[part of RFC] DOM CSS selector support: querySelector(All), matches, closest

UPGRADING.INTERNALS

@@ -247,6 +247,7 @@ PHP 8.4 INTERNALS UPGRADE NOTES
    - Removed the "properties" HashTable field from php_libxml_node_object.
    - Added a way to attached private data to a php_libxml_ref_obj.
    - Added a way to fix a class type onto php_libxml_ref_obj.
+   - Added a way to record quirks mode in php_libxml_ref_obj.
    - Added php_libxml_uses_internal_errors().
    - Added a way to override document handlers (e.g. serialization) with
      php_libxml_document_handlers.

codecov.yml

@@ -1,3 +1,11 @@
 ignore:
-  - "ext/dom/lexbor/lexbor" # bundled library
-  - "ext/pcre/pcre2lib" # bundled library
+  # bundled libraries
+  - "ext/dom/lexbor/lexbor/core"
+  - "ext/dom/lexbor/lexbor/css"
+  - "ext/dom/lexbor/lexbor/dom"
+  - "ext/dom/lexbor/lexbor/encoding"
+  - "ext/dom/lexbor/lexbor/html"
+  - "ext/dom/lexbor/lexbor/ns"
+  - "ext/dom/lexbor/lexbor/ports"
+  - "ext/dom/lexbor/lexbor/tag"
+  - "ext/pcre/pcre2lib"

ext/dom/config.m4

@@ -22,12 +22,14 @@ if test "$PHP_DOM" != "no"; then
       $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/css/state.c $LEXBOR_DIR/css/log.c $LEXBOR_DIR/css/parser.c $LEXBOR_DIR/css/selectors/state.c $LEXBOR_DIR/css/selectors/selectors.c $LEXBOR_DIR/css/selectors/selector.c $LEXBOR_DIR/css/selectors/pseudo_state.c $LEXBOR_DIR/css/selectors/pseudo.c $LEXBOR_DIR/css/syntax/tokenizer/error.c $LEXBOR_DIR/css/syntax/state.c $LEXBOR_DIR/css/syntax/parser.c $LEXBOR_DIR/css/syntax/syntax.c $LEXBOR_DIR/css/syntax/anb.c $LEXBOR_DIR/css/syntax/tokenizer.c $LEXBOR_DIR/css/syntax/token.c $LEXBOR_DIR/css/css.c \
+      $LEXBOR_DIR/selectors-adapted/selectors.c \
       $LEXBOR_DIR/ns/ns.c \
       $LEXBOR_DIR/tag/tag.c"
     PHP_NEW_EXTENSION(dom, [php_dom.c attr.c document.c infra.c \
                             xml_document.c html_document.c xml_serializer.c html5_serializer.c html5_parser.c namespace_compat.c \
-                            domexception.c parentnode.c \
+                            domexception.c \
+                            parentnode/tree.c parentnode/css_selectors.c \
                             processinginstruction.c cdatasection.c \
                             documentfragment.c domimplementation.c \
                             element.c node.c characterdata.c \
@@ -38,14 +40,18 @@ if test "$PHP_DOM" != "no"; then
                             namednodemap.c xpath_callbacks.c \
                             $LEXBOR_SOURCES],
                             $ext_shared,,$PHP_LEXBOR_CFLAGS)
+    PHP_ADD_BUILD_DIR($ext_builddir/parentnode)
     PHP_ADD_BUILD_DIR($ext_builddir/$LEXBOR_DIR/ports/posix/lexbor/core)
     PHP_ADD_BUILD_DIR($ext_builddir/$LEXBOR_DIR/core)
     PHP_ADD_BUILD_DIR($ext_builddir/$LEXBOR_DIR/dom/interfaces)
     PHP_ADD_BUILD_DIR($ext_builddir/$LEXBOR_DIR/html/tree/insertion_mode)
     PHP_ADD_BUILD_DIR($ext_builddir/$LEXBOR_DIR/html/tokenizer)
     PHP_ADD_BUILD_DIR($ext_builddir/$LEXBOR_DIR/html/interfaces)
     PHP_ADD_BUILD_DIR($ext_builddir/$LEXBOR_DIR/encoding)
-    PHP_ADD_BUILD_DIR($ext_builddir/$LEXBOR_DIR/selectors)
+    PHP_ADD_BUILD_DIR($ext_builddir/$LEXBOR_DIR/css/selectors)
+    PHP_ADD_BUILD_DIR($ext_builddir/$LEXBOR_DIR/css/tokenizer)
+    PHP_ADD_BUILD_DIR($ext_builddir/$LEXBOR_DIR/css/syntax/tokenizer)
+    PHP_ADD_BUILD_DIR($ext_builddir/$LEXBOR_DIR/selectors-adapted)
     PHP_ADD_BUILD_DIR($ext_builddir/$LEXBOR_DIR/ns)
     PHP_ADD_BUILD_DIR($ext_builddir/$LEXBOR_DIR/tag)
     PHP_SUBST(DOM_SHARED_LIBADD)

ext/dom/config.w32

@@ -9,14 +9,15 @@ if (PHP_DOM == "yes") {
 	) {
 		EXTENSION("dom", "php_dom.c attr.c document.c infra.c \
 			xml_document.c html_document.c xml_serializer.c html5_serializer.c html5_parser.c namespace_compat.c \
-			domexception.c parentnode.c processinginstruction.c \
+			domexception.c processinginstruction.c \
 			cdatasection.c documentfragment.c domimplementation.c element.c \
 			node.c characterdata.c documenttype.c \
 			entity.c nodelist.c html_collection.c text.c comment.c \
 			entityreference.c \
 			notation.c xpath.c dom_iterators.c \
 			namednodemap.c xpath_callbacks.c", null, "-Iext/dom/lexbor");
 
+		ADD_SOURCES("ext/dom/parentnode", "tree.c css_selectors.c", "dom");
 		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");
@@ -27,7 +28,11 @@ if (PHP_DOM == "yes") {
 		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/selectors-adapted", "selectors.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/css", "state.c log.c parser.c css.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/css/selectors", "state.c selectors.c selector.c pseudo_state.c pseudo.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/css/syntax", "state.c parser.c syntax.c anb.c tokenizer.c token.c", "dom");
+		ADD_SOURCES("ext/dom/lexbor/lexbor/css/syntax/tokenizer", "error.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 ");

ext/dom/domexception.h

@@ -46,4 +46,7 @@ typedef enum {
 	VALIDATION_ERR                 = 16
 } dom_exception_code;
 
+void php_dom_throw_error(dom_exception_code error_code, bool strict_error);
+void php_dom_throw_error_with_message(dom_exception_code error_code, const char *error_message, bool strict_error);
+
 #endif /* DOM_EXCEPTION_H */

ext/dom/element.c

@@ -1752,4 +1752,66 @@ PHP_METHOD(DOMElement, toggleAttribute)
 }
 /* }}} end DOMElement::prepend */
 
+static void php_dom_dispatch_query_selector(INTERNAL_FUNCTION_PARAMETERS, bool all)
+{
+	zend_string *selectors_str;
+
+	ZEND_PARSE_PARAMETERS_START(1, 1)
+		Z_PARAM_STR(selectors_str)
+	ZEND_PARSE_PARAMETERS_END();
+
+	xmlNodePtr thisp;
+	dom_object *intern;
+	zval *id;
+	DOM_GET_THIS_OBJ(thisp, id, xmlNodePtr, intern);
+
+	if (all) {
+		dom_parent_node_query_selector_all(thisp, intern, return_value, selectors_str);
+	} else {
+		dom_parent_node_query_selector(thisp, intern, return_value, selectors_str);
+	}
+}
+
+PHP_METHOD(Dom_Element, querySelector)
+{
+	php_dom_dispatch_query_selector(INTERNAL_FUNCTION_PARAM_PASSTHRU, false);
+}
+
+PHP_METHOD(Dom_Element, querySelectorAll)
+{
+	php_dom_dispatch_query_selector(INTERNAL_FUNCTION_PARAM_PASSTHRU, true);
+}
+
+PHP_METHOD(Dom_Element, matches)
+{
+	zend_string *selectors_str;
+
+	ZEND_PARSE_PARAMETERS_START(1, 1)
+		Z_PARAM_STR(selectors_str)
+	ZEND_PARSE_PARAMETERS_END();
+
+	xmlNodePtr thisp;
+	dom_object *intern;
+	zval *id;
+	DOM_GET_THIS_OBJ(thisp, id, xmlNodePtr, intern);
+
+	dom_element_matches(thisp, intern, return_value, selectors_str);
+}
+
+PHP_METHOD(Dom_Element, closest)
+{
+	zend_string *selectors_str;
+
+	ZEND_PARSE_PARAMETERS_START(1, 1)
+		Z_PARAM_STR(selectors_str)
+	ZEND_PARSE_PARAMETERS_END();
+
+	xmlNodePtr thisp;
+	dom_object *intern;
+	zval *id;
+	DOM_GET_THIS_OBJ(thisp, id, xmlNodePtr, intern);
+
+	dom_element_closest(thisp, intern, return_value, selectors_str);
+}
+
 #endif

ext/dom/html5_parser.c

@@ -24,6 +24,7 @@
 #include "html5_parser.h"
 #include <lexbor/html/parser.h>
 #include <lexbor/html/interfaces/element.h>
+#include <lexbor/dom/dom.h>
 #include <libxml/parserInternals.h>
 #include <libxml/HTMLtree.h>
 #include <Zend/zend.h>
@@ -380,6 +381,7 @@ void lexbor_libxml2_bridge_copy_observations(lxb_html_tree_t *tree, lexbor_libxm
     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;
+    observations->quirks_mode = lxb_dom_interface_document(tree->document)->compat_mode == LXB_DOM_DOCUMENT_CMODE_QUIRKS;
 }
 
 #endif  /* HAVE_LIBXML && HAVE_DOM */

ext/dom/html5_parser.h

@@ -47,6 +47,7 @@ typedef struct _lexbor_libxml2_bridge_extracted_observations {
     bool has_explicit_html_tag;
     bool has_explicit_head_tag;
     bool has_explicit_body_tag;
+    bool quirks_mode;
 } lexbor_libxml2_bridge_extracted_observations;
 
 typedef struct _lexbor_libxml2_bridge_parse_context {

ext/dom/html_document.c

@@ -917,6 +917,7 @@ PHP_METHOD(Dom_HTMLDocument, createFromString)
 		NULL
 	);
 	dom_set_xml_class(intern->document);
+	intern->document->quirks_mode = ctx.observations.quirks_mode;
 	intern->document->private_data = php_dom_libxml_ns_mapper_header(ns_mapper);
 	return;
 
@@ -1137,6 +1138,7 @@ PHP_METHOD(Dom_HTMLDocument, createFromFile)
 		NULL
 	);
 	dom_set_xml_class(intern->document);
+	intern->document->quirks_mode = ctx.observations.quirks_mode;
 	intern->document->private_data = php_dom_libxml_ns_mapper_header(ns_mapper);
 	return;