_02.phpt

@@ -0,0 +1,36 @@
+--TEST--
+DOM\HTMLDocument::registerNodeClass 02
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+class Custom extends DOM\Document {
+    public function foo() {
+    }
+}
+
+$dom = new DOMDocument();
+try {
+    $dom->registerNodeClass("DOM\\Document", "Custom");
+} catch (ValueError $e) {
+    echo $e->getMessage(), "\n";
+}
+
+$element = $dom->appendChild($dom->createElement("foo"));
+unset($dom);
+
+var_dump(get_class($element->ownerDocument));
+
+// Should fail
+$element->ownerDocument->foo();
+
+?>
+--EXPECTF--
+DOM\Document::registerNodeClass(): Argument #1 ($baseClass) must not be an abstract class
+string(11) "DOMDocument"
+
+Fatal error: Uncaught Error: Call to undefined method DOMDocument::foo() in %s:%d
+Stack trace:
+#0 {main}
+  thrown in %s on line %d

ext/dom/tests/modern/html/interactions/HTMLDocument_saveHTMLFile_empty_path.phpt

@@ -0,0 +1,18 @@
+--TEST--
+DOM\HTMLDocument::saveHTMLFile() empty path
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createEmpty();
+$root = $dom->appendChild($dom->createElement("root"));
+$dom->saveHTMLFile("");
+
+?>
+--EXPECTF--
+Fatal error: Uncaught ValueError: DOM\HTMLDocument::saveHTMLFile(): Argument #1 ($filename) must not be empty in %s:%d
+Stack trace:
+#0 %s(%d): DOM\HTMLDocument->saveHTMLFile('')
+#1 {main}
+  thrown in %s on line %d

ext/dom/tests/modern/html/interactions/HTMLDocument_saveHTML_wrong_document.phpt

@@ -0,0 +1,17 @@
+--TEST--
+DOM\HTMLDocument::saveHTML() wrong document
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createEmpty();
+$dom->saveHTML(DOM\HTMLDocument::createEmpty());
+
+?>
+--EXPECTF--
+Fatal error: Uncaught DOMException: Wrong Document Error in %s:%d
+Stack trace:
+#0 %s(%d): DOM\HTMLDocument->saveHTML(Object(DOM\HTMLDocument))
+#1 {main}
+  thrown in %s on line %d

ext/dom/tests/modern/html/interactions/HTMLDocument_should_retain_properties_and_owner_01.phpt

@@ -0,0 +1,78 @@
+--TEST--
+DOM\HTMLDocument should retain properties and ownerDocument relation 01
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromString("<p>foo</p>", LIBXML_NOERROR);
+$dom->strictErrorChecking = false;
+
+// Destroy reference to the DOM
+$child = $dom->documentElement;
+unset($dom);
+
+// Regain reference using the ownerDocument property
+// Should be a DOM\HTML5Document
+$dom = $child->ownerDocument;
+var_dump($dom);
+// Test if property is preserved (any random doc_props property will do)
+var_dump($dom->strictErrorChecking);
+
+?>
+--EXPECT--
+object(DOM\HTMLDocument)#1 (26) {
+  ["encoding"]=>
+  string(5) "UTF-8"
+  ["doctype"]=>
+  NULL
+  ["documentElement"]=>
+  string(22) "(object value omitted)"
+  ["strictErrorChecking"]=>
+  bool(false)
+  ["documentURI"]=>
+  NULL
+  ["firstElementChild"]=>
+  string(22) "(object value omitted)"
+  ["lastElementChild"]=>
+  string(22) "(object value omitted)"
+  ["childElementCount"]=>
+  int(1)
+  ["nodeName"]=>
+  string(9) "#document"
+  ["nodeValue"]=>
+  NULL
+  ["nodeType"]=>
+  int(13)
+  ["parentNode"]=>
+  NULL
+  ["parentElement"]=>
+  NULL
+  ["childNodes"]=>
+  string(22) "(object value omitted)"
+  ["firstChild"]=>
+  string(22) "(object value omitted)"
+  ["lastChild"]=>
+  string(22) "(object value omitted)"
+  ["previousSibling"]=>
+  NULL
+  ["nextSibling"]=>
+  NULL
+  ["attributes"]=>
+  NULL
+  ["isConnected"]=>
+  bool(true)
+  ["ownerDocument"]=>
+  NULL
+  ["namespaceURI"]=>
+  NULL
+  ["prefix"]=>
+  string(0) ""
+  ["localName"]=>
+  NULL
+  ["baseURI"]=>
+  NULL
+  ["textContent"]=>
+  string(3) "foo"
+}
+bool(false)

ext/dom/tests/modern/html/interactions/HTMLDocument_should_retain_properties_and_owner_02.phpt

@@ -0,0 +1,78 @@
+--TEST--
+DOM\HTMLDocument should retain properties and ownerDocument relation 02
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromString("<p>foo</p>", LIBXML_NOERROR);
+$dom->strictErrorChecking = false;
+$child = $dom->appendChild($dom->createElement('html'));
+
+// Destroy reference to the DOM
+unset($dom);
+
+// Regain reference using the ownerDocument property
+// Should be a DOM\HTMLDocument
+$dom = $child->ownerDocument;
+var_dump($dom);
+// Test if property is preserved (any random doc_props property will do)
+var_dump($dom->strictErrorChecking);
+
+?>
+--EXPECT--
+object(DOM\HTMLDocument)#1 (26) {
+  ["encoding"]=>
+  string(5) "UTF-8"
+  ["doctype"]=>
+  NULL
+  ["documentElement"]=>
+  string(22) "(object value omitted)"
+  ["strictErrorChecking"]=>
+  bool(false)
+  ["documentURI"]=>
+  NULL
+  ["firstElementChild"]=>
+  string(22) "(object value omitted)"
+  ["lastElementChild"]=>
+  string(22) "(object value omitted)"
+  ["childElementCount"]=>
+  int(2)
+  ["nodeName"]=>
+  string(9) "#document"
+  ["nodeValue"]=>
+  NULL
+  ["nodeType"]=>
+  int(13)
+  ["parentNode"]=>
+  NULL
+  ["parentElement"]=>
+  NULL
+  ["childNodes"]=>
+  string(22) "(object value omitted)"
+  ["firstChild"]=>
+  string(22) "(object value omitted)"
+  ["lastChild"]=>
+  string(22) "(object value omitted)"
+  ["previousSibling"]=>
+  NULL
+  ["nextSibling"]=>
+  NULL
+  ["attributes"]=>
+  NULL
+  ["isConnected"]=>
+  bool(true)
+  ["ownerDocument"]=>
+  NULL
+  ["namespaceURI"]=>
+  NULL
+  ["prefix"]=>
+  string(0) ""
+  ["localName"]=>
+  NULL
+  ["baseURI"]=>
+  NULL
+  ["textContent"]=>
+  string(3) "foo"
+}
+bool(false)

ext/dom/tests/modern/html/interactions/getLineNo_65536.phpt

@@ -0,0 +1,20 @@
+--TEST--
+getLineNo() returns the line number of the node >= 65536
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$html = str_repeat("\n", 65536) . <<<EOF
+<!doctype html>
+<body>
+    <p>hello</p>
+</body>
+EOF;
+
+$dom = DOM\HTMLDocument::createFromString($html);
+var_dump($dom->documentElement->firstChild->nextSibling->firstChild->nextSibling->getLineNo());
+
+?>
+--EXPECT--
+int(65538)

ext/dom/tests/modern/html/interactions/noscript.phpt

@@ -0,0 +1,46 @@
+--TEST--
+noscript behaviour
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+echo "--- Parsing ---\n";
+
+$dom = DOM\HTMLDocument::createFromString("<!doctype html><html><body><noscript><p>hi</p></noscript></body></html>", DOM\HTML_NO_DEFAULT_NS);
+var_dump($dom->documentElement->textContent);
+echo $dom->saveHTML(), "\n";
+echo $dom->saveXML();
+
+echo "--- Modifying the text content: tag ---\n";
+
+$xpath = new DOMXPath($dom);
+$noscript = $xpath->query("//noscript")[0];
+$noscript->textContent = "<p>bye</p>";
+echo $dom->saveHTML(), "\n";
+echo $dom->saveXML();
+
+echo "--- Modifying the text content: trick ---\n";
+
+$noscript->textContent = "<!-- </noscript> -->";
+echo $dom->saveHTML(), "\n";
+echo $dom->saveXML();
+
+?>
+--EXPECT--
+--- Parsing ---
+string(2) "hi"
+<!DOCTYPE html><html><head></head><body><noscript><p>hi</p></noscript></body></html>
+<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+<!DOCTYPE html>
+<html><head/><body><noscript><p>hi</p></noscript></body></html>
+--- Modifying the text content: tag ---
+<!DOCTYPE html><html><head></head><body><noscript>&lt;p&gt;bye&lt;/p&gt;</noscript></body></html>
+<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+<!DOCTYPE html>
+<html><head/><body><noscript>&lt;p&gt;bye&lt;/p&gt;</noscript></body></html>
+--- Modifying the text content: trick ---
+<!DOCTYPE html><html><head></head><body><noscript>&lt;!-- &lt;/noscript&gt; --&gt;</noscript></body></html>
+<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+<!DOCTYPE html>
+<html><head/><body><noscript>&lt;!-- &lt;/noscript&gt; --&gt;</noscript></body></html>

ext/dom/tests/modern/html/interactions/test foo.html

@@ -0,0 +1 @@
+foo

ext/dom/tests/modern/html/interactions/without_constructor.phpt

@@ -0,0 +1,20 @@
+--TEST--
+Tests without running the constructor
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+foreach (['DOM\HTMLDocument', 'DOM\XMLDocument'] as $class) {
+    try {
+        $rc = new ReflectionClass($class);
+        $rc->newInstanceWithoutConstructor();
+    } catch (ReflectionException $e) {
+        echo $e->getMessage(), "\n";
+    }
+}
+
+?>
+--EXPECT--
+Class DOM\HTMLDocument is an internal class marked as final that cannot be instantiated without invoking its constructor
+Class DOM\XMLDocument is an internal class marked as final that cannot be instantiated without invoking its constructor

ext/dom/tests/modern/html/parser/HTMLDocument_createFromFile_createFromString_BOM_buffer_edge.phpt

@@ -0,0 +1,26 @@
+--TEST--
+DOM\HTMLDocument::createFromFile()/createFromString() BOM with a buffer on the edge
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$header = "\xEF\xBB\xBF<!doctype html><html><body>";
+$trailer = "</body></html>";
+$data = $header . str_repeat("a", 4096 - strlen($header) - strlen($trailer)) . $trailer;
+
+$dom = DOM\HTMLDocument::createFromString($header . str_repeat("a", 4096 - strlen($header) - strlen($trailer)) . $trailer);
+var_dump($dom->textContent);
+
+file_put_contents(__DIR__ . "/BOM_edge.tmp", $data);
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . "/BOM_edge.tmp");
+var_dump($dom->textContent);
+
+?>
+--CLEAN--
+<?php
+@unlink(__DIR__ . "/BOM_edge.tmp");
+?>
+--EXPECT--
+string(4052) "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa"
+string(4052) "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa"

ext/dom/tests/modern/html/parser/HTMLDocument_fromFile_DOM_HTML_NO_DEFAULT_NS.phpt

@@ -0,0 +1,37 @@
+--TEST--
+DOM\HTMLDocument::createFromFile() with DOM\HTML_NO_DEFAULT_NS
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . "/paragraph.html", LIBXML_HTML_NOIMPLIED | LIBXML_NOERROR);
+$xpath = new DOMXPath($dom);
+$xpath->registerNamespace("x", "http://www.w3.org/1999/xhtml");
+var_dump($xpath->query("//p"));
+var_dump($xpath->query("//x:p"));
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . "/paragraph.html", LIBXML_HTML_NOIMPLIED | LIBXML_NOERROR | DOM\HTML_NO_DEFAULT_NS);
+$xpath = new DOMXPath($dom);
+$xpath->registerNamespace("x", "http://www.w3.org/1999/xhtml");
+var_dump($xpath->query("//p"));
+var_dump($xpath->query("//x:p"));
+
+?>
+--EXPECT--
+object(DOMNodeList)#3 (1) {
+  ["length"]=>
+  int(0)
+}
+object(DOMNodeList)#4 (1) {
+  ["length"]=>
+  int(1)
+}
+object(DOMNodeList)#3 (1) {
+  ["length"]=>
+  int(1)
+}
+object(DOMNodeList)#3 (1) {
+  ["length"]=>
+  int(0)
+}

ext/dom/tests/modern/html/parser/HTMLDocument_fromFile_empty_path.phpt

@@ -0,0 +1,16 @@
+--TEST--
+DOM\HTMLDocument::createFromFile() - empty path
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile("");
+
+?>
+--EXPECTF--
+Fatal error: Uncaught ValueError: Path cannot be empty in %s:%d
+Stack trace:
+#0 %s(%d): DOM\HTMLDocument::createFromFile('')
+#1 {main}
+  thrown in %s on line %d

ext/dom/tests/modern/html/parser/HTMLDocument_fromFile_local_existing_file.phpt

@@ -0,0 +1,22 @@
+--TEST--
+DOM\HTMLDocument::createFromFile() - local existing file
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . "/../../../test.html");
+echo $dom->saveHTML(), "\n";
+
+?>
+--EXPECTF--
+Warning: DOM\HTMLDocument::createFromFile(): tree error unexpected-token-in-initial-mode in %s on line %d
+<html><head>
+<title>Hello world</title>
+</head>
+<body>
+This is a not well-formed<br>
+html files with undeclared entities&nbsp;
+
+
+</body></html>

ext/dom/tests/modern/html/parser/HTMLDocument_fromFile_local_file_does_not_exist.phpt

@@ -0,0 +1,19 @@
+--TEST--
+DOM\HTMLDocument::createFromFile() - local file that does not exist
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__ . "/../foobar");
+echo $dom->saveHTML(), "\n";
+
+?>
+--EXPECTF--
+Warning: DOM\HTMLDocument::createFromFile(%s): Failed to open stream: No such file or directory in %s on line %d
+
+Fatal error: Uncaught Exception: Cannot open file '%s' in %s:%d
+Stack trace:
+#0 %s(%d): DOM\HTMLDocument::createFromFile('%s')
+#1 {main}
+  thrown in %s on line %d

ext/dom/tests/modern/html/parser/HTMLDocument_fromFile_nul_terminator_cases_path.phpt

@@ -0,0 +1,22 @@
+--TEST--
+DOM\HTMLDocument::createFromFile() - NUL terminator cases path
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+try {
+    DOM\HTMLDocument::createFromFile("\0");
+} catch (ValueError $e) {
+    echo $e->getMessage(), "\n";
+}
+try {
+    DOM\HTMLDocument::createFromFile('%00');
+} catch (ValueError $e) {
+    echo $e->getMessage(), "\n";
+}
+
+?>
+--EXPECT--
+DOM\HTMLDocument::createFromFile(): Argument #1 ($path) must not contain any null bytes
+DOM\HTMLDocument::createFromFile(): Argument #1 ($path) must not contain percent-encoded NUL bytes

ext/dom/tests/modern/html/parser/HTMLDocument_fromFile_parser_warning_01.phpt

@@ -0,0 +1,19 @@
+--TEST--
+DOM\HTMLDocument::createFromFile() - parser warning 01
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__."/parser_warning_01.html", LIBXML_NOERROR);
+echo $dom->saveHTML(), "\n";
+
+?>
+--EXPECT--
+<html><head><title>foo</title>
+</head><body><datalist id="fruits">
+    <option value="Apple">
+    </option><option value="Banana">
+    </option><option value="Pear">
+</option></datalist>
+<p>error</p></body></html>

ext/dom/tests/modern/html/parser/HTMLDocument_fromFile_parser_warning_02.phpt

@@ -0,0 +1,20 @@
+--TEST--
+DOM\HTMLDocument::createFromFile() - parser warning 02
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__."/parser_warning_02.html", LIBXML_NOERROR);
+echo $dom->saveHTML(), "\n";
+
+?>
+--EXPECT--
+<!DOCTYPE html><html><head></head><body>
+        <p<>>foo<p></p>
+        <p id="foo" class="bar">
+        <!--ELEMENT br EMPTY-->
+        <!-- <!-- nested --> --&gt;
+        
+    
+</p></p<>></body></html>

ext/dom/tests/modern/html/parser/HTMLDocument_fromFile_parser_warning_03.phpt

@@ -0,0 +1,16 @@
+--TEST--
+DOM\HTMLDocument::createFromFile() - parser warning 03
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromFile(__DIR__."/parser_warning_03.html", LIBXML_NOERROR);
+echo $dom->saveHTML(), "\n";
+
+?>
+--EXPECT--
+<!DOCTYPE html><html><head></head><body>
+        <p>
+    
+</p></body></html>

ext/dom/tests/modern/html/parser/HTMLDocument_fromFile_parser_warning_libxml_get_last_error.phpt

@@ -0,0 +1,30 @@
+--TEST--
+DOM\HTMLDocument::createFromFile() - parser warning libxml_get_last_error()
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+libxml_use_internal_errors(true);
+
+$html = '<>x</> <!doctype html>';
+$dom = DOM\HTMLDocument::createFromFile(__DIR__."/parser_warning_01.html");
+
+var_dump(libxml_get_last_error());
+
+?>
+--EXPECTF--
+object(LibXMLError)#2 (6) {
+  ["level"]=>
+  int(2)
+  ["code"]=>
+  int(1)
+  ["column"]=>
+  int(2)
+  ["message"]=>
+  string(%d) "tree error unexpected-token-in-initial-mode in %sparser_warning_01.html, line: 1, column: 2-6"
+  ["file"]=>
+  string(%d) "%sparser_warning_01.html"
+  ["line"]=>
+  int(1)
+}

ext/dom/tests/modern/html/parser/HTMLDocument_fromFile_with_failing_stream_wrapper.phpt

@@ -0,0 +1,44 @@
+--TEST--
+DOM\HTMLDocument::createFromFile() with failing stream wrapper
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+class FailingWrapper {
+    public $context;
+    public $fail = false;
+
+    public function stream_open($path, $mode, $options, &$opened_path) {
+        return true;
+    }
+
+    public function stream_read($count) {
+        if ($this->fail) {
+            throw new Error("fail");
+        }
+        $this->fail = true;
+        return str_repeat("X", $count);
+    }
+
+    public function stream_eof() {
+        return false;
+    }
+
+    public function stream_close() {
+        return true;
+    }
+}
+
+stream_wrapper_register("fail", FailingWrapper::class, 0);
+
+DOM\HTMLDocument::createFromFile("fail://x");
+
+?>
+--EXPECTF--
+Fatal error: Uncaught Error: fail in %s:%d
+Stack trace:
+#0 [internal function]: FailingWrapper->stream_read(8192)
+#1 %s(%d): DOM\HTMLDocument::createFromFile('fail://x')
+#2 {main}
+  thrown in %s on line %d

ext/dom/tests/modern/html/parser/HTMLDocument_fromFile_with_working_stream_wrapper.phpt

@@ -0,0 +1,60 @@
+--TEST--
+DOM\HTMLDocument::createFromFile() with working stream wrapper
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+class EchoUriWrapper {
+    private int $position = 0;
+    private string $data;
+    public $context;
+
+    public function stream_open($path, $mode, $options, &$opened_path) {
+        $this->data = substr($path, 6);
+        return true;
+    }
+
+    public function stream_read($count) {
+        $ret = substr($this->data, $this->position, $count);
+        $this->position += $count;
+        return $ret;
+    }
+
+    public function stream_eof() {
+        return $this->position >= strlen($this->data);
+    }
+
+    public function stream_close() {
+        return true;
+    }
+}
+
+stream_wrapper_register("euw", EchoUriWrapper::class, 0);
+
+echo "--- Stream wrapper case ---\n";
+
+$dom = DOM\HTMLDocument::createFromFile("euw://<p>hello</p>");
+echo $dom->saveHTML(), "\n";
+
+echo "--- Stream wrapper in two chunks case ---\n";
+
+libxml_use_internal_errors(true);
+// To properly test this, keep the 4096 in sync with document.c's input stream buffer size.
+$dom = DOM\HTMLDocument::createFromFile("euw://<!doctype html><html>" . str_repeat("\n", 4096-22) . "<></html>");
+echo $dom->saveHTML(), "\n";
+
+foreach (libxml_get_errors() as $error) {
+    var_dump($error->line, $error->column);
+}
+
+?>
+--EXPECTF--
+--- Stream wrapper case ---
+
+Warning: DOM\HTMLDocument::createFromFile(): tree error unexpected-token-in-initial-mode in euw://<p>hello</p>, line: 1, column: 2 in %s on line %d
+<html><head></head><body><p>hello</p></body></html>
+--- Stream wrapper in two chunks case ---
+<!DOCTYPE html><html><head></head><body>&lt;&gt;</body></html>
+int(4075)
+int(2)

ext/dom/tests/modern/html/parser/HTMLDocument_fromString_DOM_HTML_NO_DEFAULT_NS.phpt

@@ -0,0 +1,37 @@
+--TEST--
+DOM\HTMLDocument::createFromString() with DOM\HTML_NO_DEFAULT_NS
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromString(file_get_contents(__DIR__ . "/paragraph.html"), LIBXML_HTML_NOIMPLIED | LIBXML_NOERROR);
+$xpath = new DOMXPath($dom);
+$xpath->registerNamespace("x", "http://www.w3.org/1999/xhtml");
+var_dump($xpath->query("//p"));
+var_dump($xpath->query("//x:p"));
+
+$dom = DOM\HTMLDocument::createFromString(file_get_contents(__DIR__ . "/paragraph.html"), LIBXML_HTML_NOIMPLIED | LIBXML_NOERROR | DOM\HTML_NO_DEFAULT_NS);
+$xpath = new DOMXPath($dom);
+$xpath->registerNamespace("x", "http://www.w3.org/1999/xhtml");
+var_dump($xpath->query("//p"));
+var_dump($xpath->query("//x:p"));
+
+?>
+--EXPECT--
+object(DOMNodeList)#3 (1) {
+  ["length"]=>
+  int(0)
+}
+object(DOMNodeList)#4 (1) {
+  ["length"]=>
+  int(1)
+}
+object(DOMNodeList)#3 (1) {
+  ["length"]=>
+  int(1)
+}
+object(DOMNodeList)#3 (1) {
+  ["length"]=>
+  int(0)
+}

ext/dom/tests/modern/html/parser/HTMLDocument_fromString_LIBXML_COMPACT.phpt

@@ -0,0 +1,40 @@
+--TEST--
+DOM\HTMLDocument::createFromString() with LIBXML_COMPACT
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromString(<<<HTML
+<!DOCTYPE HTML>
+<html>
+<head>
+</head>
+<body>
+    <p> x </p>
+    <p>foo</p>
+    <p>foox</p>
+    <p>fooxx</p>
+    <p>fooxxx</p>
+    <p>fooxxxx</p>
+    <p>fooxxxxx</p>
+    <p>this does not fit</p>
+</body>
+</html>
+HTML, LIBXML_COMPACT);
+
+$xpath = new DOMXPath($dom);
+foreach ($xpath->query("//*[name()='p']") as $p) {
+    echo $p->textContent, "\n";
+}
+
+?>
+--EXPECT--
+x 
+foo
+foox
+fooxx
+fooxxx
+fooxxxx
+fooxxxxx
+this does not fit

ext/dom/tests/modern/html/parser/HTMLDocument_fromString_LIBXML_HTML_NOIMPLIED_error.phpt

@@ -0,0 +1,13 @@
+--TEST--
+DOM\HTMLDocument::createFromString() with LIBXML_HTML_NOIMPLIED - tree error should not happen
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromString("<div>foo</div>", LIBXML_HTML_NOIMPLIED);
+echo $dom->saveHTML();
+
+?>
+--EXPECT--
+<div>foo</div>

ext/dom/tests/modern/html/parser/HTMLDocument_fromString_LIBXML_HTML_NOIMPLIED_namespace.phpt

@@ -0,0 +1,45 @@
+--TEST--
+DOM\HTMLDocument::createFromString() with LIBXML_HTML_NOIMPLIED namespace check
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+echo "--- No elements ---\n";
+
+$dom = DOM\HTMLDocument::createFromString("", LIBXML_HTML_NOIMPLIED | LIBXML_NOERROR);
+echo $dom->saveXML();
+
+echo "--- Single element ---\n";
+
+$dom = DOM\HTMLDocument::createFromString("<p>foo</p>", LIBXML_HTML_NOIMPLIED | LIBXML_NOERROR);
+echo $dom->saveXML();
+var_dump($dom->documentElement->namespaceURI);
+var_dump($dom->documentElement->prefix);
+
+echo "--- Multiple elements ---\n";
+
+$dom = DOM\HTMLDocument::createFromString("<p>foo</p><strong>bar</strong>", LIBXML_HTML_NOIMPLIED | LIBXML_NOERROR);
+echo $dom->saveXML();
+var_dump($dom->documentElement->namespaceURI);
+var_dump($dom->documentElement->prefix);
+var_dump($dom->documentElement->nextSibling->namespaceURI);
+var_dump($dom->documentElement->nextSibling->prefix);
+
+?>
+--EXPECT--
+--- No elements ---
+<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+--- Single element ---
+<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+<p xmlns="http://www.w3.org/1999/xhtml">foo</p>
+string(28) "http://www.w3.org/1999/xhtml"
+string(0) ""
+--- Multiple elements ---
+<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+<p xmlns="http://www.w3.org/1999/xhtml">foo</p>
+<strong xmlns="http://www.w3.org/1999/xhtml">bar</strong>
+string(28) "http://www.w3.org/1999/xhtml"
+string(0) ""
+string(28) "http://www.w3.org/1999/xhtml"
+string(0) ""

ext/dom/tests/modern/html/parser/HTMLDocument_fromString_empty.phpt

@@ -0,0 +1,13 @@
+--TEST--
+DOM\HTMLDocument::createFromString() - empty document
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromString('');
+echo $dom->saveHTML(), "\n";
+
+?>
+--EXPECT--
+<html><head></head><body></body></html>

ext/dom/tests/modern/html/parser/HTMLDocument_fromString_fromFile_LIBXML_HTML_NOIMPLIED.phpt

@@ -0,0 +1,90 @@
+--TEST--
+DOM\HTMLDocument::createFromString()/createFromFile() with LIBXML_HTML_NOIMPLIED
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+function test(string $html) {
+    echo "Testing: $html\n";
+    $dom = DOM\HTMLDocument::createFromString($html, LIBXML_HTML_NOIMPLIED | LIBXML_NOERROR);
+    $output = $dom->saveHTML();
+    echo $output, "\n";
+
+    // Also test the loadHTMLFile variation. We won't print out the result, just checking the result is the same.
+    $temp = fopen(__DIR__."/DOM_HTMLDocument_loadHTML_LIBXML_HTML_NOIMPLIED_input.tmp", "w");
+    fwrite($temp, $html);
+    fclose($temp);
+    $dom = DOM\HTMLDocument::createFromFile(__DIR__."/DOM_HTMLDocument_loadHTML_LIBXML_HTML_NOIMPLIED_input.tmp", LIBXML_HTML_NOIMPLIED | LIBXML_NOERROR);
+    var_dump($output === $dom->saveHTML());
+}
+
+echo "--- Missing html, head, body ---\n";
+test("");
+test("foobarbaz");
+test("<p>foo</p>");
+echo "--- Missing html, head ---\n";
+test("<body><p>foo</p></body>");
+test("<title>x</title><p>foo</p>");
+echo "--- Missing html, body ---\n";
+test("<!-- bar --><head><title>x</title><!-- foo --></head><p>foo</p><!-- bar -->");
+echo "--- Missing html ---\n";
+test("<head><title>x</title></head><body><p>foo</p></body>");
+echo "--- Missing head, body ---\n";
+test("<html>foobar</html>");
+test("<html><!-- foo --><title>a</title><p>foo</p></html><!-- bar -->");
+echo "--- Missing head ---\n";
+test("<html><!-- foo --><body>hi</body></html>");
+echo "--- Missing nothing ---\n";
+test("<html><head><title>x</title></head><!-- foo --><body><p>foo</p></body></html>");
+echo "--- Malformed document ---\n";
+test("<!-- start --><body><head><html>foo</html></head></body><!-- end -->");
+?>
+--CLEAN--
+<?php
+@unlink(__DIR__."/DOM_HTMLDocument_loadHTML_LIBXML_HTML_NOIMPLIED_input.tmp");
+?>
+--EXPECT--
+--- Missing html, head, body ---
+Testing: 
+
+bool(true)
+Testing: foobarbaz
+foobarbaz
+bool(true)
+Testing: <p>foo</p>
+<p>foo</p>
+bool(true)
+--- Missing html, head ---
+Testing: <body><p>foo</p></body>
+<body><p>foo</p></body>
+bool(true)
+Testing: <title>x</title><p>foo</p>
+<title>x</title><p>foo</p>
+bool(true)
+--- Missing html, body ---
+Testing: <!-- bar --><head><title>x</title><!-- foo --></head><p>foo</p><!-- bar -->
+<!-- bar --><head><title>x</title><!-- foo --></head><p>foo</p><!-- bar -->
+bool(true)
+--- Missing html ---
+Testing: <head><title>x</title></head><body><p>foo</p></body>
+<head><title>x</title></head><body><p>foo</p></body>
+bool(true)
+--- Missing head, body ---
+Testing: <html>foobar</html>
+<html>foobar</html>
+bool(true)
+Testing: <html><!-- foo --><title>a</title><p>foo</p></html><!-- bar -->
+<html><!-- foo --><title>a</title><p>foo</p></html><!-- bar -->
+bool(true)
+--- Missing head ---
+Testing: <html><!-- foo --><body>hi</body></html>
+<html><!-- foo --><body>hi</body></html>
+bool(true)
+--- Missing nothing ---
+Testing: <html><head><title>x</title></head><!-- foo --><body><p>foo</p></body></html>
+<html><head><title>x</title></head><!-- foo --><body><p>foo</p></body></html>
+bool(true)
+--- Malformed document ---
+Testing: <!-- start --><body><head><html>foo</html></head></body><!-- end -->
+<!-- start --><body>foo</body><!-- end -->
+bool(true)

ext/dom/tests/modern/html/parser/HTMLDocument_fromString_line_column.phpt

@@ -0,0 +1,57 @@
+--TEST--
+DOM\HTMLDocument::createFromString() - line and column test
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromString(<<<HTML
+<!doctype html>
+<html>
+    <head>
+        <title>foo</title>
+    </head>
+    <body>
+        <div id="mydiv" x="foo">
+            <p>
+                <strong>This is my paragraph</strong>
+                <!-- my comment -->
+            </p>
+        </div>
+    </body>
+</html>
+HTML);
+
+$xpath = new DOMXPath($dom);
+
+foreach ($xpath->query("//*") as $element) {
+    echo "Element: '", $element->tagName, "', ", $element->getLineNo(), "\n";
+}
+
+foreach ($xpath->query("//*[name()='strong']") as $element) {
+    echo "Text: '", $element->textContent, "', ", $element->firstChild->getLineNo(), "\n";
+}
+
+foreach ($xpath->query("//*[name()='div']") as $element) {
+    foreach ($element->attributes as $attribute) {
+        echo "Attribute: '", $attribute->nodeName, "', ", $attribute->getLineNo(), "\n";
+    }
+}
+
+foreach ($xpath->query("//comment()") as $comment) {
+    echo "Comment: '", $comment->data, "', ", $comment->getLineNo(), "\n";
+}
+
+?>
+--EXPECT--
+Element: 'html', 1
+Element: 'head', 2
+Element: 'title', 3
+Element: 'body', 5
+Element: 'div', 6
+Element: 'p', 7
+Element: 'strong', 8
+Text: 'This is my paragraph', 8
+Attribute: 'id', 6
+Attribute: 'x', 6
+Comment: ' my comment ', 9

ext/dom/tests/modern/html/parser/HTMLDocument_fromString_normal_no_error.phpt

@@ -0,0 +1,39 @@
+--TEST--
+DOM\HTMLDocument::createFromString() - normal document, no error
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+// The closing p tag breaks libxml2's HTML parser, but doesn't break the HTML5 parser due to the script context parsing rules.
+$html = <<<HTML
+<!DOCTYPE HTML>
+<html>
+    <head>
+        <meta charset="utf-8">
+        <title>foo</title>
+    </head>
+    <body>
+        <script>
+        var foo = "</p>";
+        </script>
+        <p test="<script>">bar <!-- hi --></p>
+    </body>
+</html>
+HTML;
+$dom = DOM\HTMLDocument::createFromString($html);
+echo $dom->saveHTML(), "\n";
+
+?>
+--EXPECT--
+<!DOCTYPE html><html><head>
+        <meta charset="utf-8">
+        <title>foo</title>
+    </head>
+    <body>
+        <script>
+        var foo = "</p>";
+        </script>
+        <p test="<script>">bar <!-- hi --></p>
+    
+</body></html>

ext/dom/tests/modern/html/parser/HTMLDocument_fromString_old_dtd.phpt

@@ -0,0 +1,39 @@
+--TEST--
+DOM\HTMLDocument::createFromString(): Old DTD
+--EXTENSIONS--
+dom
+--FILE--
+<?php
+
+$dom = DOM\HTMLDocument::createFromString(<<<HTML
+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<html>
+<head>
+</head>
+<body>
+</body>
+</html>
+HTML);
+
+echo "--- HTML serialization ---\n";
+echo $dom->saveHTML(), "\n";
+echo "--- XML serialization ---\n";
+echo $dom->saveXML();
+
+?>
+--EXPECTF--
+Warning: DOM\HTMLDocument::createFromString(): tree error bad-doctype-token-in-initial-mode in Entity, line: 1, column: 3-9 in %s on line %d
+--- HTML serialization ---
+<!DOCTYPE html><html><head>
+</head>
+<body>
+
+</body></html>
+--- XML serialization ---
+<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
+<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml"><head>
+</head>
+<body>
+
+</body></html>