Create separate lexbor extension (#18538)

An always enabled lexbor extension is added, containing the lexbor library that was separated from ext/dom extension in preparation of https://wiki.php.net/rfc/url_parsing_api. While at it, the lexbor library is upgraded to 2.5.0.

Co-authored-by: Niels Dossche <7771979+nielsdos@users.noreply.github.com>
Co-authored-by: Gina Peter Banyard <girgias@php.net>

Author: 3 people

.gitattributes

@@ -30,4 +30,4 @@
 /ext/date/tests/ossfuzz*.txt binary
 
 # Vendored libraries
-/ext/dom/lexbor/lexbor linguist-vendored
+/ext/lexbor/lexbor linguist-vendored

.gitignore

@@ -306,6 +306,6 @@ tmp-php.ini
 !/ext/fileinfo/libmagic/config.h
 !/ext/fileinfo/libmagic.patch
 !/ext/fileinfo/magicdata.patch
-!/ext/dom/lexbor/patches/*.patch
+!/ext/lexbor/patches/*.patch
 !/ext/pcre/pcre2lib/config.h
 !/win32/build/Makefile

EXTENSIONS

@@ -195,6 +195,13 @@ PRIMARY MAINTAINER:  Thies C. Arntzen <thies@thieso.net> (1999 - 2002)
 MAINTENANCE:         Maintained
 STATUS:              Working
 -------------------------------------------------------------------------------
+EXTENSION:           lexbor
+PRIMARY MAINTAINER:  Niels Dossche <nielsdos@php.net> (2023 - 2025)
+                     Mate Kocsis <kocsismate@php.net> (2025 - 2025)
+MAINTENANCE:         Maintained
+STATUS:              Working
+SINCE:               8.5
+-------------------------------------------------------------------------------
 EXTENSION:           libxml
 PRIMARY MAINTAINER:  Rob Richards <rrichards@php.net> (2003 - 2009)
                      Christian Stocker <chregu@php.net> (2004 - 2011)

README.REDIST.BINS

@@ -17,7 +17,7 @@
 17. ext/mbstring/ucgendat portions based on the ucgendat.c from the OpenLDAP
 18. avifinfo (ext/standard/libavifinfo) see ext/standard/libavifinfo/LICENSE
 19. xxHash (ext/hash/xxhash)
-20. Lexbor (ext/dom/lexbor/lexbor) see ext/dom/lexbor/LICENSE
+20. Lexbor (ext/lexbor/lexbor) see ext/lexbor/LICENSE
 21. Portions of libcperciva (ext/hash/hash_sha_{ni,sse2}.c) see the header in the source file
 
 3. pcre2lib (ext/pcre)

UPGRADING

@@ -400,6 +400,10 @@ PHP 8.5 UPGRADE NOTES
   . The return type of finfo_close() has been changed to true, rather
     than bool.
 
+- Lexbor:
+  . An always enabled lexbor extension is added. It contains the lexbor
+    library that was separated from ext/dom for being reused among other
+    extensions. The new extension is not directly exposed to userland.
 
 - PCRE:
   . Upgraded to pcre2lib from 10.44 to 10.45.

codecov.yml

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

ext/dom/config.m4

@@ -8,185 +8,6 @@ if test "$PHP_DOM" != "no"; then
   PHP_SETUP_LIBXML([DOM_SHARED_LIBADD], [
     AC_DEFINE([HAVE_DOM], [1],
       [Define to 1 if the PHP extension 'dom' is available.])
-    PHP_LEXBOR_CFLAGS="-I@ext_srcdir@/lexbor -DLEXBOR_STATIC"
-    LEXBOR_DIR="lexbor/lexbor"
-    LEXBOR_SOURCES=m4_normalize(["
-      $LEXBOR_DIR/core/array_obj.c
-      $LEXBOR_DIR/core/array.c
-      $LEXBOR_DIR/core/avl.c
-      $LEXBOR_DIR/core/bst.c
-      $LEXBOR_DIR/core/conv.c
-      $LEXBOR_DIR/core/diyfp.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/css/css.c
-      $LEXBOR_DIR/css/log.c
-      $LEXBOR_DIR/css/parser.c
-      $LEXBOR_DIR/css/selectors/pseudo_state.c
-      $LEXBOR_DIR/css/selectors/pseudo.c
-      $LEXBOR_DIR/css/selectors/selector.c
-      $LEXBOR_DIR/css/selectors/selectors.c
-      $LEXBOR_DIR/css/selectors/state.c
-      $LEXBOR_DIR/css/state.c
-      $LEXBOR_DIR/css/syntax/anb.c
-      $LEXBOR_DIR/css/syntax/parser.c
-      $LEXBOR_DIR/css/syntax/state.c
-      $LEXBOR_DIR/css/syntax/syntax.c
-      $LEXBOR_DIR/css/syntax/token.c
-      $LEXBOR_DIR/css/syntax/tokenizer.c
-      $LEXBOR_DIR/css/syntax/tokenizer/error.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_fragment.c
-      $LEXBOR_DIR/dom/interfaces/document_type.c
-      $LEXBOR_DIR/dom/interfaces/document.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/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/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/d_list_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/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/o_list_element.c
-      $LEXBOR_DIR/html/interfaces/object_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/html/parser.c
-      $LEXBOR_DIR/html/token_attr.c
-      $LEXBOR_DIR/html/token.c
-      $LEXBOR_DIR/html/tokenizer.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.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_noscript.c
-      $LEXBOR_DIR/html/tree/insertion_mode/in_head.c
-      $LEXBOR_DIR/html/tree/insertion_mode/in_row.c
-      $LEXBOR_DIR/html/tree/insertion_mode/in_select_in_table.c
-      $LEXBOR_DIR/html/tree/insertion_mode/in_select.c
-      $LEXBOR_DIR/html/tree/insertion_mode/in_table_body.c
-      $LEXBOR_DIR/html/tree/insertion_mode/in_table_text.c
-      $LEXBOR_DIR/html/tree/insertion_mode/in_table.c
-      $LEXBOR_DIR/html/tree/insertion_mode/in_template.c
-      $LEXBOR_DIR/html/tree/insertion_mode/initial.c
-      $LEXBOR_DIR/html/tree/insertion_mode/text.c
-      $LEXBOR_DIR/html/tree/open_elements.c
-      $LEXBOR_DIR/ns/ns.c
-      $LEXBOR_DIR/ports/posix/lexbor/core/memory.c
-      $LEXBOR_DIR/selectors-adapted/selectors.c
-      $LEXBOR_DIR/tag/tag.c
-    "])
     PHP_NEW_EXTENSION([dom], m4_normalize([
         attr.c
         cdatasection.c
@@ -223,25 +44,13 @@ if test "$PHP_DOM" != "no"; then
         xml_serializer.c
         xpath_callbacks.c
         xpath.c
-        $LEXBOR_SOURCES
+        lexbor/selectors-adapted/selectors.c
       ]),
       [$ext_shared],,
-      [$PHP_LEXBOR_CFLAGS])
+      [])
     PHP_ADD_BUILD_DIR([
       $ext_builddir/parentnode
-      $ext_builddir/$LEXBOR_DIR/core
-      $ext_builddir/$LEXBOR_DIR/css/selectors
-      $ext_builddir/$LEXBOR_DIR/css/syntax/tokenizer
-      $ext_builddir/$LEXBOR_DIR/css/tokenizer
-      $ext_builddir/$LEXBOR_DIR/dom/interfaces
-      $ext_builddir/$LEXBOR_DIR/encoding
-      $ext_builddir/$LEXBOR_DIR/html/interfaces
-      $ext_builddir/$LEXBOR_DIR/html/tokenizer
-      $ext_builddir/$LEXBOR_DIR/html/tree/insertion_mode
-      $ext_builddir/$LEXBOR_DIR/ns
-      $ext_builddir/$LEXBOR_DIR/ports/posix/lexbor/core
-      $ext_builddir/$LEXBOR_DIR/selectors-adapted
-      $ext_builddir/$LEXBOR_DIR/tag
+      $ext_builddir/lexbor/selectors-adapted
     ])
     PHP_SUBST([DOM_SHARED_LIBADD])
     PHP_INSTALL_HEADERS([ext/dom], m4_normalize([
@@ -251,5 +60,6 @@ if test "$PHP_DOM" != "no"; then
       xpath_callbacks.h
     ]))
     PHP_ADD_EXTENSION_DEP(dom, libxml)
+    PHP_ADD_EXTENSION_DEP(dom, lexbor)
   ])
 fi

ext/dom/config.w32

@@ -16,32 +16,17 @@ if (PHP_DOM == "yes") {
 			entityreference.c \
 			token_list.c \
 			notation.c xpath.c dom_iterators.c \
-			namednodemap.c xpath_callbacks.c", null, "-Iext/dom/lexbor");
+			namednodemap.c xpath_callbacks.c", null, "/I ext/lexbor");
+
+		ADD_EXTENSION_DEP('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");
-		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-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 ");
+		ADD_SOURCES("ext/dom/lexbor/selectors-adapted", "selectors.c", "dom");
 
 		AC_DEFINE("HAVE_DOM", 1, "Define to 1 if the PHP extension 'dom' is available.");
 
 		if (!PHP_DOM_SHARED) {
-			ADD_FLAG("CFLAGS_DOM", "/D LIBXML_STATIC ");
+			ADD_FLAG("CFLAGS_DOM", "/D LIBXML_STATIC /D LEXBOR_STATIC ");
 		} else {
 			if (!CHECK_LIB("libxml2.lib", "dom")) {
 				WARNING("dom support can't be enabled, libxml is not found")
@@ -51,7 +36,8 @@ if (PHP_DOM == "yes") {
 			"dom_ce.h " +
 			"namespace_compat.h " +
 			"xml_common.h " +
-			"xpath_callbacks.h "
+			"xpath_callbacks.h " +
+			"lexbor/selectors-adapted/selectors.h "
 		);
 	} else {
 		WARNING("dom support can't be enabled, libxml is not enabled")