[WIP] Implement the URL Pattern WHATWG specification

.github/workflows/website-build.yml

@@ -30,6 +30,7 @@ jobs:
           -DSOURCEMETA_CORE_JSONSCHEMA:BOOL=OFF
           -DSOURCEMETA_CORE_JSONPOINTER:BOOL=OFF
           -DSOURCEMETA_CORE_YAML:BOOL=OFF
+          -DSOURCEMETA_CORE_URLPATTERN:BOOL=OFF
           -DSOURCEMETA_CORE_EXTENSION_ALTERSCHEMA:BOOL=OFF
           -DSOURCEMETA_CORE_EXTENSION_EDITORSCHEMA:BOOL=OFF
           -DSOURCEMETA_CORE_EXTENSION_SCHEMACONFIG:BOOL=OFF

.github/workflows/website-deploy.yml

@@ -40,6 +40,7 @@ jobs:
           -DSOURCEMETA_CORE_JSONSCHEMA:BOOL=OFF
           -DSOURCEMETA_CORE_JSONPOINTER:BOOL=OFF
           -DSOURCEMETA_CORE_YAML:BOOL=OFF
+          -DSOURCEMETA_CORE_URLPATTERN:BOOL=OFF
           -DSOURCEMETA_CORE_EXTENSION_ALTERSCHEMA:BOOL=OFF
           -DSOURCEMETA_CORE_EXTENSION_EDITORSCHEMA:BOOL=OFF
           -DSOURCEMETA_CORE_EXTENSION_SCHEMACONFIG:BOOL=OFF

CMakeLists.txt

@@ -18,6 +18,7 @@ option(SOURCEMETA_CORE_JSONSCHEMA "Build the Sourcemeta Core JSON Schema library
 option(SOURCEMETA_CORE_JSONPOINTER "Build the Sourcemeta Core JSON Pointer library" ON)
 option(SOURCEMETA_CORE_JSONL "Build the Sourcemeta Core JSONL library" ON)
 option(SOURCEMETA_CORE_YAML "Build the Sourcemeta Core YAML library" ON)
+option(SOURCEMETA_CORE_URLPATTERN "Build the Sourcemeta Core URL Pattern library" ON)
 option(SOURCEMETA_CORE_EXTENSION_ALTERSCHEMA "Build the Sourcemeta Core AlterSchema library" ON)
 option(SOURCEMETA_CORE_EXTENSION_EDITORSCHEMA "Build the Sourcemeta Core EditorSchema library" ON)
 option(SOURCEMETA_CORE_EXTENSION_SCHEMACONFIG "Build the Sourcemeta Core SchemaConfig library" ON)
@@ -121,6 +122,10 @@ if(SOURCEMETA_CORE_YAML)
   add_subdirectory(src/core/yaml)
 endif()
 
+if(SOURCEMETA_CORE_URLPATTERN)
+  add_subdirectory(src/core/urlpattern)
+endif()
+
 if(SOURCEMETA_CORE_EXTENSION_ALTERSCHEMA)
   add_subdirectory(src/extension/alterschema)
 endif()
@@ -232,6 +237,10 @@ if(SOURCEMETA_CORE_TESTS)
     add_subdirectory(test/yaml)
   endif()
 
+  if(SOURCEMETA_CORE_URLPATTERN)
+    add_subdirectory(test/urlpattern)
+  endif()
+
   if(SOURCEMETA_CORE_EXTENSION_ALTERSCHEMA)
     add_subdirectory(test/alterschema)
   endif()

DEPENDENCIES

@@ -15,3 +15,4 @@ yaml https://github.com/yaml/libyaml 0.2.5
 pcre2 https://github.com/PCRE2Project/pcre2 pcre2-10.47
 googletest https://github.com/google/googletest a7f443b80b105f940225332ed3c31f2790092f47
 googlebenchmark https://github.com/google/benchmark 378fe693a1ef51500db21b11ff05a8018c5f0e55
+wpt https://github.com/web-platform-tests/wpt 71154f49e6cdda82c43e11bb74cfe3ad7b3f9368

config.cmake.in

@@ -19,6 +19,7 @@ if(NOT SOURCEMETA_CORE_COMPONENTS)
   list(APPEND SOURCEMETA_CORE_COMPONENTS jsonpointer)
   list(APPEND SOURCEMETA_CORE_COMPONENTS jsonschema)
   list(APPEND SOURCEMETA_CORE_COMPONENTS yaml)
+  list(APPEND SOURCEMETA_CORE_COMPONENTS urlpattern)
   list(APPEND SOURCEMETA_CORE_COMPONENTS alterschema)
   list(APPEND SOURCEMETA_CORE_COMPONENTS editorschema)
   list(APPEND SOURCEMETA_CORE_COMPONENTS schemaconfig)
@@ -87,6 +88,11 @@ foreach(component ${SOURCEMETA_CORE_COMPONENTS})
     find_dependency(yaml CONFIG)
     include("${CMAKE_CURRENT_LIST_DIR}/sourcemeta_core_io.cmake")
     include("${CMAKE_CURRENT_LIST_DIR}/sourcemeta_core_yaml.cmake")
+  elseif(component STREQUAL "urlpattern")
+    include("${CMAKE_CURRENT_LIST_DIR}/sourcemeta_core_json.cmake")
+    include("${CMAKE_CURRENT_LIST_DIR}/sourcemeta_core_regex.cmake")
+    include("${CMAKE_CURRENT_LIST_DIR}/sourcemeta_core_punycode.cmake")
+    include("${CMAKE_CURRENT_LIST_DIR}/sourcemeta_core_urlpattern.cmake")
   elseif(component STREQUAL "alterschema")
     include("${CMAKE_CURRENT_LIST_DIR}/sourcemeta_core_uri.cmake")
     find_dependency(mpdecimal CONFIG)

patches/wpt/0001-Remove-invalid-surrogate-pairs-tests.patch

@@ -0,0 +1,44 @@
+From ec3c41d5eb68cf24e86ef67b1067fc43df859ab0 Mon Sep 17 00:00:00 2001
+From: Juan Cruz Viotti <jv@jviotti.com>
+Date: Fri, 28 Nov 2025 15:42:16 -0400
+Subject: [PATCH] Remove invalid surrogate pairs tests
+
+Signed-off-by: Juan Cruz Viotti <jv@jviotti.com>
+---
+ urlpattern/resources/urlpatterntestdata.json | 20 --------------------
+ 1 file changed, 20 deletions(-)
+
+diff --git a/urlpattern/resources/urlpatterntestdata.json b/urlpattern/resources/urlpatterntestdata.json
+index 4b1b9ee5f6..ae10c412ad 100644
+--- a/urlpattern/resources/urlpatterntestdata.json
++++ b/urlpattern/resources/urlpatterntestdata.json
+@@ -1136,26 +1136,6 @@
+       "pathname": { "input": "/", "groups": {}}
+     }
+   },
+-  {
+-    "pattern": ["http://\uD83D \uDEB2"],
+-    "expected_obj": "error"
+-  },
+-  {
+-    "pattern": [{"hostname":"\uD83D \uDEB2"}],
+-    "expected_obj": "error"
+-  },
+-  {
+-    "pattern": [{"pathname":"\uD83D \uDEB2"}],
+-    "inputs": [],
+-    "expected_obj": {
+-      "pathname": "%EF%BF%BD%20%EF%BF%BD"
+-    },
+-    "expected_match": null
+-  },
+-  {
+-    "pattern": [{"pathname":":\uD83D \uDEB2"}],
+-    "expected_obj": "error"
+-  },
+   {
+     "pattern": [{"pathname":":a\uDB40\uDD00b"}],
+     "inputs": [],
+-- 
+2.52.0
+

src/core/urlpattern/CMakeLists.txt

@@ -0,0 +1,17 @@
+sourcemeta_library(NAMESPACE sourcemeta PROJECT core NAME urlpattern
+  PRIVATE_HEADERS error.h part.h component.h
+  SOURCES urlpattern.cc urlpattern_part.cc urlpattern_component.cc)
+
+if(SOURCEMETA_CORE_INSTALL)
+  sourcemeta_library_install(NAMESPACE sourcemeta PROJECT core NAME urlpattern)
+endif()
+
+target_link_libraries(sourcemeta_core_urlpattern PUBLIC
+  sourcemeta::core::json)
+target_link_libraries(sourcemeta_core_urlpattern PUBLIC
+  sourcemeta::core::regex)
+target_link_libraries(sourcemeta_core_urlpattern PRIVATE
+  sourcemeta::core::punycode)
+
+# TODO: Find a way to get rid of this?
+target_link_libraries(sourcemeta_core_urlpattern PRIVATE PCRE2::pcre2)

src/core/urlpattern/include/sourcemeta/core/urlpattern.h

@@ -0,0 +1,86 @@
+#ifndef SOURCEMETA_CORE_URLPATTERN_H_
+#define SOURCEMETA_CORE_URLPATTERN_H_
+
+#ifndef SOURCEMETA_CORE_URLPATTERN_EXPORT
+#include <sourcemeta/core/urlpattern_export.h>
+#endif
+
+// NOLINTBEGIN(misc-include-cleaner)
+#include <sourcemeta/core/urlpattern_component.h>
+#include <sourcemeta/core/urlpattern_error.h>
+#include <sourcemeta/core/urlpattern_part.h>
+// NOLINTEND(misc-include-cleaner)
+
+#include <sourcemeta/core/json.h>
+
+#include <compare>     // std::strong_ordering
+#include <optional>    // std::optional
+#include <string>      // std::string
+#include <string_view> // std::string_view
+
+/// @defgroup urlpattern URL Pattern
+/// @brief A WHATWG URL Pattern implementation.
+///
+/// This functionality is included as follows:
+///
+/// ```cpp
+/// #include <sourcemeta/core/urlpattern.h>
+/// ```
+
+namespace sourcemeta::core {
+
+/// @ingroup urlpattern
+struct URLPatternResult {
+  std::optional<URLPatternComponentResult> protocol;
+  std::optional<URLPatternComponentResult> username;
+  std::optional<URLPatternComponentResult> password;
+  std::optional<URLPatternComponentResult> hostname;
+  std::optional<URLPatternComponentResult> port;
+  std::optional<URLPatternComponentResult> pathname;
+  std::optional<URLPatternComponentResult> search;
+  std::optional<URLPatternComponentResult> hash;
+};
+
+/// @ingroup urlpattern
+struct SOURCEMETA_CORE_URLPATTERN_EXPORT URLPatternInput {
+  std::string protocol;
+  std::string username;
+  std::string password;
+  std::string hostname;
+  std::string port;
+  std::string pathname;
+  std::string search;
+  std::string hash;
+
+  [[nodiscard]] static auto parse(const JSON &input)
+      -> std::optional<URLPatternInput>;
+  [[nodiscard]] static auto parse(const std::string_view input)
+      -> std::optional<URLPatternInput>;
+};
+
+/// @ingroup urlpattern
+/// See https://urlpattern.spec.whatwg.org/#url-pattern-struct
+struct SOURCEMETA_CORE_URLPATTERN_EXPORT URLPattern {
+  URLPatternProtocol protocol;
+  URLPatternUsername username;
+  URLPatternPassword password;
+  URLPatternHostname hostname;
+  URLPatternPort port;
+  URLPatternPathname pathname;
+  URLPatternSearch search;
+  URLPatternHash hash;
+
+  auto operator==(const URLPattern &other) const -> bool = default;
+  auto operator<=>(const URLPattern &other) const -> std::strong_ordering;
+
+  [[nodiscard]] static auto parse(const std::string_view input) -> URLPattern;
+  [[nodiscard]] static auto parse(const JSON &input)
+      -> std::optional<URLPattern>;
+
+  [[nodiscard]] auto match(const URLPatternInput &input) const
+      -> URLPatternResult;
+};
+
+} // namespace sourcemeta::core
+
+#endif