From 320ffa2819950ef8b79717d480b008ba0c72660b Mon Sep 17 00:00:00 2001 From: Karl Seguin Date: Fri, 29 May 2026 17:58:45 +0800 Subject: [PATCH 1/9] Improve WPT /url/ tests This is a bit all over the place. 1 - Replace libidn2 with rust-idna. It looks like there are different idna profiles, and rust-idna (from the servo project) implements the whatwg one. libidn2 would be too strict in some cases and not strict enough in others. (Gemini says I could use libidn2 for this, but what it suggested didn't work, and I couldn't figure it out myself, and claude insisted it _did not_ have the correct implementation for what we want). 2 - We previously only ran a URL through idna if it wasn't ascii. Turns out we also need to run it if there's a "xn--" (aka, an IDNA ACE prefix) in there. This helps us pass hundreds of WPT cases, and it's pretty cheap. 3 - Implement more of the Area WebAPI. Mostly copied from Anchor. 4 - Add username/password accessor to Anchor/Area 5 - window.open validates the URL (i.e. tries to resolve it and handles the error) 6 - Invalid idna conversion maps to a TypeError 7 - Cleanup closed popups on the next tick (like destroyed pages), rather than at an interval or on shutdown. This one seems unrelated, but some of these tests are opening hundreds (thousands?) of popups and then closing them. --- build.zig | 180 +- build.zig.zon | 4 - src/browser/Page.zig | 15 - src/browser/Runner.zig | 3 +- src/browser/Session.zig | 50 +- src/browser/URL.zig | 110 +- src/browser/js/Caller.zig | 1 + src/browser/tests/url.html | 22 + src/browser/webapi/Window.zig | 15 +- src/browser/webapi/element/html/Anchor.zig | 24 + src/browser/webapi/element/html/Area.zig | 181 ++ src/html5ever/Cargo.lock | 250 ++- src/html5ever/Cargo.toml | 1 + src/html5ever/lib.rs | 9 +- src/html5ever/url.rs | 83 + src/sys/idna.zig | 74 +- vendor/libidn2/config.h | 1915 -------------------- vendor/libidn2/darwin/strchrnul.c | 20 - 18 files changed, 779 insertions(+), 2178 deletions(-) create mode 100644 src/html5ever/url.rs delete mode 100644 vendor/libidn2/config.h delete mode 100644 vendor/libidn2/darwin/strchrnul.c diff --git a/build.zig b/build.zig index 428ca7ef..a160719a 100644 --- a/build.zig +++ b/build.zig @@ -227,6 +227,7 @@ fn linkHtml5Ever(b: *Build, mod: *Build.Module) !void { "src/html5ever/lib.rs", "src/html5ever/sink.rs", "src/html5ever/types.rs", + "src/html5ever/url.rs", }) |path| { exec_cargo.addFileInput(b.path(path)); } @@ -310,13 +311,6 @@ fn linkCurl(b: *Build, mod: *Build.Module, is_tsan: bool) !void { const boringssl = buildBoringSsl(b, target, mod.optimize.?); for (boringssl) |lib| curl.root_module.linkLibrary(lib); - const libidn2 = buildLibidn2(b, target, mod.optimize.?, is_tsan); - curl.root_module.linkLibrary(libidn2); - // Also expose libidn2 to the consuming module so src/sys/idna.zig's - // @cImport of resolves. Without this, lightpanda_module only - // sees idn2.h transitively if a system libidn2 happens to be installed. - mod.linkLibrary(libidn2); - switch (target.result.os.tag) { .macos => { // needed for proxying on mac @@ -471,168 +465,6 @@ fn buildNghttp2(b: *Build, target: Build.ResolvedTarget, optimize: std.builtin.O return lib; } -fn buildLibidn2( - b: *Build, - target: Build.ResolvedTarget, - optimize: std.builtin.OptimizeMode, - is_tsan: bool, -) *Build.Step.Compile { - const dep = b.dependency("libidn2", .{}); - - const os = target.result.os.tag; - const is_darwin = os.isDarwin(); - - const mod = b.createModule(.{ - .target = target, - .optimize = optimize, - .link_libc = true, - .sanitize_thread = is_tsan, - }); - - // libidn2's autoconf+gnulib stack expects a config.h with hundreds of - // HAVE_*/_GL_ATTRIBUTE_* defines — including ~800 lines of attribute- - // detection macros emitted from gnulib-common.m4 via AH_VERBATIM. We - // vendor a single autoconf-generated config.h rather than try to - // reproduce that machinery in the Zig build system. - mod.addIncludePath(b.path("vendor/libidn2")); - - // Substitute the gnulib-style .in.h templates. All @VAR@ in them are - // either DLL-visibility markers (empty for static POSIX) or - // HAVE_UNISTRING_WOE32DLL_H (0). - inline for (.{ "unitypes", "unistr", "uniconv", "unictype", "uninorm" }) |name| { - mod.addConfigHeader(renderUnistringHeader(b, dep, name)); - } - - mod.addIncludePath(dep.path("lib")); - mod.addIncludePath(dep.path("unistring")); - // gl/ holds gnulib helpers — only malloca and version-etc headers are - // referenced from the sources we compile; we don't need the full gl/ shim - // layer (system header replacements). - mod.addIncludePath(dep.path("gl")); - - const lib = b.addLibrary(.{ .name = "idn2", .root_module = mod }); - lib.installHeader(dep.path("lib/idn2.h"), "idn2.h"); - - if (is_darwin) { - // unistring's striconveh.c calls real iconv_*, which on macOS lives - // in libiconv (separate from libSystem). On glibc Linux iconv is in - // libc itself; on musl it would also need a separate -liconv. - mod.linkSystemLibrary("iconv", .{}); - - // libidn2's lib/lookup.c calls strchrnul() without including - // ; the prototype is declared in vendor/libidn2/config.h - // alongside the existing strverscmp shim. macOS libc lacked the - // symbol entirely before 15.4 — provide it here so the link - // succeeds. Mirrors how gl/strverscmp.c is wired up below. - lib.addCSourceFile(.{ - .file = b.path("vendor/libidn2/darwin/strchrnul.c"), - .flags = &.{}, - }); - } - - lib.addCSourceFiles(.{ - .root = dep.path("lib"), - .flags = &.{ "-DHAVE_CONFIG_H", "-DIDN2_STATIC" }, - .files = &.{ - "bidi.c", "context.c", "data.c", "decode.c", - "error.c", "free.c", "idna.c", "lookup.c", - "punycode.c", "register.c", "tables.c", "tr46map.c", - "version.c", - }, - }); - lib.addCSourceFiles(.{ - .root = dep.path("gl"), - .flags = &.{"-DHAVE_CONFIG_H"}, - // malloca.c provides striconveha's stack-or-heap allocator; strverscmp - // is a glibc extension absent on macOS that lib/version.c needs. - .files = &.{ "malloca.c", "strverscmp.c" }, - }); - lib.addCSourceFiles(.{ - .root = dep.path("unistring"), - .flags = &.{"-DHAVE_CONFIG_H"}, - .files = &.{ - "c-ctype.c", "c-strcasecmp.c", "c-strncasecmp.c", - "free.c", "iconv.c", "iconv_close.c", - "iconv_open.c", "localcharset.c", "stdlib.c", - "striconveh.c", "striconveha.c", "unistd.c", - "uniconv/u8-conv-from-enc.c", "uniconv/u8-strconv-from-enc.c", "uniconv/u8-strconv-from-locale.c", - "uniconv/u8-strconv-to-enc.c", "uniconv/u8-strconv-to-locale.c", "unictype/bidi_of.c", - "unictype/categ_M.c", "unictype/categ_none.c", "unictype/categ_of.c", - "unictype/categ_test.c", "unictype/combiningclass.c", "unictype/joiningtype_of.c", - "unictype/scripts.c", "uninorm/canonical-decomposition.c", "uninorm/composition.c", - "uninorm/decompose-internal.c", "uninorm/decomposition-table.c", "uninorm/nfc.c", - "uninorm/nfd.c", "uninorm/u32-normalize.c", "unistr/u32-cmp.c", - "unistr/u32-cpy-alloc.c", "unistr/u32-cpy.c", "unistr/u32-mbtouc-unsafe.c", - "unistr/u32-strlen.c", "unistr/u32-to-u8.c", "unistr/u32-uctomb.c", - "unistr/u8-check.c", "unistr/u8-mblen.c", "unistr/u8-mbtouc.c", - "unistr/u8-mbtouc-aux.c", "unistr/u8-mbtouc-unsafe.c", "unistr/u8-mbtouc-unsafe-aux.c", - "unistr/u8-mbtoucr.c", "unistr/u8-prev.c", "unistr/u8-strlen.c", - "unistr/u8-to-u32.c", "unistr/u8-uctomb.c", "unistr/u8-uctomb-aux.c", - }, - }); - - return lib; -} - -/// Process one of unistring's `.in.h` template headers into a real `.h`. -/// All `@VAR@` substitutions in these headers are either DLL-visibility markers -/// (empty for static POSIX builds) or `HAVE_UNISTRING_WOE32DLL_H` (0). -fn renderUnistringHeader(b: *Build, dep: *Build.Dependency, name: []const u8) *Build.Step.ConfigHeader { - const in_rel = b.fmt("unistring/{s}.in.h", .{name}); - const out_name = b.fmt("{s}.h", .{name}); - const lazy = dep.path(in_rel); - const path = lazy.getPath3(b, null); - - const file = path.root_dir.handle.openFile(path.sub_path, .{}) catch |e| { - std.debug.panic("openFile {s}: {s}", .{ path.sub_path, @errorName(e) }); - }; - defer file.close(); - const contents = file.readToEndAlloc(b.allocator, 4 << 20) catch @panic("OOM"); - - const ch = b.addConfigHeader(.{ - .include_path = out_name, - .style = .{ .autoconf_at = lazy }, - }, .{}); - - var seen = std.StringHashMap(void).init(b.allocator); - var i: usize = 0; - while (std.mem.indexOfScalarPos(u8, contents, i, '@')) |s| { - const a = s + 1; - const e = std.mem.indexOfScalarPos(u8, contents, a, '@') orelse break; - const var_name = contents[a..e]; - if (!isAtConfigName(var_name)) { - // Stray '@' (e.g. an email address in a comment); advance past it - // alone so we don't mis-pair with a later '@'. - i = s + 1; - continue; - } - const owned = b.allocator.dupe(u8, var_name) catch @panic("OOM"); - const gop = seen.getOrPut(owned) catch @panic("OOM"); - if (!gop.found_existing) { - if (std.mem.eql(u8, var_name, "HAVE_UNISTRING_WOE32DLL_H")) { - ch.addValue(owned, c_int, 0); - } else { - ch.addValue(owned, []const u8, ""); - } - } - i = e + 1; - } - return ch; -} - -fn isAtConfigName(s: []const u8) bool { - if (s.len == 0) return false; - for (s, 0..) |c, idx| { - const ok = switch (c) { - 'A'...'Z', '_' => true, - '0'...'9' => idx > 0, - else => false, - }; - if (!ok) return false; - } - return true; -} - fn buildCurl( b: *Build, target: Build.ResolvedTarget, @@ -709,11 +541,11 @@ fn buildCurl( ._FILE_OFFSET_BITS = 64, .USE_IPV6 = true, - // Route IDN hostnames through libidn2 (vendored, see buildLibidn2). - // Without this, libcurl ships UTF-8 host bytes to SNI/cert validation - // and breaks for non-ASCII hostnames like räksmörgås.se. - .HAVE_LIBIDN2 = true, - .HAVE_IDN2_H = true, + // IDN is handled before libcurl (HttpClient calls URL.ensureHostAscii, + // backed by rust-url), so libcurl always receives an ASCII host and + // does not link libidn2. + .HAVE_LIBIDN2 = false, + .HAVE_IDN2_H = false, .CURL_OS = switch (os) { .linux => if (is_android) "\"android\"" else "\"linux\"", else => std.fmt.allocPrint(b.allocator, "\"{s}\"", .{@tagName(os)}) catch @panic("OOM"), diff --git a/build.zig.zon b/build.zig.zon index d891d903..4d2d7087 100644 --- a/build.zig.zon +++ b/build.zig.zon @@ -34,10 +34,6 @@ .url = "https://github.com/allyourcodebase/sqlite3/archive/8f840560eae88ab66668c6827c64ffbd0d74ef37.tar.gz", .hash = "sqlite3-3.51.0-DMxLWssOAABZ8cAvU_LfBIbp0kZjm824PU8sSLXpEDdr", }, - .libidn2 = .{ - .url = "https://ftp.gnu.org/gnu/libidn/libidn2-2.3.8.tar.gz", - .hash = "N-V-__8AABGOuAC_dhAN07kfoP4dycCFi8Bka4O-tuhriNH8", - }, }, .paths = .{""}, } diff --git a/src/browser/Page.zig b/src/browser/Page.zig index 9a3f764d..5292355f 100644 --- a/src/browser/Page.zig +++ b/src/browser/Page.zig @@ -109,12 +109,6 @@ frame: Frame, // to the original page like this. popups: std.ArrayList(*Frame) = .empty, -// Popups that have called window.close() but whose teardown is deferred to -// Page.deinit. We can't deinit synchronously from window.close() because -// that's invoked from JS still running on top of the Frame's V8 context (or -// from a script eval whose parser still holds the Frame). -queued_close: std.ArrayList(*Frame) = .empty, - // Lifecycle state. A Page is `.pending` while we hold it as the in-flight // destination of a root navigation — its V8 context exists but is not yet the // session's active context. Flipped to `.active` by Session.commitPendingPage @@ -142,8 +136,6 @@ pub fn init(self: *Page, session: *Session, frame_id: u32) !void { // Tear down the Page and its root Frame. Equivalent to the old // Session.removePage + Session.resetFrameResources. pub fn deinit(self: *Page) void { - self.cleanupClosedPopups(); - for (self.popups.items) |popup| { popup.deinit(); } @@ -197,13 +189,6 @@ pub fn deinit(self: *Page) void { session.arena_pool.release(self.frame_arena); } -pub fn cleanupClosedPopups(self: *Page) void { - for (self.queued_close.items) |popup| { - popup.deinit(); - } - self.queued_close = .empty; -} - pub fn getArena(self: *Page, size_or_bucket: anytype, debug: []const u8) !Allocator { return self.session.getArena(size_or_bucket, debug); } diff --git a/src/browser/Runner.zig b/src/browser/Runner.zig index d225469b..592f9dd5 100644 --- a/src/browser/Runner.zig +++ b/src/browser/Runner.zig @@ -86,10 +86,9 @@ fn _wait(self: *Runner, comptime is_cdp: bool, opts: WaitOpts) !void { while (true) { if (gc_hint_timer.read() >= gc_hint_period_ns) { gc_hint_timer.reset(); - self.frame._page.cleanupClosedPopups(); browser.env.memoryPressureNotification(.moderate); } - session.processQueuedDestroyed(); + session.processDestroyQueues(); const tick_result = self._tick(is_cdp, tick_opts) catch |err| { switch (err) { diff --git a/src/browser/Session.zig b/src/browser/Session.zig index f6f85347..faaf44cf 100644 --- a/src/browser/Session.zig +++ b/src/browser/Session.zig @@ -70,7 +70,8 @@ _active: ?*Page = null, // In-flight root navigation _pending: ?*Page = null, -_queued_destroy: std.ArrayList(*Page) = .{}, +_page_descruction_queue: std.ArrayList(*Page) = .{}, +_frame_destruction_queue: std.ArrayList(*Frame) = .{}, // Loader IDs are scoped to the Session: each new BrowserContext gets a // fresh counter. Frame IDs (`frame_id_gen`) live on `Browser` instead so @@ -127,7 +128,7 @@ pub fn deinit(self: *Session) void { if (self._active != null) { self.removePage(); } - self.processQueuedDestroyed(); + self.processDestroyQueues(); self.cookie_jar.deinit(); @@ -137,12 +138,27 @@ pub fn deinit(self: *Session) void { self.arena_pool.release(self.arena); } -pub fn processQueuedDestroyed(self: *Session) void { - for (self._queued_destroy.items) |page| { - page.deinit(); - self.browser.page_pool.destroy(page); +pub fn processDestroyQueues(self: *Session) void { + { + const queue = self._frame_destruction_queue.items; + if (queue.len > 0) { + for (queue) |frame| { + frame.deinit(); + } + self._frame_destruction_queue.clearRetainingCapacity(); + } + } + + { + const queue = self._page_descruction_queue.items; + if (queue.len > 0) { + for (queue) |page| { + page.deinit(); + self.browser.page_pool.destroy(page); + } + self._page_descruction_queue.clearRetainingCapacity(); + } } - self._queued_destroy.clearRetainingCapacity(); } // True iff there is an active Page. CDP / external callers should use this @@ -161,8 +177,12 @@ fn allocatePage(self: *Session, frame_id: u32) !*Page { } // Tear down and free a Page allocated via allocatePage. -fn destroyPage(self: *Session, page: *Page) void { - self._queued_destroy.append(self.arena, page) catch @panic("OOM"); +fn queuePageDestruction(self: *Session, page: *Page) void { + self._page_descruction_queue.append(self.arena, page) catch @panic("OOM"); +} + +pub fn queueFrameDestruction(self: *Session, frame: *Frame) void { + self._frame_destruction_queue.append(self.arena, frame) catch @panic("OOM"); } // Tear down the currently-active Page. Dispatches `frame_remove` first @@ -193,7 +213,7 @@ fn tearDownActivePage(self: *Session) void { }; page.frame.abortTransfers(); - self.destroyPage(page); + self.queuePageDestruction(page); self._active = null; self.navigation.onRemoveFrame(); } @@ -209,7 +229,7 @@ fn tearDownActivePage(self: *Session) void { // for any prior teardown of an old page). fn installNewActivePage(self: *Session, frame_id: u32) !*Frame { const page = try self.allocatePage(frame_id); - errdefer self.destroyPage(page); + errdefer self.queuePageDestruction(page); self._active = page; errdefer self._active = null; @@ -227,7 +247,7 @@ pub fn createPage(self: *Session) !*Frame { lp.assert(self._active == null, "Session.createPage - page not null", .{}); // Drain any pending Page deinits now, while we're at a known-safe point - self.processQueuedDestroyed(); + self.processDestroyQueues(); if (comptime IS_DEBUG) { log.debug(.browser, "create page", .{}); @@ -522,7 +542,7 @@ pub fn initiateRootNavigation(self: *Session, frame_id: u32, url: [:0]const u8, } const page = try self.allocatePage(frame_id); - errdefer self.destroyPage(page); + errdefer self.queuePageDestruction(page); page._state = .pending; self._pending = page; @@ -604,7 +624,7 @@ pub fn commitPendingPage(self: *Session) !void { // done_callback after this point would re-enter against the new // _active and trip the half-torn-down session. old_active.frame.abortTransfers(); - self.destroyPage(old_active); + self.queuePageDestruction(old_active); } // Discard a pending Page without committing. Used for failure paths @@ -622,7 +642,7 @@ pub fn discardPendingPage(self: *Session) void { page.frame.abortTransfers(); self._pending = null; - self.destroyPage(page); + self.queuePageDestruction(page); } // Frame IDs come from `Browser` (per-CDP-connection scope), not diff --git a/src/browser/URL.zig b/src/browser/URL.zig index 620e63a2..57628202 100644 --- a/src/browser/URL.zig +++ b/src/browser/URL.zig @@ -35,6 +35,20 @@ pub fn resolve(allocator: Allocator, base: [:0]const u8, source_path: anytype, o const needs_dupe = comptime !isNullTerminated(PT); var path: [:0]const u8 = if (needs_dupe or opts.always_dupe) try allocator.dupeZ(u8, source_path) else source_path; + if (std.mem.indexOfAny(u8, path, "\t\r\n")) |first| { + path = blk: { + var buf: std.ArrayList(u8) = try .initCapacity(allocator, path.len); + buf.appendSliceAssumeCapacity(path[0..first]); + for (path[first + 1 ..]) |c| { + if (c != '\t' and c != '\r' and c != '\n') { + buf.appendAssumeCapacity(c); + } + } + buf.appendAssumeCapacity(0); + break :blk buf.items[0 .. buf.items.len - 1 :0]; + }; + } + if (base.len == 0) { return processResolved(allocator, path, opts); } @@ -196,11 +210,13 @@ fn processResolved(allocator: Allocator, url: [:0]const u8, opts: ResolveOpts) ! return ensureEncoded(allocator, url, encoding); } -/// IDNA-only pass: converts a non-ASCII host (`räksmörgås.se`) to its -/// punycode form (`xn--rksmrgs-5wao1o.se`) and leaves everything else alone. +/// IDNA pass: converts a non-ASCII host (`räksmörgås.se`) to its punycode form +/// (`xn--rksmrgs-5wao1o.se`), validates any ASCII punycode (`xn--…`) labels, +/// and leaves everything else alone. Returns `error.Idna` for an invalid +/// domain (e.g. malformed punycode), which surfaces as a URL parse failure. fn ensureHostAscii(allocator: Allocator, url: [:0]const u8) ![:0]const u8 { const hostname = getHostname(url); - if (hostname.len == 0 or !idna.needsAscii(hostname)) { + if (hostname.len == 0 or (!idna.needsAscii(hostname) and !hasAceLabel(hostname))) { return url; } @@ -217,6 +233,30 @@ fn ensureHostAscii(allocator: Allocator, url: [:0]const u8) ![:0]const u8 { return buf.items[0 .. buf.items.len - 1 :0]; } +/// True if any dot-separated label of `host` begins with the IDNA ACE prefix +/// "xn--" (case-insensitive). Such labels are punycode: even though they're +/// pure ASCII, UTS#46 must decode and validate them, so they can't take the +/// `needsAscii` fast path. +fn hasAceLabel(host: []const u8) bool { + var pos: usize = 0; + while (std.mem.indexOfScalarPos(u8, host, pos, '-')) |i| { + pos = i + 1; + if (i < 2 or i + 1 >= host.len or host[i + 1] != '-') { + continue; + } + + if (!std.ascii.eqlIgnoreCase(host[i - 2 .. i], "xn")) { + continue; + } + + const label_start = i - 2; + if (label_start == 0 or host[label_start - 1] == '.') { + return true; + } + } + return false; +} + pub fn ensureEncoded(allocator: Allocator, url_in: [:0]const u8, encoding: []const u8) ![:0]const u8 { // Resolve any IDN host first; everything below operates on the ASCII form. const url = try ensureHostAscii(allocator, url_in); @@ -1086,6 +1126,70 @@ test "URL: resolve" { } } +test "URL: resolve strips tab and newline from input" { + defer testing.reset(); + + const Case = struct { + base: [:0]const u8, + path: [:0]const u8, + expected: [:0]const u8, + }; + + const cases = [_]Case{ + // Control char inside the host of an absolute URL. + .{ .base = "https://x/", .path = "https://exa\tmple.com/p", .expected = "https://example.com/p" }, + .{ .base = "https://x/", .path = "https://example.com/\n\rp", .expected = "https://example.com/p" }, + // Leading control char (first == 0). + .{ .base = "https://example/", .path = "\tfoo.js", .expected = "https://example/foo.js" }, + // Consecutive control chars. + .{ .base = "https://example/", .path = "a\t\r\nb.js", .expected = "https://example/ab.js" }, + // Control chars spread through the path. + .{ .base = "https://example/", .path = "a\tb\nc\rd.js", .expected = "https://example/abcd.js" }, + // Trailing control char. + .{ .base = "https://example/", .path = "foo.js\n", .expected = "https://example/foo.js" }, + // All-strippable relative path collapses to the base. + .{ .base = "https://example/dir/", .path = "\t\r\n", .expected = "https://example/dir/" }, + // No control chars: unchanged (the fast path). + .{ .base = "https://example/", .path = "clean.js", .expected = "https://example/clean.js" }, + }; + + for (cases) |case| { + const result = try resolve(testing.arena_allocator, case.base, case.path, .{}); + try testing.expectString(case.expected, result); + } +} + +test "URL: resolve validates ASCII punycode (xn--) labels" { + defer testing.reset(); + + // Valid punycode is left untouched (the needsAscii fast path would skip it, + // so this exercises the xn-- gate going through toAscii and back). + const ok = try resolve(testing.arena_allocator, "", "https://xn--rksmrgs-5wao1o.se/x", .{}); + try testing.expectString("https://xn--rksmrgs-5wao1o.se/x", ok); + + // Malformed punycode must be rejected rather than passed through verbatim. + // (URL.init remaps this error.Idna to TypeError for `new URL`.) + try testing.expectError(error.Idna, resolve(testing.arena_allocator, "", "https://xn--0.pt/x", .{})); + try testing.expectError(error.Idna, resolve(testing.arena_allocator, "", "https://xn--a.pt/x", .{})); +} + +test "URL: hasAceLabel" { + // ACE prefix at a label start (case-insensitive). + try testing.expectEqual(true, hasAceLabel("xn--a")); + try testing.expectEqual(true, hasAceLabel("xn--rksmrgs-5wao1o.se")); + try testing.expectEqual(true, hasAceLabel("a.xn--b.com")); + try testing.expectEqual(true, hasAceLabel("XN--ab.com")); + try testing.expectEqual(true, hasAceLabel("foo.example.xn--p1ai")); + + // Has '-', but no ACE label. + try testing.expectEqual(false, hasAceLabel("example.com")); + try testing.expectEqual(false, hasAceLabel("my-site.com")); + try testing.expectEqual(false, hasAceLabel("axn--b.com")); // xn-- not at a label start + try testing.expectEqual(false, hasAceLabel("x-n--a.com")); // not "xn" before the '-' + try testing.expectEqual(false, hasAceLabel("-.com")); + try testing.expectEqual(false, hasAceLabel("")); +} + test "URL: ensureEncoded" { defer testing.reset(); diff --git a/src/browser/js/Caller.zig b/src/browser/js/Caller.zig index 48d24a78..a6c600e4 100644 --- a/src/browser/js/Caller.zig +++ b/src/browser/js/Caller.zig @@ -372,6 +372,7 @@ fn handleError(comptime T: type, comptime F: type, local: *const Local, err: any error.TryCatchRethrow => return, error.InvalidArgument => isolate.createTypeError("invalid argument"), error.TypeError => isolate.createTypeError(""), + error.Idna => isolate.createTypeError("invalid domain"), error.RangeError => isolate.createRangeError(""), error.OutOfMemory => isolate.createError("out of memory"), error.IllegalConstructor => isolate.createError("Illegal Constructor"), diff --git a/src/browser/tests/url.html b/src/browser/tests/url.html index 60f991ea..f1e626e4 100644 --- a/src/browser/tests/url.html +++ b/src/browser/tests/url.html @@ -97,6 +97,28 @@ testing.expectEqual('http://example.com/a/b/foo', url.toString()); } + { + // IDN hosts are converted to punycode (UTS#46). + const url = new URL('https://räksmörgås.se/x'); + testing.expectEqual('xn--rksmrgs-5wao1o.se', url.hostname); + testing.expectEqual('https://xn--rksmrgs-5wao1o.se/x', url.href); + } + + { + // Valid punycode passes through unchanged. + const url = new URL('https://xn--rksmrgs-5wao1o.se/x'); + testing.expectEqual('xn--rksmrgs-5wao1o.se', url.hostname); + } + + { + // An invalid domain (malformed punycode) is a parse failure -> TypeError. + testing.withError((err) => { + testing.expectEqual(true, err.toString().includes('TypeError')); + }, () => { + const url = new URL('https://xn--0.pt/x'); + }); + } + { const base = 'http://example.com/a/b/c/d'; const url = new URL('../../../../../foo', base); diff --git a/src/browser/webapi/Window.zig b/src/browser/webapi/Window.zig index f49e1f95..df467e75 100644 --- a/src/browser/webapi/Window.zig +++ b/src/browser/webapi/Window.zig @@ -21,6 +21,7 @@ const lp = @import("lightpanda"); const builtin = @import("builtin"); const js = @import("../js/js.zig"); +const URL = @import("../URL.zig"); const Frame = @import("../Frame.zig"); const Console = @import("Console.zig"); const History = @import("History.zig"); @@ -500,6 +501,14 @@ pub fn open(self: *Window, url_: ?[]const u8, target_: ?[]const u8, features_: ? const no_opener = hasFeatureToken(features, "noopener") or hasFeatureToken(features, "noreferrer"); + if (raw_url.len > 0) { + // Per spec, we should validate the url + _ = URL.resolve(frame.call_arena, frame.base(), raw_url, .{}) catch |err| switch (err) { + error.OutOfMemory => |e| return e, + else => return error.SyntaxError, + }; + } + // _self / _parent / _top navigate the current browsing context. if (std.ascii.eqlIgnoreCase(target, "_self") or std.ascii.eqlIgnoreCase(target, "_parent") or @@ -604,9 +613,7 @@ pub fn close(self: *Window) void { // eval whose parser is still holding the Frame. Destroying the context // now leaves dangling pointers in the unwinding script eval (load event // dispatch, runMacrotasks, etc.). Defer to Page.deinit instead. - page.queued_close.append(page.frame_arena, frame) catch |err| { - log.err(.frame, "queue popup close", .{ .err = err }); - }; + page.session.queueFrameDestruction(frame); } pub fn postMessage(self: *Window, message: js.Value.Temp, target_origin: ?[]const u8, transfer: ?[]const *MessagePort, frame: *Frame) !void { @@ -995,7 +1002,7 @@ pub const JsApi = struct { pub const opener = bridge.accessor(Window.getOpener, null, .{}); pub const closed = bridge.accessor(Window.getClosed, null, .{}); pub const name = bridge.accessor(Window.getName, Window.setName, .{}); - pub const open = bridge.function(Window.open, .{}); + pub const open = bridge.function(Window.open, .{ .dom_exception = true }); pub const close = bridge.function(Window.close, .{}); pub const alert = bridge.function(struct { diff --git a/src/browser/webapi/element/html/Anchor.zig b/src/browser/webapi/element/html/Anchor.zig index 62223112..65acd009 100644 --- a/src/browser/webapi/element/html/Anchor.zig +++ b/src/browser/webapi/element/html/Anchor.zig @@ -165,6 +165,28 @@ pub fn setProtocol(self: *Anchor, value: []const u8, frame: *Frame) !void { try setHref(self, new_href, frame); } +pub fn getUsername(self: *Anchor, frame: *Frame) ![]const u8 { + const href = try getResolvedHref(self, frame) orelse return ""; + return URL.getUsername(href); +} + +pub fn setUsername(self: *Anchor, value: []const u8, frame: *Frame) !void { + const href = try getResolvedHref(self, frame) orelse return; + const new_href = try URL.setUsername(href, value, frame.call_arena); + try setHref(self, new_href, frame); +} + +pub fn getPassword(self: *Anchor, frame: *Frame) ![]const u8 { + const href = try getResolvedHref(self, frame) orelse return ""; + return URL.getPassword(href); +} + +pub fn setPassword(self: *Anchor, value: []const u8, frame: *Frame) !void { + const href = try getResolvedHref(self, frame) orelse return; + const new_href = try URL.setPassword(href, value, frame.call_arena); + try setHref(self, new_href, frame); +} + pub fn getType(self: *Anchor) []const u8 { return self.asElement().getAttributeSafe(comptime .wrap("type")) orelse ""; } @@ -221,6 +243,8 @@ pub const JsApi = struct { pub const protocol = bridge.accessor(Anchor.getProtocol, Anchor.setProtocol, .{ .ce_reactions = true }); pub const host = bridge.accessor(Anchor.getHost, Anchor.setHost, .{ .ce_reactions = true }); pub const hostname = bridge.accessor(Anchor.getHostname, Anchor.setHostname, .{ .ce_reactions = true }); + pub const username = bridge.accessor(Anchor.getUsername, Anchor.setUsername, .{ .ce_reactions = true }); + pub const password = bridge.accessor(Anchor.getPassword, Anchor.setPassword, .{ .ce_reactions = true }); pub const port = bridge.accessor(Anchor.getPort, Anchor.setPort, .{ .ce_reactions = true }); pub const pathname = bridge.accessor(Anchor.getPathname, Anchor.setPathname, .{ .ce_reactions = true }); pub const search = bridge.accessor(Anchor.getSearch, Anchor.setSearch, .{ .ce_reactions = true }); diff --git a/src/browser/webapi/element/html/Area.zig b/src/browser/webapi/element/html/Area.zig index a74e705c..d00e120e 100644 --- a/src/browser/webapi/element/html/Area.zig +++ b/src/browser/webapi/element/html/Area.zig @@ -1,4 +1,23 @@ +// Copyright (C) 2023-2026 Lightpanda (Selecy SAS) +// +// This program is free software: you can redistribute it and/or modify +// it under the terms of the GNU Affero General Public License as +// published by the Free Software Foundation, either version 3 of the +// License, or (at your option) any later version. +// +// This program is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +// GNU Affero General Public License for more details. +// +// You should have received a copy of the GNU Affero General Public License +// along with this program. If not, see . + +const std = @import("std"); const js = @import("../../../js/js.zig"); +const Frame = @import("../../../Frame.zig"); + +const URL = @import("../../../URL.zig"); const Node = @import("../../Node.zig"); const Element = @import("../../Element.zig"); const HtmlElement = @import("../Html.zig"); @@ -14,6 +33,155 @@ pub fn asNode(self: *Area) *Node { return self.asElement().asNode(); } +pub fn getHref(self: *Area, frame: *Frame) ![]const u8 { + const href = self.asElement().getAttributeSafe(comptime .wrap("href")) orelse return ""; + if (href.len == 0) { + return ""; + } + return self.asNode().resolveURL(href, frame, .{}); +} + +pub fn setHref(self: *Area, value: []const u8, frame: *Frame) !void { + try self.asElement().setAttributeSafe(comptime .wrap("href"), .wrap(value), frame); +} + +pub fn getOrigin(self: *Area, frame: *Frame) ![]const u8 { + const href = try getResolvedHref(self, frame) orelse return ""; + return (try URL.getOrigin(frame.call_arena, href)) orelse "null"; +} + +pub fn getHost(self: *Area, frame: *Frame) ![]const u8 { + const href = try getResolvedHref(self, frame) orelse return ""; + const host = URL.getHost(href); + const protocol = URL.getProtocol(href); + const port = URL.getPort(href); + + // Strip default ports + if (port.len > 0) { + if ((std.mem.eql(u8, protocol, "https:") and std.mem.eql(u8, port, "443")) or + (std.mem.eql(u8, protocol, "http:") and std.mem.eql(u8, port, "80"))) + { + return URL.getHostname(href); + } + } + + return host; +} + +pub fn setHost(self: *Area, value: []const u8, frame: *Frame) !void { + const href = try getResolvedHref(self, frame) orelse return; + const new_href = try URL.setHost(href, value, frame.call_arena); + try setHref(self, new_href, frame); +} + +pub fn getHostname(self: *Area, frame: *Frame) ![]const u8 { + const href = try getResolvedHref(self, frame) orelse return ""; + return URL.getHostname(href); +} + +pub fn setHostname(self: *Area, value: []const u8, frame: *Frame) !void { + const href = try getResolvedHref(self, frame) orelse return; + const new_href = try URL.setHostname(href, value, frame.call_arena); + try setHref(self, new_href, frame); +} + +pub fn getUsername(self: *Area, frame: *Frame) ![]const u8 { + const href = try getResolvedHref(self, frame) orelse return ""; + return URL.getUsername(href); +} + +pub fn setUsername(self: *Area, value: []const u8, frame: *Frame) !void { + const href = try getResolvedHref(self, frame) orelse return; + const new_href = try URL.setUsername(href, value, frame.call_arena); + try setHref(self, new_href, frame); +} + +pub fn getPassword(self: *Area, frame: *Frame) ![]const u8 { + const href = try getResolvedHref(self, frame) orelse return ""; + return URL.getPassword(href); +} + +pub fn setPassword(self: *Area, value: []const u8, frame: *Frame) !void { + const href = try getResolvedHref(self, frame) orelse return; + const new_href = try URL.setPassword(href, value, frame.call_arena); + try setHref(self, new_href, frame); +} + +pub fn getPort(self: *Area, frame: *Frame) ![]const u8 { + const href = try getResolvedHref(self, frame) orelse return ""; + const port = URL.getPort(href); + const protocol = URL.getProtocol(href); + + // Return empty string for default ports + if (port.len > 0) { + if ((std.mem.eql(u8, protocol, "https:") and std.mem.eql(u8, port, "443")) or + (std.mem.eql(u8, protocol, "http:") and std.mem.eql(u8, port, "80"))) + { + return ""; + } + } + + return port; +} + +pub fn setPort(self: *Area, value: ?[]const u8, frame: *Frame) !void { + const href = try getResolvedHref(self, frame) orelse return; + const new_href = try URL.setPort(href, value, frame.call_arena); + try setHref(self, new_href, frame); +} + +pub fn getSearch(self: *Area, frame: *Frame) ![]const u8 { + const href = try getResolvedHref(self, frame) orelse return ""; + return URL.getSearch(href); +} + +pub fn setSearch(self: *Area, value: []const u8, frame: *Frame) !void { + const href = try getResolvedHref(self, frame) orelse return; + const new_href = try URL.setSearch(href, value, frame.call_arena); + try setHref(self, new_href, frame); +} + +pub fn getHash(self: *Area, frame: *Frame) ![]const u8 { + const href = try getResolvedHref(self, frame) orelse return ""; + return URL.getHash(href); +} + +pub fn setHash(self: *Area, value: []const u8, frame: *Frame) !void { + const href = try getResolvedHref(self, frame) orelse return; + const new_href = try URL.setHash(href, value, frame.call_arena); + try setHref(self, new_href, frame); +} + +pub fn getPathname(self: *Area, frame: *Frame) ![]const u8 { + const href = try getResolvedHref(self, frame) orelse return ""; + return URL.getPathname(href); +} + +pub fn setPathname(self: *Area, value: []const u8, frame: *Frame) !void { + const href = try getResolvedHref(self, frame) orelse return; + const new_href = try URL.setPathname(href, value, frame.call_arena); + try setHref(self, new_href, frame); +} + +pub fn getProtocol(self: *Area, frame: *Frame) ![]const u8 { + const href = try getResolvedHref(self, frame) orelse return ""; + return URL.getProtocol(href); +} + +pub fn setProtocol(self: *Area, value: []const u8, frame: *Frame) !void { + const href = try getResolvedHref(self, frame) orelse return; + const new_href = try URL.setProtocol(href, value, frame.call_arena); + try setHref(self, new_href, frame); +} + +fn getResolvedHref(self: *Area, frame: *Frame) !?[:0]const u8 { + const href = self.asElement().getAttributeSafe(comptime .wrap("href")) orelse return null; + if (href.len == 0) { + return null; + } + return try self.asNode().resolveURL(href, frame, .{}); +} + pub const JsApi = struct { pub const bridge = js.Bridge(Area); @@ -22,4 +190,17 @@ pub const JsApi = struct { pub const prototype_chain = bridge.prototypeChain(); pub var class_id: bridge.ClassId = undefined; }; + + pub const href = bridge.accessor(Area.getHref, Area.setHref, .{ .ce_reactions = true }); + pub const origin = bridge.accessor(Area.getOrigin, null, .{}); + pub const protocol = bridge.accessor(Area.getProtocol, Area.setProtocol, .{ .ce_reactions = true }); + pub const host = bridge.accessor(Area.getHost, Area.setHost, .{ .ce_reactions = true }); + pub const hostname = bridge.accessor(Area.getHostname, Area.setHostname, .{ .ce_reactions = true }); + pub const username = bridge.accessor(Area.getUsername, Area.setUsername, .{ .ce_reactions = true }); + pub const password = bridge.accessor(Area.getPassword, Area.setPassword, .{ .ce_reactions = true }); + pub const port = bridge.accessor(Area.getPort, Area.setPort, .{ .ce_reactions = true }); + pub const pathname = bridge.accessor(Area.getPathname, Area.setPathname, .{ .ce_reactions = true }); + pub const search = bridge.accessor(Area.getSearch, Area.setSearch, .{ .ce_reactions = true }); + pub const hash = bridge.accessor(Area.getHash, Area.setHash, .{ .ce_reactions = true }); + pub const toString = bridge.function(Area.getHref, .{}); }; diff --git a/src/html5ever/Cargo.lock b/src/html5ever/Cargo.lock index 89a69176..d1646acb 100644 --- a/src/html5ever/Cargo.lock +++ b/src/html5ever/Cargo.lock @@ -30,6 +30,17 @@ version = "1.0.0" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd" +[[package]] +name = "displaydoc" +version = "0.2.6" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "1ac70aa55017e108007fbaf5aa0f54b021c98f92ff8af59d42eda9da96e3dd4f" +dependencies = [ + "proc-macro2", + "quote", + "syn", +] + [[package]] name = "encoding_rs" version = "0.8.35" @@ -61,6 +72,109 @@ dependencies = [ "markup5ever", ] +[[package]] +name = "icu_collections" +version = "2.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "2984d1cd16c883d7935b9e07e44071dca8d917fd52ecc02c04d5fa0b5a3f191c" +dependencies = [ + "displaydoc", + "potential_utf", + "utf8_iter", + "yoke", + "zerofrom", + "zerovec", +] + +[[package]] +name = "icu_locale_core" +version = "2.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "92219b62b3e2b4d88ac5119f8904c10f8f61bf7e95b640d25ba3075e6cac2c29" +dependencies = [ + "displaydoc", + "litemap", + "tinystr", + "writeable", + "zerovec", +] + +[[package]] +name = "icu_normalizer" +version = "2.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "c56e5ee99d6e3d33bd91c5d85458b6005a22140021cc324cea84dd0e72cff3b4" +dependencies = [ + "icu_collections", + "icu_normalizer_data", + "icu_properties", + "icu_provider", + "smallvec", + "zerovec", +] + +[[package]] +name = "icu_normalizer_data" +version = "2.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "da3be0ae77ea334f4da67c12f149704f19f81d1adf7c51cf482943e84a2bad38" + +[[package]] +name = "icu_properties" +version = "2.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "bee3b67d0ea5c2cca5003417989af8996f8604e34fb9ddf96208a033901e70de" +dependencies = [ + "icu_collections", + "icu_locale_core", + "icu_properties_data", + "icu_provider", + "zerotrie", + "zerovec", +] + +[[package]] +name = "icu_properties_data" +version = "2.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "8e2bbb201e0c04f7b4b3e14382af113e17ba4f63e2c9d2ee626b720cbce54a14" + +[[package]] +name = "icu_provider" +version = "2.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "139c4cf31c8b5f33d7e199446eff9c1e02decfc2f0eec2c8d71f65befa45b421" +dependencies = [ + "displaydoc", + "icu_locale_core", + "writeable", + "yoke", + "zerofrom", + "zerotrie", + "zerovec", +] + +[[package]] +name = "idna" +version = "1.1.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "3b0875f23caa03898994f6ddc501886a45c7d3d62d04d2d90788d47be1b1e4de" +dependencies = [ + "idna_adapter", + "smallvec", + "utf8_iter", +] + +[[package]] +name = "idna_adapter" +version = "1.2.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "cb68373c0d6620ef8105e855e7745e18b0d00d3bdb07fb532e434244cdb9a714" +dependencies = [ + "icu_normalizer", + "icu_properties", +] + [[package]] name = "libc" version = "0.2.172" @@ -73,6 +187,7 @@ version = "0.1.0" dependencies = [ "encoding_rs", "html5ever", + "idna", "string_cache", "tikv-jemalloc-ctl", "tikv-jemallocator", @@ -80,6 +195,12 @@ dependencies = [ "xml5ever", ] +[[package]] +name = "litemap" +version = "0.8.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "92daf443525c4cce67b150400bc2316076100ce0b3686209eb8cf3c31612e6f0" + [[package]] name = "lock_api" version = "0.4.13" @@ -181,6 +302,15 @@ dependencies = [ "siphasher", ] +[[package]] +name = "potential_utf" +version = "0.1.5" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "0103b1cef7ec0cf76490e969665504990193874ea05c85ff9bab8b911d0a0564" +dependencies = [ + "zerovec", +] + [[package]] name = "precomputed-hash" version = "0.1.1" @@ -198,9 +328,9 @@ dependencies = [ [[package]] name = "quote" -version = "1.0.40" +version = "1.0.45" source = "registry+https://github.com/rust-lang/crates.io-index" -checksum = "1885c039570dc00dcb4ff087a89e185fd56bae234ddc7f056a945bf36467248d" +checksum = "41f2619966050689382d2b44f664f4bc593e129785a36d6ee376ddf37259b924" dependencies = [ "proc-macro2", ] @@ -258,6 +388,12 @@ version = "1.15.1" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "67b1b7a3b5fe4f1376887184045fcf45c69e92af734b7aaddc05fb777b6fbd03" +[[package]] +name = "stable_deref_trait" +version = "1.2.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "6ce2be8dc25455e1f91df71bfa12ad37d7af1092ae736f3a6cd0e37bc7810596" + [[package]] name = "string_cache" version = "0.9.0" @@ -294,6 +430,17 @@ dependencies = [ "unicode-ident", ] +[[package]] +name = "synstructure" +version = "0.13.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "728a70f3dbaf5bab7f0c4b1ac8d7ae5ea60a4b5549c8a5914361c99147a709d2" +dependencies = [ + "proc-macro2", + "quote", + "syn", +] + [[package]] name = "tendril" version = "0.5.0" @@ -335,6 +482,16 @@ dependencies = [ "tikv-jemalloc-sys", ] +[[package]] +name = "tinystr" +version = "0.8.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "c8323304221c2a851516f22236c5722a72eaa19749016521d6dff0824447d96d" +dependencies = [ + "displaydoc", + "zerovec", +] + [[package]] name = "typed-arena" version = "2.0.2" @@ -353,6 +510,12 @@ version = "0.7.6" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "09cc8ee72d2a9becf2f2febe0205bbed8fc6615b7cb429ad062dc7b7ddd036a9" +[[package]] +name = "utf8_iter" +version = "1.0.4" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "b6c140620e7ffbb22c2dee59cafe6084a59b5ffc27a8859a5f0d494b5d52b6be" + [[package]] name = "web_atoms" version = "0.2.3" @@ -429,6 +592,12 @@ version = "0.52.6" source = "registry+https://github.com/rust-lang/crates.io-index" checksum = "589f6da84c646204747d1270a2a5661ea66ed1cced2631d546fdfb155959f9ec" +[[package]] +name = "writeable" +version = "0.6.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "1ffae5123b2d3fc086436f8834ae3ab053a283cfac8fe0a0b8eaae044768a4c4" + [[package]] name = "xml5ever" version = "0.39.0" @@ -438,3 +607,80 @@ dependencies = [ "log", "markup5ever", ] + +[[package]] +name = "yoke" +version = "0.8.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "abe8c5fda708d9ca3df187cae8bfb9ceda00dd96231bed36e445a1a48e66f9ca" +dependencies = [ + "stable_deref_trait", + "yoke-derive", + "zerofrom", +] + +[[package]] +name = "yoke-derive" +version = "0.8.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "de844c262c8848816172cef550288e7dc6c7b7814b4ee56b3e1553f275f1858e" +dependencies = [ + "proc-macro2", + "quote", + "syn", + "synstructure", +] + +[[package]] +name = "zerofrom" +version = "0.1.8" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "0ec05a11813ea801ff6d75110ad09cd0824ddba17dfe17128ea0d5f68e6c5272" +dependencies = [ + "zerofrom-derive", +] + +[[package]] +name = "zerofrom-derive" +version = "0.1.7" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "11532158c46691caf0f2593ea8358fed6bbf68a0315e80aae9bd41fbade684a1" +dependencies = [ + "proc-macro2", + "quote", + "syn", + "synstructure", +] + +[[package]] +name = "zerotrie" +version = "0.2.4" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "0f9152d31db0792fa83f70fb2f83148effb5c1f5b8c7686c3459e361d9bc20bf" +dependencies = [ + "displaydoc", + "yoke", + "zerofrom", +] + +[[package]] +name = "zerovec" +version = "0.11.6" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "90f911cbc359ab6af17377d242225f4d75119aec87ea711a880987b18cd7b239" +dependencies = [ + "yoke", + "zerofrom", + "zerovec-derive", +] + +[[package]] +name = "zerovec-derive" +version = "0.11.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "625dc425cab0dca6dc3c3319506e6593dcb08a9f387ea3b284dbd52a92c40555" +dependencies = [ + "proc-macro2", + "quote", + "syn", +] diff --git a/src/html5ever/Cargo.toml b/src/html5ever/Cargo.toml index 70c434af..2cc913cf 100644 --- a/src/html5ever/Cargo.toml +++ b/src/html5ever/Cargo.toml @@ -16,6 +16,7 @@ tikv-jemallocator = {version = "0.6.1", features = ["stats"]} tikv-jemalloc-ctl = {version = "0.6.1", features = ["stats"]} xml5ever = "0.39.0" encoding_rs = "0.8" +idna = "1.1.0" [profile.release] lto = true diff --git a/src/html5ever/lib.rs b/src/html5ever/lib.rs index 0fccf9de..709d508c 100644 --- a/src/html5ever/lib.rs +++ b/src/html5ever/lib.rs @@ -18,6 +18,7 @@ mod sink; mod types; +mod url; #[cfg(debug_assertions)] #[global_allocator] @@ -159,8 +160,7 @@ pub extern "C" fn html5ever_parse_document_with_encoding( }; // Parse directly from decoded string - parse_document(sink, Default::default()) - .one(StrTendril::from(decoded.as_ref())); + parse_document(sink, Default::default()).one(StrTendril::from(decoded.as_ref())); } // === Encoding API for TextDecoder === @@ -180,10 +180,7 @@ pub struct EncodingInfo { /// Look up an encoding by its label (case-insensitive, whitespace-trimmed) #[no_mangle] -pub extern "C" fn encoding_for_label( - label: *const c_uchar, - label_len: usize, -) -> EncodingInfo { +pub extern "C" fn encoding_for_label(label: *const c_uchar, label_len: usize) -> EncodingInfo { if label.is_null() || label_len == 0 { return EncodingInfo { found: 0, diff --git a/src/html5ever/url.rs b/src/html5ever/url.rs new file mode 100644 index 00000000..98a83b60 --- /dev/null +++ b/src/html5ever/url.rs @@ -0,0 +1,83 @@ +// Copyright (C) 2023-2026 Lightpanda (Selecy SAS) +// +// This program is free software: you can redistribute it and/or modify +// it under the terms of the GNU Affero General Public License as +// published by the Free Software Foundation, either version 3 of the +// License, or (at your option) any later version. +// +// This program is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +// GNU Affero General Public License for more details. +// +// You should have received a copy of the GNU Affero General Public License +// along with this program. If not, see . + +// WHATWG "domain to ASCII" backed by the `idna` crate (UTS#46, the same engine +// rust-url/Servo use). Pairs with src/sys/idna.zig. Replaced libidn2, whose +// IDNA-2008 behavior diverged from the spec. Value-in / value-out: a UTF-8 +// host string becomes its punycode form, or an error. + +use std::os::raw::c_uchar; +use std::slice; + +fn str_from(ptr: *const c_uchar, len: usize) -> Option<&'static str> { + // Zig hands empty slices a non-null but dangling pointer, so length must + // be checked before forming a slice from raw parts. + if ptr.is_null() || len == 0 { + return Some(""); + } + let bytes = unsafe { slice::from_raw_parts(ptr, len) }; + std::str::from_utf8(bytes).ok() +} + +// Catch any panic from the IDNA code so it never unwinds across the extern "C" +// boundary and aborts the whole process; a panic becomes error code 1. +fn ffi_guard i32>(f: F) -> i32 { + std::panic::catch_unwind(std::panic::AssertUnwindSafe(f)).unwrap_or(1) +} + +/// WHATWG "domain to ASCII" (UTS#46, non-transitional, beStrict=false). Writes +/// a NUL-terminated owned buffer to *out_ptr / *out_len (caller frees with +/// lpurl_free). Returns 0 on success, 1 if `host` is not a valid domain. +#[no_mangle] +pub extern "C" fn lpurl_domain_to_ascii( + host_ptr: *const c_uchar, + host_len: usize, + out_ptr: *mut *mut c_uchar, + out_len: *mut usize, +) -> i32 { + ffi_guard(move || { + let host = match str_from(host_ptr, host_len) { + Some(s) => s, + None => return 1, + }; + let ascii = match idna::domain_to_ascii(host) { + Ok(s) => s, + Err(_) => return 1, + }; + let len = ascii.len(); + let mut bytes = ascii.into_bytes(); + bytes.push(0); + let boxed = bytes.into_boxed_slice(); + unsafe { + *out_ptr = Box::into_raw(boxed) as *mut c_uchar; + *out_len = len; + } + 0 + }) +} + +/// Free a NUL-terminated buffer handed out by lpurl_domain_to_ascii. +#[no_mangle] +pub extern "C" fn lpurl_free(ptr: *mut c_uchar, len: usize) { + if ptr.is_null() { + return; + } + // The buffer included a NUL terminator, so its length is len + 1 and its + // capacity matches exactly (it was a boxed slice). + unsafe { + let slice = std::ptr::slice_from_raw_parts_mut(ptr, len + 1); + drop(Box::from_raw(slice)); + } +} diff --git a/src/sys/idna.zig b/src/sys/idna.zig index 368928e5..39e8c65f 100644 --- a/src/sys/idna.zig +++ b/src/sys/idna.zig @@ -15,12 +15,19 @@ const std = @import("std"); -const c = @cImport({ - @cInclude("idn2.h"); -}); - const Allocator = std.mem.Allocator; -pub const Error = error{Idna} || Allocator.Error; + +// WHATWG "domain to ASCII" lives in the rust-url FFI (src/html5ever/url.rs), +// which uses the UTS#46-conformant `idna` crate — the same engine rust-url +// itself uses. +extern "c" fn lpurl_domain_to_ascii( + host_ptr: [*]const u8, + host_len: usize, + out_ptr: *?[*]u8, + out_len: *usize, +) i32; + +extern "c" fn lpurl_free(ptr: ?[*]u8, len: usize) void; /// True if `host` contains any non-ASCII byte and therefore needs IDNA /// processing. Pure-ASCII hostnames are returned unchanged by `toAscii`, @@ -35,21 +42,16 @@ pub fn needsAscii(host: []const u8) bool { } /// Convert a UTF-8 hostname to its ASCII (Punycode) form per UTS#46 -/// IDNA 2008 with non-transitional processing — the algorithm WHATWG URL -/// invokes as "domain to ASCII". Returns an allocator-owned slice. -pub fn toAscii(allocator: Allocator, host: []const u8) Error![]u8 { - const host_z = try allocator.dupeZ(u8, host); - defer allocator.free(host_z); - - var out_ptr: [*c]u8 = undefined; - const flags: c_int = c.IDN2_NFC_INPUT | c.IDN2_NONTRANSITIONAL; - const rc = c.idn2_to_ascii_8z(host_z.ptr, &out_ptr, flags); - if (rc != c.IDN2_OK) { +/// non-transitional processing — the algorithm WHATWG URL invokes as +/// "domain to ASCII". Returns an allocator-owned slice. +pub fn toAscii(allocator: Allocator, host: []const u8) ![]u8 { + var out_len: usize = 0; + var out_ptr: ?[*]u8 = null; + if (lpurl_domain_to_ascii(host.ptr, host.len, &out_ptr, &out_len) != 0) { return error.Idna; } - defer c.idn2_free(out_ptr); - - return try allocator.dupe(u8, std.mem.span(@as([*:0]const u8, @ptrCast(out_ptr)))); + defer lpurl_free(out_ptr, out_len); + return allocator.dupe(u8, out_ptr.?[0..out_len]); } const testing = @import("../testing.zig"); @@ -74,3 +76,39 @@ test "idna: German sharp s with non-transitional processing" { defer testing.allocator.free(out); try testing.expectString("xn--fa-hia.de", out); } + +test "idna: needsAscii" { + try testing.expectEqual(false, needsAscii("")); + try testing.expectEqual(false, needsAscii("xn--fa-hia.de")); + try testing.expectEqual(true, needsAscii("faß.de")); + try testing.expectEqual(true, needsAscii("\xff")); +} + +test "idna: UTS#46 lowercases ASCII" { + const out = try toAscii(testing.allocator, "EXAMPLE.COM"); + defer testing.allocator.free(out); + try testing.expectString("example.com", out); +} + +test "idna: already-punycode is idempotent" { + const out = try toAscii(testing.allocator, "xn--rksmrgs-5wao1o.se"); + defer testing.allocator.free(out); + try testing.expectString("xn--rksmrgs-5wao1o.se", out); +} + +test "idna: mixed ASCII and non-ASCII labels" { + const out = try toAscii(testing.allocator, "münchen.example.com"); + defer testing.allocator.free(out); + try testing.expectString("xn--mnchen-3ya.example.com", out); +} + +test "idna: multi-label CJK" { + const out = try toAscii(testing.allocator, "日本.jp"); + defer testing.allocator.free(out); + try testing.expectString("xn--wgv71a.jp", out); +} + +test "idna: invalid domain returns error" { + // U+FFFD (REPLACEMENT CHARACTER) is disallowed under UTS#46. + try testing.expectError(error.Idna, toAscii(testing.allocator, "\u{FFFD}.com")); +} diff --git a/vendor/libidn2/config.h b/vendor/libidn2/config.h deleted file mode 100644 index 539177ab..00000000 --- a/vendor/libidn2/config.h +++ /dev/null @@ -1,1915 +0,0 @@ -/* config.h. Generated from config.h.in by configure. */ -/* config.h.in. Generated from configure.ac by autoheader. */ - -/* Witness that has been included. */ -#define _GL_CONFIG_H_INCLUDED 1 - - -/* Define to the number of bits in type 'ptrdiff_t'. */ -/* #undef BITSIZEOF_PTRDIFF_T */ - -/* Define to the number of bits in type 'sig_atomic_t'. */ -/* #undef BITSIZEOF_SIG_ATOMIC_T */ - -/* Define to the number of bits in type 'size_t'. */ -/* #undef BITSIZEOF_SIZE_T */ - -/* Define to the number of bits in type 'wchar_t'. */ -/* #undef BITSIZEOF_WCHAR_T */ - -/* Define to the number of bits in type 'wint_t'. */ -/* #undef BITSIZEOF_WINT_T */ - -/* Define to 1 if using 'alloca.c'. */ -/* #undef C_ALLOCA */ - -/* Define to 1 if // is a file system root distinct from /. */ -/* #undef DOUBLE_SLASH_IS_DISTINCT_ROOT */ - -/* Define to 1 if translation of program messages to the user's native - language is requested. */ -/* #undef ENABLE_NLS */ - -/* Define this to 1 if F_DUPFD behavior does not match POSIX */ -/* #undef FCNTL_DUPFD_BUGGY */ - -/* Define to a C preprocessor expression that evaluates to 1 or 0, depending - whether the gnulib module close shall be considered present. */ -#define GNULIB_CLOSE 1 - -/* Define to a C preprocessor expression that evaluates to 1 or 0, depending - whether the gnulib module fscanf shall be considered present. */ -#define GNULIB_FSCANF 1 - -/* Define to a C preprocessor expression that evaluates to 1 or 0, depending - whether the gnulib module fstat shall be considered present. */ -#define GNULIB_FSTAT 1 - -/* Define to the directory where to find the localizations of the translation - domain 'gnulib', as a C string. */ -#define GNULIB_LOCALEDIR "/usr/local/share/locale" - -/* Define to a C preprocessor expression that evaluates to 1 or 0, depending - whether the gnulib module msvc-nothrow shall be considered present. */ -#define GNULIB_MSVC_NOTHROW 1 - -/* Disable VLA usage in gettext.h. */ -#define GNULIB_NO_VLA 1 - -/* Define to 1 if printf and friends should be labeled with attribute - "__gnu_printf__" instead of "__printf__" */ -/* #undef GNULIB_PRINTF_ATTRIBUTE_FLAVOR_GNU */ - -/* Define to a C preprocessor expression that evaluates to 1 or 0, depending - whether the gnulib module scanf shall be considered present. */ -#define GNULIB_SCANF 1 - -/* Define to a C preprocessor expression that evaluates to 1 or 0, depending - whether the gnulib module stat shall be considered present. */ -#define GNULIB_STAT 1 - -/* Define to a C preprocessor expression that evaluates to 1 or 0, depending - whether the gnulib module strerror shall be considered present. */ -#define GNULIB_STRERROR 1 - -/* Define to 1 when the gnulib module cloexec should be tested. */ -#define GNULIB_TEST_CLOEXEC 1 - -/* Define to 1 when the gnulib module close should be tested. */ -#define GNULIB_TEST_CLOSE 1 - -/* Define to 1 when the gnulib module dup2 should be tested. */ -#define GNULIB_TEST_DUP2 1 - -/* Define to 1 when the gnulib module fcntl should be tested. */ -#define GNULIB_TEST_FCNTL 1 - -/* Define to 1 when the gnulib module fgetc should be tested. */ -#define GNULIB_TEST_FGETC 1 - -/* Define to 1 when the gnulib module fgets should be tested. */ -#define GNULIB_TEST_FGETS 1 - -/* Define to 1 when the gnulib module fprintf should be tested. */ -#define GNULIB_TEST_FPRINTF 1 - -/* Define to 1 when the gnulib module fputc should be tested. */ -#define GNULIB_TEST_FPUTC 1 - -/* Define to 1 when the gnulib module fputs should be tested. */ -#define GNULIB_TEST_FPUTS 1 - -/* Define to 1 when the gnulib module fread should be tested. */ -#define GNULIB_TEST_FREAD 1 - -/* Define to 1 when the gnulib module free-posix should be tested. */ -#define GNULIB_TEST_FREE_POSIX 1 - -/* Define to 1 when the gnulib module fscanf should be tested. */ -#define GNULIB_TEST_FSCANF 1 - -/* Define to 1 when the gnulib module fstat should be tested. */ -#define GNULIB_TEST_FSTAT 1 - -/* Define to 1 when the gnulib module fwrite should be tested. */ -#define GNULIB_TEST_FWRITE 1 - -/* Define to 1 when the gnulib module getc should be tested. */ -#define GNULIB_TEST_GETC 1 - -/* Define to 1 when the gnulib module getchar should be tested. */ -#define GNULIB_TEST_GETCHAR 1 - -/* Define to 1 when the gnulib module getdelim should be tested. */ -#define GNULIB_TEST_GETDELIM 1 - -/* Define to 1 when the gnulib module getdtablesize should be tested. */ -#define GNULIB_TEST_GETDTABLESIZE 1 - -/* Define to 1 when the gnulib module getline should be tested. */ -#define GNULIB_TEST_GETLINE 1 - -/* Define to 1 when the gnulib module getopt-posix should be tested. */ -#define GNULIB_TEST_GETOPT_POSIX 1 - -/* Define to 1 when the gnulib module getprogname should be tested. */ -#define GNULIB_TEST_GETPROGNAME 1 - -/* Define to 1 when the gnulib module open should be tested. */ -#define GNULIB_TEST_OPEN 1 - -/* Define to 1 when the gnulib module printf should be tested. */ -#define GNULIB_TEST_PRINTF 1 - -/* Define to 1 when the gnulib module putc should be tested. */ -#define GNULIB_TEST_PUTC 1 - -/* Define to 1 when the gnulib module putchar should be tested. */ -#define GNULIB_TEST_PUTCHAR 1 - -/* Define to 1 when the gnulib module puts should be tested. */ -#define GNULIB_TEST_PUTS 1 - -/* Define to 1 when the gnulib module rawmemchr should be tested. */ -#define GNULIB_TEST_RAWMEMCHR 1 - -/* Define to 1 when the gnulib module scanf should be tested. */ -#define GNULIB_TEST_SCANF 1 - -/* Define to 1 when the gnulib module stat should be tested. */ -#define GNULIB_TEST_STAT 1 - -/* Define to 1 when the gnulib module strchrnul should be tested. */ -#define GNULIB_TEST_STRCHRNUL 1 - -/* Define to 1 when the gnulib module strerror should be tested. */ -#define GNULIB_TEST_STRERROR 1 - -/* Define to 1 when the gnulib module strverscmp should be tested. */ -#define GNULIB_TEST_STRVERSCMP 1 - -/* Define to 1 when the gnulib module uninorm/u32-normalize should be tested. - */ -#define GNULIB_TEST_UNINORM_U32_NORMALIZE 1 - -/* Define to 1 when the gnulib module vfprintf should be tested. */ -#define GNULIB_TEST_VFPRINTF 1 - -/* Define to 1 when the gnulib module vprintf should be tested. */ -#define GNULIB_TEST_VPRINTF 1 - -/* Define to a C preprocessor expression that evaluates to 1 or 0, depending - whether the gnulib module unistr/u32-mbtouc-unsafe shall be considered - present. */ -#define GNULIB_UNISTR_U32_MBTOUC_UNSAFE 1 - -/* Define to a C preprocessor expression that evaluates to 1 or 0, depending - whether the gnulib module unistr/u32-uctomb shall be considered present. */ -#define GNULIB_UNISTR_U32_UCTOMB 1 - -/* Define to a C preprocessor expression that evaluates to 1 or 0, depending - whether the gnulib module unistr/u8-mbtouc shall be considered present. */ -#define GNULIB_UNISTR_U8_MBTOUC 1 - -/* Define to a C preprocessor expression that evaluates to 1 or 0, depending - whether the gnulib module unistr/u8-mbtoucr shall be considered present. */ -#define GNULIB_UNISTR_U8_MBTOUCR 1 - -/* Define to a C preprocessor expression that evaluates to 1 or 0, depending - whether the gnulib module unistr/u8-mbtouc-unsafe shall be considered - present. */ -#define GNULIB_UNISTR_U8_MBTOUC_UNSAFE 1 - -/* Define to a C preprocessor expression that evaluates to 1 or 0, depending - whether the gnulib module unistr/u8-uctomb shall be considered present. */ -#define GNULIB_UNISTR_U8_UCTOMB 1 - -/* Define to 1 if you have 'alloca' after including , a header that - may be supplied by this distribution. */ -#define HAVE_ALLOCA 1 - -/* Define to 1 if works. */ -#define HAVE_ALLOCA_H 1 - -/* Define to 1 if you have the Mac OS X function CFLocaleCopyCurrent in the - CoreFoundation framework. */ -#define HAVE_CFLOCALECOPYCURRENT 1 - -/* Define to 1 if you have the Mac OS X function CFPreferencesCopyAppValue in - the CoreFoundation framework. */ -#define HAVE_CFPREFERENCESCOPYAPPVALUE 1 - -/* Define to 1 if you have the header file. */ -/* #undef HAVE_CRTDEFS_H */ - -/* Define to 1 if bool, true and false work as per C2023. */ -/* #undef HAVE_C_BOOL */ - -/* Define to 1 if the static_assert keyword works. */ -/* #undef HAVE_C_STATIC_ASSERT */ - -/* Define to 1 if C supports variable-length arrays. */ -#define HAVE_C_VARARRAYS 1 - -/* Define if the GNU dcgettext() function is already present or preinstalled. - */ -/* #undef HAVE_DCGETTEXT */ - -/* Define to 1 if you have the declaration of `ecvt', and to 0 if you don't. - */ -#define HAVE_DECL_ECVT 1 - -/* Define to 1 if you have the declaration of `execvpe', and to 0 if you - don't. */ -#define HAVE_DECL_EXECVPE 0 - -/* Define to 1 if you have the declaration of `fcloseall', and to 0 if you - don't. */ -#define HAVE_DECL_FCLOSEALL 0 - -/* Define to 1 if you have the declaration of `fcvt', and to 0 if you don't. - */ -#define HAVE_DECL_FCVT 1 - -/* Define to 1 if you have the declaration of `gcvt', and to 0 if you don't. - */ -#define HAVE_DECL_GCVT 1 - -/* Define to 1 if you have the declaration of `getc_unlocked', and to 0 if you - don't. */ -#define HAVE_DECL_GETC_UNLOCKED 1 - -/* Define to 1 if you have the declaration of `getdelim', and to 0 if you - don't. */ -#define HAVE_DECL_GETDELIM 1 - -/* Define to 1 if you have the declaration of `getdtablesize', and to 0 if you - don't. */ -#define HAVE_DECL_GETDTABLESIZE 1 - -/* Define to 1 if you have the declaration of `getline', and to 0 if you - don't. */ -#define HAVE_DECL_GETLINE 1 - -/* Define to 1 if you have the declaration of `getw', and to 0 if you don't. - */ -#define HAVE_DECL_GETW 1 - -/* Define to 1 if you have the declaration of `program_invocation_name', and - to 0 if you don't. */ -#define HAVE_DECL_PROGRAM_INVOCATION_NAME 0 - -/* Define to 1 if you have the declaration of `program_invocation_short_name', - and to 0 if you don't. */ -#define HAVE_DECL_PROGRAM_INVOCATION_SHORT_NAME 0 - -/* Define to 1 if you have the declaration of `putw', and to 0 if you don't. - */ -#define HAVE_DECL_PUTW 1 - -/* Define to 1 if you have the declaration of `strerror_r', and to 0 if you - don't. */ -#define HAVE_DECL_STRERROR_R 1 - -/* Define to 1 if you have the declaration of `wcsdup', and to 0 if you don't. - */ -#define HAVE_DECL_WCSDUP 1 - -/* Define to 1 if you have the declaration of `__argv', and to 0 if you don't. - */ -/* #undef HAVE_DECL___ARGV */ - -/* Define to 1 if you have the header file. */ -#define HAVE_DLFCN_H 1 - -/* Define to 1 if you have the `error' function. */ -/* #undef HAVE_ERROR */ - -/* Define to 1 if you have the header file. */ -/* #undef HAVE_ERROR_H */ - -/* Define to 1 if you have the `fcntl' function. */ -#define HAVE_FCNTL 1 - -/* Define to 1 if you have the header file. */ -/* #undef HAVE_FEATURES_H */ - -/* Define to 1 if you have the `flockfile' function. */ -#define HAVE_FLOCKFILE 1 - -/* Define if the 'free' function is guaranteed to preserve errno. */ -/* #undef HAVE_FREE_POSIX */ - -/* Define to 1 if you have the `funlockfile' function. */ -#define HAVE_FUNLOCKFILE 1 - -/* Define to 1 if you have the `getdelim' function. */ -#define HAVE_GETDELIM 1 - -/* Define to 1 if you have the `getdtablesize' function. */ -#define HAVE_GETDTABLESIZE 1 - -/* Define to 1 if you have the `getexecname' function. */ -/* #undef HAVE_GETEXECNAME */ - -/* Define to 1 if you have the `getline' function. */ -#define HAVE_GETLINE 1 - -/* Define to 1 if you have the header file. */ -#define HAVE_GETOPT_H 1 - -/* Define to 1 if you have the `getopt_long_only' function. */ -#define HAVE_GETOPT_LONG_ONLY 1 - -/* Define to 1 if you have the `getprogname' function. */ -#define HAVE_GETPROGNAME 1 - -/* Define if the GNU gettext() function is already present or preinstalled. */ -/* #undef HAVE_GETTEXT */ - -/* Define if you have the iconv() function and it works. */ -/* #undef HAVE_ICONV */ - -/* Define to 1 if you have the header file. */ -#define HAVE_ICONV_H 1 - -/* Define to 1 if the compiler supports one of the keywords 'inline', - '__inline__', '__inline' and effectively inlines functions marked as such. - */ -#define HAVE_INLINE 1 - -/* Define to 1 if you have the header file. */ -#define HAVE_INTTYPES_H 1 - -/* Define if you have and nl_langinfo(CODESET). */ -#define HAVE_LANGINFO_CODESET 1 - -/* Define if you have the libunistring library. */ -/* #undef HAVE_LIBUNISTRING */ - -/* Define to 1 if you have the header file. */ -#define HAVE_LIMITS_H 1 - -/* Define to 1 if the system has the type 'long long int'. */ -#define HAVE_LONG_LONG_INT 1 - -/* Define to 1 if you have the `lstat' function. */ -#define HAVE_LSTAT 1 - -/* Define to 1 if malloc (0) returns nonnull. */ -#define HAVE_MALLOC_0_NONNULL 1 - -/* Define if malloc, realloc, and calloc set errno on allocation failure. */ -#define HAVE_MALLOC_POSIX 1 - -/* Define to 1 if malloc-like functions do not allocate objects larger than - PTRDIFF_MAX bytes. */ -#define HAVE_MALLOC_PTRDIFF 1 - -/* Define to 1 if you have the header file. */ -/* #undef HAVE_MINIX_CONFIG_H */ - -/* Define to 1 on MSVC platforms that have the "invalid parameter handler" - concept. */ -/* #undef HAVE_MSVC_INVALID_PARAMETER_HANDLER */ - -/* Define to 1 if you have the `rawmemchr' function. */ -/* #undef HAVE_RAWMEMCHR */ - -/* Define to 1 if you have the header file. */ -/* #undef HAVE_SDKDDKVER_H */ - -/* Define to 1 if you have the `setdtablesize' function. */ -/* #undef HAVE_SETDTABLESIZE */ - -/* Define to 1 if 'sig_atomic_t' is a signed integer type. */ -/* #undef HAVE_SIGNED_SIG_ATOMIC_T */ - -/* Define to 1 if 'wchar_t' is a signed integer type. */ -/* #undef HAVE_SIGNED_WCHAR_T */ - -/* Define to 1 if 'wint_t' is a signed integer type. */ -/* #undef HAVE_SIGNED_WINT_T */ - -/* Define to 1 if you have the header file. */ -#define HAVE_STDBOOL_H 1 - -/* Define to 1 if you have the header file. */ -#define HAVE_STDCKDINT_H 1 - -/* Define to 1 if you have the header file. */ -#define HAVE_STDINT_H 1 - -/* Define to 1 if you have the header file. */ -#define HAVE_STDIO_H 1 - -/* Define to 1 if you have the header file. */ -#define HAVE_STDLIB_H 1 - -/* Define to 1 if you have the `strchrnul' function. */ -#define HAVE_STRCHRNUL 1 - -/* Define if you have `strerror_r'. */ -#define HAVE_STRERROR_R 1 - -/* Define to 1 if you have the header file. */ -#define HAVE_STRINGS_H 1 - -/* Define to 1 if you have the header file. */ -#define HAVE_STRING_H 1 - -/* Define to 1 if `st_atimensec' is a member of `struct stat'. */ -/* #undef HAVE_STRUCT_STAT_ST_ATIMENSEC */ - -/* Define to 1 if `st_atimespec.tv_nsec' is a member of `struct stat'. */ -#define HAVE_STRUCT_STAT_ST_ATIMESPEC_TV_NSEC 1 - -/* Define to 1 if `st_atim.st__tim.tv_nsec' is a member of `struct stat'. */ -/* #undef HAVE_STRUCT_STAT_ST_ATIM_ST__TIM_TV_NSEC */ - -/* Define to 1 if `st_atim.tv_nsec' is a member of `struct stat'. */ -/* #undef HAVE_STRUCT_STAT_ST_ATIM_TV_NSEC */ - -/* Define to 1 if `st_birthtimensec' is a member of `struct stat'. */ -/* #undef HAVE_STRUCT_STAT_ST_BIRTHTIMENSEC */ - -/* Define to 1 if `st_birthtimespec.tv_nsec' is a member of `struct stat'. */ -#define HAVE_STRUCT_STAT_ST_BIRTHTIMESPEC_TV_NSEC 1 - -/* Define to 1 if `st_birthtim.tv_nsec' is a member of `struct stat'. */ -/* #undef HAVE_STRUCT_STAT_ST_BIRTHTIM_TV_NSEC */ - -/* Define to 1 if you have the `strverscmp' function. */ -/* #undef HAVE_STRVERSCMP */ - -/* Define to 1 if you have the `symlink' function. */ -#define HAVE_SYMLINK 1 - -/* The toolchain supports aliases and .symver. */ -/* #undef HAVE_SYMVER_ALIAS_SUPPORT */ - -/* Define to 1 if you have the header file. */ -/* #undef HAVE_SYS_BITYPES_H */ - -/* Define to 1 if you have the header file. */ -/* #undef HAVE_SYS_INTTYPES_H */ - -/* Define to 1 if you have the header file. */ -#define HAVE_SYS_PARAM_H 1 - -/* Define to 1 if you have the header file. */ -#define HAVE_SYS_SOCKET_H 1 - -/* Define to 1 if you have the header file. */ -#define HAVE_SYS_STAT_H 1 - -/* Define to 1 if you have the header file. */ -#define HAVE_SYS_TIME_H 1 - -/* Define to 1 if you have the header file. */ -#define HAVE_SYS_TYPES_H 1 - -/* Define to 1 if you have the header file. */ -#define HAVE_UNISTD_H 1 - -/* Define to 1 if you have the header file. */ -/* #undef HAVE_UNISTRING_WOE32DLL_H */ - -/* Define to 1 if the system has the type 'unsigned long long int'. */ -#define HAVE_UNSIGNED_LONG_LONG_INT 1 - -/* Define if you have a global __progname variable */ -/* #undef HAVE_VAR___PROGNAME */ - -/* Define to 1 or 0, depending whether the compiler supports simple visibility - declarations. */ -#define HAVE_VISIBILITY 1 - -/* Define to 1 if you have the header file. */ -#define HAVE_WCHAR_H 1 - -/* Define to 1 if you have the header file. */ -/* #undef HAVE_WINSOCK2_H */ - -/* Define if you have the 'wint_t' type. */ -#define HAVE_WINT_T 1 - -/* Define to 1 if O_NOATIME works. */ -#define HAVE_WORKING_O_NOATIME 1 - -/* Define to 1 if O_NOFOLLOW works. */ -#define HAVE_WORKING_O_NOFOLLOW 1 - -/* Define to 1 if you have the `_set_invalid_parameter_handler' function. */ -/* #undef HAVE__SET_INVALID_PARAMETER_HANDLER */ - -/* Define to 1 if ctype.h defines __header_inline. */ -#define HAVE___HEADER_INLINE 1 - -/* Please see the Gnulib manual for how to use these macros. - - Suppress extern inline with HP-UX cc, as it appears to be broken; see - . - - Suppress extern inline with Sun C in standards-conformance mode, as it - mishandles inline functions that call each other. E.g., for 'inline void f - (void) { } inline void g (void) { f (); }', c99 incorrectly complains - 'reference to static identifier "f" in extern inline function'. - This bug was observed with Oracle Developer Studio 12.6 - (Sun C 5.15 SunOS_sparc 2017/05/30). - - Suppress extern inline (with or without __attribute__ ((__gnu_inline__))) - on configurations that mistakenly use 'static inline' to implement - functions or macros in standard C headers like . For example, - if isdigit is mistakenly implemented via a static inline function, - a program containing an extern inline function that calls isdigit - may not work since the C standard prohibits extern inline functions - from calling static functions (ISO C 99 section 6.7.4.(3). - This bug is known to occur on: - - OS X 10.8 and earlier; see: - https://lists.gnu.org/r/bug-gnulib/2012-12/msg00023.html - - DragonFly; see - http://muscles.dragonflybsd.org/bulk/clang-master-potential/20141111_102002/logs/ah-tty-0.3.12.log - - FreeBSD; see: - https://lists.gnu.org/r/bug-gnulib/2014-07/msg00104.html - - OS X 10.9 has a macro __header_inline indicating the bug is fixed for C and - for clang but remains for g++; see . - Assume DragonFly and FreeBSD will be similar. - - GCC 4.3 and above with -std=c99 or -std=gnu99 implements ISO C99 - inline semantics, unless -fgnu89-inline is used. It defines a macro - __GNUC_STDC_INLINE__ to indicate this situation or a macro - __GNUC_GNU_INLINE__ to indicate the opposite situation. - GCC 4.2 with -std=c99 or -std=gnu99 implements the GNU C inline - semantics but warns, unless -fgnu89-inline is used: - warning: C99 inline functions are not supported; using GNU89 - warning: to disable this warning use -fgnu89-inline or the gnu_inline function attribute - It defines a macro __GNUC_GNU_INLINE__ to indicate this situation. - */ -#if (((defined __APPLE__ && defined __MACH__) \ - || defined __DragonFly__ || defined __FreeBSD__) \ - && (defined HAVE___HEADER_INLINE \ - ? (defined __cplusplus && defined __GNUC_STDC_INLINE__ \ - && ! defined __clang__) \ - : ((! defined _DONT_USE_CTYPE_INLINE_ \ - && (defined __GNUC__ || defined __cplusplus)) \ - || (defined _FORTIFY_SOURCE && 0 < _FORTIFY_SOURCE \ - && defined __GNUC__ && ! defined __cplusplus)))) -# define _GL_EXTERN_INLINE_STDHEADER_BUG -#endif -#if ((__GNUC__ \ - ? (defined __GNUC_STDC_INLINE__ && __GNUC_STDC_INLINE__ \ - && !defined __PCC__) \ - : (199901L <= __STDC_VERSION__ \ - && !defined __HP_cc \ - && !defined __PGI \ - && !(defined __SUNPRO_C && __STDC__))) \ - && !defined _GL_EXTERN_INLINE_STDHEADER_BUG) -# define _GL_INLINE inline -# define _GL_EXTERN_INLINE extern inline -# define _GL_EXTERN_INLINE_IN_USE -#elif (2 < __GNUC__ + (7 <= __GNUC_MINOR__) && !defined __STRICT_ANSI__ \ - && !defined __PCC__ \ - && !defined _GL_EXTERN_INLINE_STDHEADER_BUG) -# if defined __GNUC_GNU_INLINE__ && __GNUC_GNU_INLINE__ - /* __gnu_inline__ suppresses a GCC 4.2 diagnostic. */ -# define _GL_INLINE extern inline __attribute__ ((__gnu_inline__)) -# else -# define _GL_INLINE extern inline -# endif -# define _GL_EXTERN_INLINE extern -# define _GL_EXTERN_INLINE_IN_USE -#else -# define _GL_INLINE _GL_UNUSED static -# define _GL_EXTERN_INLINE _GL_UNUSED static -#endif - -/* In GCC 4.6 (inclusive) to 5.1 (exclusive), - suppress bogus "no previous prototype for 'FOO'" - and "no previous declaration for 'FOO'" diagnostics, - when FOO is an inline function in the header; see - and - . */ -#if __GNUC__ == 4 && 6 <= __GNUC_MINOR__ -# if defined __GNUC_STDC_INLINE__ && __GNUC_STDC_INLINE__ -# define _GL_INLINE_HEADER_CONST_PRAGMA -# else -# define _GL_INLINE_HEADER_CONST_PRAGMA \ - _Pragma ("GCC diagnostic ignored \"-Wsuggest-attribute=const\"") -# endif -# define _GL_INLINE_HEADER_BEGIN \ - _Pragma ("GCC diagnostic push") \ - _Pragma ("GCC diagnostic ignored \"-Wmissing-prototypes\"") \ - _Pragma ("GCC diagnostic ignored \"-Wmissing-declarations\"") \ - _GL_INLINE_HEADER_CONST_PRAGMA -# define _GL_INLINE_HEADER_END \ - _Pragma ("GCC diagnostic pop") -#else -# define _GL_INLINE_HEADER_BEGIN -# define _GL_INLINE_HEADER_END -#endif - -/* Define to 1 if the compiler supports the keyword '__inline'. */ -#define HAVE___INLINE 1 - -/* Define as const if the declaration of iconv() needs const. */ -#define ICONV_CONST - -/* Define to a symbolic name denoting the flavor of iconv_open() - implementation. */ -/* #undef ICONV_FLAVOR */ - -/* Define to the sub-directory where libtool stores uninstalled libraries. */ -#define LT_OBJDIR ".libs/" - -/* Use GNU style printf and scanf. */ -#ifndef __USE_MINGW_ANSI_STDIO -# define __USE_MINGW_ANSI_STDIO 1 -#endif - - -/* Define to 1 on musl libc. */ -/* #undef MUSL_LIBC */ - -/* Define to 1 if open() fails to recognize a trailing slash. */ -/* #undef OPEN_TRAILING_SLASH_BUG */ - -/* Name of package */ -#define PACKAGE "libidn2" - -/* Define to the address where bug reports for this package should be sent. */ -#define PACKAGE_BUGREPORT "help-libidn@gnu.org" - -/* Define to the full name of this package. */ -#define PACKAGE_NAME "Libidn2" - -/* String identifying the packager of this software */ -/* #undef PACKAGE_PACKAGER */ - -/* Packager info for bug reports (URL/e-mail/...) */ -/* #undef PACKAGE_PACKAGER_BUG_REPORTS */ - -/* Packager-specific version information */ -/* #undef PACKAGE_PACKAGER_VERSION */ - -/* Define to the full name and version of this package. */ -#define PACKAGE_STRING "Libidn2 2.3.8" - -/* Define to the one symbol short name of this package. */ -#define PACKAGE_TARNAME "libidn2" - -/* Define to the home page for this package. */ -#define PACKAGE_URL "https://www.gnu.org/software/libidn/#libidn2" - -/* Define to the version of this package. */ -#define PACKAGE_VERSION "2.3.8" - -/* Define to the type that is the result of default argument promotions of - type mode_t. */ -#define PROMOTED_MODE_T int - -/* Define to l, ll, u, ul, ull, etc., as suitable for constants of type - 'ptrdiff_t'. */ -/* #undef PTRDIFF_T_SUFFIX */ - -/* Define to 1 if stat needs help when passed a file name with a trailing - slash */ -/* #undef REPLACE_FUNC_STAT_FILE */ - -/* Define to 1 if strerror(0) does not return a message implying success. */ -#define REPLACE_STRERROR_0 1 - -/* Define to l, ll, u, ul, ull, etc., as suitable for constants of type - 'sig_atomic_t'. */ -/* #undef SIG_ATOMIC_T_SUFFIX */ - -/* Define to l, ll, u, ul, ull, etc., as suitable for constants of type - 'size_t'. */ -/* #undef SIZE_T_SUFFIX */ - -/* If using the C implementation of alloca, define if you know the - direction of stack growth for your system; otherwise it will be - automatically deduced at runtime. - STACK_DIRECTION > 0 => grows toward higher addresses - STACK_DIRECTION < 0 => grows toward lower addresses - STACK_DIRECTION = 0 => direction of growth unknown */ -/* #undef STACK_DIRECTION */ - -/* Define to 1 if the `S_IS*' macros in do not work properly. */ -/* #undef STAT_MACROS_BROKEN */ - -/* Define to 1 if all of the C90 standard headers exist (not just the ones - required in a freestanding environment). This macro is provided for - backward compatibility; new code need not use it. */ -#define STDC_HEADERS 1 - -/* Define to 1 if strerror_r returns char *. */ -/* #undef STRERROR_R_CHAR_P */ - -/* Define to 1 if the type of the st_atim member of a struct stat is struct - timespec. */ -/* #undef TYPEOF_STRUCT_STAT_ST_ATIM_IS_STRUCT_TIMESPEC */ - -/* Define to enable the declarations of ISO C 23 Annex K types and functions. */ -#if !(defined __STDC_WANT_LIB_EXT1__ && __STDC_WANT_LIB_EXT1__) -#undef/**/__STDC_WANT_LIB_EXT1__ -#define __STDC_WANT_LIB_EXT1__ 1 -#endif - - -/* Enable extensions on AIX 3, Interix. */ -#ifndef _ALL_SOURCE -# define _ALL_SOURCE 1 -#endif -/* Enable general extensions on macOS. */ -#ifndef _DARWIN_C_SOURCE -# define _DARWIN_C_SOURCE 1 -#endif -/* Enable general extensions on Solaris. */ -#ifndef __EXTENSIONS__ -# define __EXTENSIONS__ 1 -#endif -/* Enable GNU extensions on systems that have them. */ -#ifndef _GNU_SOURCE -# define _GNU_SOURCE 1 -#endif -/* Enable X/Open compliant socket functions that do not require linking - with -lxnet on HP-UX 11.11. */ -#ifndef _HPUX_ALT_XOPEN_SOCKET_API -# define _HPUX_ALT_XOPEN_SOCKET_API 1 -#endif -/* Identify the host operating system as Minix. - This macro does not affect the system headers' behavior. - A future release of Autoconf may stop defining this macro. */ -#ifndef _MINIX -/* # undef _MINIX */ -#endif -/* Enable general extensions on NetBSD. - Enable NetBSD compatibility extensions on Minix. */ -#ifndef _NETBSD_SOURCE -# define _NETBSD_SOURCE 1 -#endif -/* Enable OpenBSD compatibility extensions on NetBSD. - Oddly enough, this does nothing on OpenBSD. */ -#ifndef _OPENBSD_SOURCE -# define _OPENBSD_SOURCE 1 -#endif -/* Define to 1 if needed for POSIX-compatible behavior. */ -#ifndef _POSIX_SOURCE -/* # undef _POSIX_SOURCE */ -#endif -/* Define to 2 if needed for POSIX-compatible behavior. */ -#ifndef _POSIX_1_SOURCE -/* # undef _POSIX_1_SOURCE */ -#endif -/* Enable POSIX-compatible threading on Solaris. */ -#ifndef _POSIX_PTHREAD_SEMANTICS -# define _POSIX_PTHREAD_SEMANTICS 1 -#endif -/* Enable extensions specified by ISO/IEC TS 18661-5:2014. */ -#ifndef __STDC_WANT_IEC_60559_ATTRIBS_EXT__ -# define __STDC_WANT_IEC_60559_ATTRIBS_EXT__ 1 -#endif -/* Enable extensions specified by ISO/IEC TS 18661-1:2014. */ -#ifndef __STDC_WANT_IEC_60559_BFP_EXT__ -# define __STDC_WANT_IEC_60559_BFP_EXT__ 1 -#endif -/* Enable extensions specified by ISO/IEC TS 18661-2:2015. */ -#ifndef __STDC_WANT_IEC_60559_DFP_EXT__ -# define __STDC_WANT_IEC_60559_DFP_EXT__ 1 -#endif -/* Enable extensions specified by C23 Annex F. */ -#ifndef __STDC_WANT_IEC_60559_EXT__ -# define __STDC_WANT_IEC_60559_EXT__ 1 -#endif -/* Enable extensions specified by ISO/IEC TS 18661-4:2015. */ -#ifndef __STDC_WANT_IEC_60559_FUNCS_EXT__ -# define __STDC_WANT_IEC_60559_FUNCS_EXT__ 1 -#endif -/* Enable extensions specified by C23 Annex H and ISO/IEC TS 18661-3:2015. */ -#ifndef __STDC_WANT_IEC_60559_TYPES_EXT__ -# define __STDC_WANT_IEC_60559_TYPES_EXT__ 1 -#endif -/* Enable extensions specified by ISO/IEC TR 24731-2:2010. */ -#ifndef __STDC_WANT_LIB_EXT2__ -# define __STDC_WANT_LIB_EXT2__ 1 -#endif -/* Enable extensions specified by ISO/IEC 24747:2009. */ -#ifndef __STDC_WANT_MATH_SPEC_FUNCS__ -# define __STDC_WANT_MATH_SPEC_FUNCS__ 1 -#endif -/* Enable extensions on HP NonStop. */ -#ifndef _TANDEM_SOURCE -# define _TANDEM_SOURCE 1 -#endif -/* Enable X/Open extensions. Define to 500 only if necessary - to make mbstate_t available. */ -#ifndef _XOPEN_SOURCE -/* # undef _XOPEN_SOURCE */ -#endif - - -/* Version number of package */ -#define VERSION "2.3.8" - -/* Define to l, ll, u, ul, ull, etc., as suitable for constants of type - 'wchar_t'. */ -/* #undef WCHAR_T_SUFFIX */ - -/* Define to l, ll, u, ul, ull, etc., as suitable for constants of type - 'wint_t'. */ -/* #undef WINT_T_SUFFIX */ - -/* Number of bits in a file offset, on hosts where this is settable. */ -/* #undef _FILE_OFFSET_BITS */ - -/* True if the compiler says it groks GNU C version MAJOR.MINOR. - Except that - - clang groks GNU C 4.2, even on Windows, where it does not define - __GNUC__. - - The OpenMandriva-modified clang compiler pretends that it groks - GNU C version 13.1, but it doesn't: It does not support - __attribute__ ((__malloc__ (f, i))), nor does it support - __attribute__ ((__warning__ (message))) on a function redeclaration. - - Users can make clang lie as well, through the -fgnuc-version option. */ -#if defined __GNUC__ && defined __GNUC_MINOR__ && !defined __clang__ -# define _GL_GNUC_PREREQ(major, minor) \ - ((major) < __GNUC__ + ((minor) <= __GNUC_MINOR__)) -#elif defined __clang__ - /* clang really only groks GNU C 4.2. */ -# define _GL_GNUC_PREREQ(major, minor) \ - ((major) < 4 + ((minor) <= 2)) -#else -# define _GL_GNUC_PREREQ(major, minor) 0 -#endif - - -/* Define to enable the declarations of ISO C 11 types and functions. */ -/* #undef _ISOC11_SOURCE */ - -/* Define to 1 on platforms where this makes off_t a 64-bit type. */ -/* #undef _LARGE_FILES */ - -/* Define so that AIX headers are more compatible with GNU/Linux. */ -#define _LINUX_SOURCE_COMPAT 1 - -/* The _Noreturn keyword of C11. */ -#ifndef _Noreturn -# if (defined __cplusplus \ - && ((201103 <= __cplusplus && !(__GNUC__ == 4 && __GNUC_MINOR__ == 7)) \ - || (defined _MSC_VER && 1900 <= _MSC_VER)) \ - && 0) - /* [[noreturn]] is not practically usable, because with it the syntax - extern _Noreturn void func (...); - would not be valid; such a declaration would only be valid with 'extern' - and '_Noreturn' swapped, or without the 'extern' keyword. However, some - AIX system header files and several gnulib header files use precisely - this syntax with 'extern'. */ -# define _Noreturn [[noreturn]] -# elif (defined __clang__ && __clang_major__ < 16 \ - && defined _GL_WORK_AROUND_LLVM_BUG_59792) - /* Compile with -D_GL_WORK_AROUND_LLVM_BUG_59792 to work around - that rare LLVM bug, though you may get many false-alarm warnings. */ -# define _Noreturn -# elif ((!defined __cplusplus || defined __clang__) \ - && (201112 <= (defined __STDC_VERSION__ ? __STDC_VERSION__ : 0) \ - || (!defined __STRICT_ANSI__ \ - && (_GL_GNUC_PREREQ (4, 7) \ - || (defined __apple_build_version__ \ - ? 6000000 <= __apple_build_version__ \ - : 3 < __clang_major__ + (5 <= __clang_minor__)))))) - /* _Noreturn works as-is. */ -# elif _GL_GNUC_PREREQ (2, 8) || defined __clang__ || 0x5110 <= __SUNPRO_C -# define _Noreturn __attribute__ ((__noreturn__)) -# elif 1200 <= (defined _MSC_VER ? _MSC_VER : 0) -# define _Noreturn __declspec (noreturn) -# else -# define _Noreturn -# endif -#endif - - -/* Number of bits in time_t, on hosts where this is settable. */ -/* #undef _TIME_BITS */ - -/* For standard stat data types on VMS. */ -#define _USE_STD_STAT 1 - -/* Define to rpl_ if the getopt replacement functions and variables should be - used. */ -#define __GETOPT_PREFIX rpl_ - -/* Define to 1 on platforms where this makes time_t a 64-bit type. */ -/* #undef __MINGW_USE_VC2005_COMPAT */ - -/* Define to 1 if the system predates C++11. */ -/* #undef __STDC_CONSTANT_MACROS */ - -/* Define to 1 if the system predates C++11. */ -/* #undef __STDC_LIMIT_MACROS */ - -/* Define to 1 if C does not support variable-length arrays, and if the - compiler does not already define this. */ -/* #undef __STDC_NO_VLA__ */ - -/* The _GL_ASYNC_SAFE marker should be attached to functions that are - signal handlers (for signals other than SIGABRT, SIGPIPE) or can be - invoked from such signal handlers. Such functions have some restrictions: - * All functions that it calls should be marked _GL_ASYNC_SAFE as well, - or should be listed as async-signal-safe in POSIX - - section 2.4.3. Note that malloc(), sprintf(), and fwrite(), in - particular, are NOT async-signal-safe. - * All memory locations (variables and struct fields) that these functions - access must be marked 'volatile'. This holds for both read and write - accesses. Otherwise the compiler might optimize away stores to and - reads from such locations that occur in the program, depending on its - data flow analysis. For example, when the program contains a loop - that is intended to inspect a variable set from within a signal handler - while (!signal_occurred) - ; - the compiler is allowed to transform this into an endless loop if the - variable 'signal_occurred' is not declared 'volatile'. - Additionally, recall that: - * A signal handler should not modify errno (except if it is a handler - for a fatal signal and ends by raising the same signal again, thus - provoking the termination of the process). If it invokes a function - that may clobber errno, it needs to save and restore the value of - errno. */ -#define _GL_ASYNC_SAFE - - -/* Attributes. */ -/* Define _GL_HAS_ATTRIBUTE only once, because on FreeBSD, with gcc < 5, if - gets included once again after , __has_attribute(x) - expands to 0 always, and redefining _GL_HAS_ATTRIBUTE would turn off all - attributes. */ -#ifndef _GL_HAS_ATTRIBUTE -# if (defined __has_attribute \ - && (!defined __clang_minor__ \ - || (defined __apple_build_version__ \ - ? 7000000 <= __apple_build_version__ \ - : 5 <= __clang_major__))) -# define _GL_HAS_ATTRIBUTE(attr) __has_attribute (__##attr##__) -# else -# define _GL_HAS_ATTRIBUTE(attr) _GL_ATTR_##attr -# define _GL_ATTR_alloc_size _GL_GNUC_PREREQ (4, 3) -# define _GL_ATTR_always_inline _GL_GNUC_PREREQ (3, 2) -# define _GL_ATTR_artificial _GL_GNUC_PREREQ (4, 3) -# define _GL_ATTR_cold _GL_GNUC_PREREQ (4, 3) -# define _GL_ATTR_const _GL_GNUC_PREREQ (2, 95) -# define _GL_ATTR_deprecated _GL_GNUC_PREREQ (3, 1) -# define _GL_ATTR_diagnose_if 0 -# define _GL_ATTR_error _GL_GNUC_PREREQ (4, 3) -# define _GL_ATTR_externally_visible _GL_GNUC_PREREQ (4, 1) -# define _GL_ATTR_fallthrough _GL_GNUC_PREREQ (7, 0) -# define _GL_ATTR_format _GL_GNUC_PREREQ (2, 7) -# define _GL_ATTR_leaf _GL_GNUC_PREREQ (4, 6) -# define _GL_ATTR_malloc _GL_GNUC_PREREQ (3, 0) -# ifdef _ICC -# define _GL_ATTR_may_alias 0 -# else -# define _GL_ATTR_may_alias _GL_GNUC_PREREQ (3, 3) -# endif -# define _GL_ATTR_noinline _GL_GNUC_PREREQ (3, 1) -# define _GL_ATTR_nonnull _GL_GNUC_PREREQ (3, 3) -# define _GL_ATTR_nonstring _GL_GNUC_PREREQ (8, 0) -# define _GL_ATTR_nothrow _GL_GNUC_PREREQ (3, 3) -# define _GL_ATTR_packed _GL_GNUC_PREREQ (2, 7) -# define _GL_ATTR_pure _GL_GNUC_PREREQ (2, 96) -# define _GL_ATTR_reproducible 0 /* not yet supported, as of GCC 14 */ -# define _GL_ATTR_returns_nonnull _GL_GNUC_PREREQ (4, 9) -# define _GL_ATTR_sentinel _GL_GNUC_PREREQ (4, 0) -# define _GL_ATTR_unsequenced 0 /* not yet supported, as of GCC 14 */ -# define _GL_ATTR_unused _GL_GNUC_PREREQ (2, 7) -# define _GL_ATTR_warn_unused_result _GL_GNUC_PREREQ (3, 4) -# endif -#endif - -/* Use __has_c_attribute if available. However, do not use with - pre-C23 GCC, which can issue false positives if -Wpedantic. */ -#if (defined __has_c_attribute \ - && ! (_GL_GNUC_PREREQ (4, 6) \ - && (defined __STDC_VERSION__ ? __STDC_VERSION__ : 0) <= 201710)) -# define _GL_HAVE___HAS_C_ATTRIBUTE 1 -#else -# define _GL_HAVE___HAS_C_ATTRIBUTE 0 -#endif - -/* Attributes in bracket syntax [[...]] vs. attributes in __attribute__((...)) - syntax, in function declarations. There are two problems here. - (Last tested with gcc/g++ 14 and clang/clang++ 18.) - - 1) We want that the _GL_ATTRIBUTE_* can be cumulated on the same declaration - in any order. - =========================== foo.c = foo.cc =========================== - __attribute__ ((__deprecated__)) [[__nodiscard__]] int bar1 (int); - [[__nodiscard__]] __attribute__ ((__deprecated__)) int bar2 (int); - ====================================================================== - This gives a syntax error - - in C mode with gcc - , and - - in C++ mode with clang++ version < 16, and - - in C++ mode, inside extern "C" {}, still in newer clang++ versions - . - */ -/* Define if, in a function declaration, the attributes in bracket syntax - [[...]] must come before the attributes in __attribute__((...)) syntax. - If this is defined, it is best to avoid the bracket syntax, so that the - various _GL_ATTRIBUTE_* can be cumulated on the same declaration in any - order. */ -#ifdef __cplusplus -# if defined __clang__ -# define _GL_BRACKET_BEFORE_ATTRIBUTE 1 -# endif -#else -# if defined __GNUC__ && !defined __clang__ -# define _GL_BRACKET_BEFORE_ATTRIBUTE 1 -# endif -#endif -/* - 2) We want that the _GL_ATTRIBUTE_* can be placed in a declaration - - without 'extern', in C as well as in C++, - - with 'extern', in C, - - with 'extern "C"', in C++ - in the same position. That is, we don't want to be forced to use a - macro which arranges for the attribute to come before 'extern' in - one case and after 'extern' in the other case, because such a macro - would make the source code of .h files pretty ugly. - =========================== foo.c = foo.cc =========================== - #ifdef __cplusplus - # define CC "C" - #else - # define CC - #endif - - #define ND [[__nodiscard__]] - #define WUR __attribute__((__warn_unused_result__)) - - #ifdef __cplusplus - extern "C" { - #endif - // gcc clang g++ clang++ - - ND int foo (int); - int ND foo (int); // warn error warn error - int foo ND (int); - int foo (int) ND; // warn error warn error - - WUR int foo (int); - int WUR foo (int); - int fo1 WUR (int); // error error error error - int foo (int) WUR; - - #ifdef __cplusplus - } - #endif - - // gcc clang g++ clang++ - - ND extern CC int foo (int); // error error - extern CC ND int foo (int); // error error - extern CC int ND foo (int); // warn error warn error - extern CC int foo ND (int); - extern CC int foo (int) ND; // warn error warn error - - WUR extern CC int foo (int); // warn - extern CC WUR int foo (int); - extern CC int WUR foo (int); - extern CC int foo WUR (int); // error error error error - extern CC int foo (int) WUR; - - ND EXTERN_C_FUNC int foo (int); // error error - EXTERN_C_FUNC ND int foo (int); - EXTERN_C_FUNC int ND foo (int); // warn error warn error - EXTERN_C_FUNC int foo ND (int); - EXTERN_C_FUNC int foo (int) ND; // warn error warn error - - WUR EXTERN_C_FUNC int foo (int); // warn - EXTERN_C_FUNC WUR int foo (int); - EXTERN_C_FUNC int WUR foo (int); - EXTERN_C_FUNC int fo2 WUR (int); // error error error error - EXTERN_C_FUNC int foo (int) WUR; - ====================================================================== - So, if we insist on using the 'extern' keyword ('extern CC' idiom): - * If _GL_ATTRIBUTE_* expands to bracket syntax [[...]] - in both C and C++, there is one available position: - - between the function name and the parameter list. - * If _GL_ATTRIBUTE_* expands to __attribute__((...)) syntax - in both C and C++, there are several available positions: - - before the return type, - - between return type and function name, - - at the end of the declaration. - * If _GL_ATTRIBUTE_* expands to bracket syntax [[...]] in C and to - __attribute__((...)) syntax in C++, there is no available position: - it would need to come before 'extern' in C but after 'extern "C"' - in C++. - * If _GL_ATTRIBUTE_* expands to __attribute__((...)) syntax in C and - to bracket syntax [[...]] in C++, there is one available position: - - before the return type. - Whereas, if we use the 'EXTERN_C_FUNC' idiom, which conditionally - omits the 'extern' keyword: - * If _GL_ATTRIBUTE_* expands to bracket syntax [[...]] - in both C and C++, there are two available positions: - - before the return type, - - between the function name and the parameter list. - * If _GL_ATTRIBUTE_* expands to __attribute__((...)) syntax - in both C and C++, there are several available positions: - - before the return type, - - between return type and function name, - - at the end of the declaration. - * If _GL_ATTRIBUTE_* expands to bracket syntax [[...]] in C and to - __attribute__((...)) syntax in C++, there is one available position: - - before the return type. - * If _GL_ATTRIBUTE_* expands to __attribute__((...)) syntax in C and - to bracket syntax [[...]] in C++, there is one available position: - - before the return type. - The best choice is therefore to use the 'EXTERN_C_FUNC' idiom and - put the attributes before the return type. This works regardless - to what the _GL_ATTRIBUTE_* macros expand. - */ - -/* Attributes in bracket syntax [[...]] vs. attributes in __attribute__((...)) - syntax, in static/inline function definitions. - - There are similar constraints as for function declarations. However, here, - we cannot omit the storage-class specifier. Therefore, the following rule - applies: - * The macros - _GL_ATTRIBUTE_CONST - _GL_ATTRIBUTE_DEPRECATED - _GL_ATTRIBUTE_MAYBE_UNUSED - _GL_ATTRIBUTE_NODISCARD - _GL_ATTRIBUTE_PURE - _GL_ATTRIBUTE_REPRODUCIBLE - _GL_ATTRIBUTE_UNSEQUENCED - which may expand to bracket syntax [[...]], must come first, before the - storage-class specifier. - * Other _GL_ATTRIBUTE_* macros, that expand to __attribute__((...)) syntax, - are better placed between the storage-class specifier and the return - type. - */ - -/* Attributes in bracket syntax [[...]] vs. attributes in __attribute__((...)) - syntax, in variable declarations. - - At which position can they be placed? - (Last tested with gcc/g++ 14 and clang/clang++ 18.) - - =========================== foo.c = foo.cc =========================== - #ifdef __cplusplus - # define CC "C" - #else - # define CC - #endif - - #define BD [[__deprecated__]] - #define AD __attribute__ ((__deprecated__)) - - // gcc clang g++ clang++ - - BD extern CC int var; // error error - extern CC BD int var; // error error - extern CC int BD var; // warn error warn error - extern CC int var BD; - - AD extern CC int var; // warn - extern CC AD int var; - extern CC int AD var; - extern CC int var AD; - - BD extern CC int z[]; // error error - extern CC BD int z[]; // error error - extern CC int BD z[]; // warn error warn error - extern CC int z1 BD []; - extern CC int z[] BD; // warn error error - - AD extern CC int z[]; // warn - extern CC AD int z[]; - extern CC int AD z[]; - extern CC int z2 AD []; // error error error error - extern CC int z[] AD; - ====================================================================== - - * For non-array variables, the only good position is after the variable name, - that is, at the end of the declaration. - * For array variables, you will need to distinguish C and C++: - - In C, before the 'extern' keyword. - - In C++, between the 'extern "C"' and the variable's type. - */ - -/* _GL_ATTRIBUTE_ALLOC_SIZE ((N)) declares that the Nth argument of the function - is the size of the returned memory block. - _GL_ATTRIBUTE_ALLOC_SIZE ((M, N)) declares that the Mth argument multiplied - by the Nth argument of the function is the size of the returned memory block. - */ -/* Applies to: functions, pointer to functions, function types. */ -#ifndef _GL_ATTRIBUTE_ALLOC_SIZE -# if _GL_HAS_ATTRIBUTE (alloc_size) -# define _GL_ATTRIBUTE_ALLOC_SIZE(args) __attribute__ ((__alloc_size__ args)) -# else -# define _GL_ATTRIBUTE_ALLOC_SIZE(args) -# endif -#endif - -/* _GL_ATTRIBUTE_ALWAYS_INLINE tells that the compiler should always inline the - function and report an error if it cannot do so. */ -/* Applies to: functions. */ -#ifndef _GL_ATTRIBUTE_ALWAYS_INLINE -# if _GL_HAS_ATTRIBUTE (always_inline) -# define _GL_ATTRIBUTE_ALWAYS_INLINE __attribute__ ((__always_inline__)) -# else -# define _GL_ATTRIBUTE_ALWAYS_INLINE -# endif -#endif - -/* _GL_ATTRIBUTE_ARTIFICIAL declares that the function is not important to show - in stack traces when debugging. The compiler should omit the function from - stack traces. */ -/* Applies to: functions. */ -#ifndef _GL_ATTRIBUTE_ARTIFICIAL -# if _GL_HAS_ATTRIBUTE (artificial) -# define _GL_ATTRIBUTE_ARTIFICIAL __attribute__ ((__artificial__)) -# else -# define _GL_ATTRIBUTE_ARTIFICIAL -# endif -#endif - -/* _GL_ATTRIBUTE_COLD declares that the function is rarely executed. */ -/* Applies to: functions. */ -/* Avoid __attribute__ ((cold)) on MinGW; see thread starting at - . - Also, Oracle Studio 12.6 requires 'cold' not '__cold__'. */ -#ifndef _GL_ATTRIBUTE_COLD -# if _GL_HAS_ATTRIBUTE (cold) && !defined __MINGW32__ -# ifndef __SUNPRO_C -# define _GL_ATTRIBUTE_COLD __attribute__ ((__cold__)) -# else -# define _GL_ATTRIBUTE_COLD __attribute__ ((cold)) -# endif -# else -# define _GL_ATTRIBUTE_COLD -# endif -#endif - -/* _GL_ATTRIBUTE_CONST declares: - It is OK for a compiler to move calls to the function and to omit - calls to the function if another call has the same arguments or the - result is not used. - This attribute is safe for a function that neither depends on - nor affects state, and always returns exactly once - - e.g., does not raise an exception, call longjmp, or loop forever. - (This attribute is stricter than _GL_ATTRIBUTE_PURE because the - function cannot observe state. It is stricter than - _GL_ATTRIBUTE_UNSEQUENCED because the function must return exactly - once and cannot depend on state addressed by its arguments.) */ -/* Applies to: functions. */ -#ifndef _GL_ATTRIBUTE_CONST -# if _GL_HAS_ATTRIBUTE (const) -# define _GL_ATTRIBUTE_CONST __attribute__ ((__const__)) -# else -# define _GL_ATTRIBUTE_CONST _GL_ATTRIBUTE_UNSEQUENCED -# endif -#endif - -/* _GL_ATTRIBUTE_DEALLOC (F, I) declares that the function returns pointers - that can be freed by passing them as the Ith argument to the - function F. - _GL_ATTRIBUTE_DEALLOC_FREE declares that the function returns pointers that - can be freed via 'free'; it can be used only after declaring 'free'. */ -/* Applies to: functions. Cannot be used on inline functions. */ -#ifndef _GL_ATTRIBUTE_DEALLOC -# if _GL_GNUC_PREREQ (11, 0) -# define _GL_ATTRIBUTE_DEALLOC(f, i) __attribute__ ((__malloc__ (f, i))) -# else -# define _GL_ATTRIBUTE_DEALLOC(f, i) -# endif -#endif -/* If gnulib's or has already defined this macro, continue - to use this earlier definition, since may not have been included - yet. */ -#ifndef _GL_ATTRIBUTE_DEALLOC_FREE -# if defined __cplusplus && defined __GNUC__ && !defined __clang__ -/* Work around GCC bug */ -# define _GL_ATTRIBUTE_DEALLOC_FREE \ - _GL_ATTRIBUTE_DEALLOC ((void (*) (void *)) free, 1) -# else -# define _GL_ATTRIBUTE_DEALLOC_FREE \ - _GL_ATTRIBUTE_DEALLOC (free, 1) -# endif -#endif - -/* _GL_ATTRIBUTE_DEPRECATED: Declares that an entity is deprecated. - The compiler may warn if the entity is used. */ -/* Applies to: - - function, variable, - - struct, union, struct/union member, - - enumeration, enumeration item, - - typedef, - in C++ also: namespace, class, template specialization. */ -#ifndef _GL_ATTRIBUTE_DEPRECATED -# ifndef _GL_BRACKET_BEFORE_ATTRIBUTE -# if _GL_HAVE___HAS_C_ATTRIBUTE -# if __has_c_attribute (__deprecated__) -# define _GL_ATTRIBUTE_DEPRECATED [[__deprecated__]] -# endif -# endif -# endif -# if !defined _GL_ATTRIBUTE_DEPRECATED && _GL_HAS_ATTRIBUTE (deprecated) -# define _GL_ATTRIBUTE_DEPRECATED __attribute__ ((__deprecated__)) -# endif -# ifndef _GL_ATTRIBUTE_DEPRECATED -# define _GL_ATTRIBUTE_DEPRECATED -# endif -#endif - -/* _GL_ATTRIBUTE_ERROR(msg) requests an error if a function is called and - the function call is not optimized away. - _GL_ATTRIBUTE_WARNING(msg) requests a warning if a function is called and - the function call is not optimized away. */ -/* Applies to: functions. */ -#if !(defined _GL_ATTRIBUTE_ERROR && defined _GL_ATTRIBUTE_WARNING) -# if _GL_HAS_ATTRIBUTE (error) -# define _GL_ATTRIBUTE_ERROR(msg) __attribute__ ((__error__ (msg))) -# define _GL_ATTRIBUTE_WARNING(msg) __attribute__ ((__warning__ (msg))) -# elif _GL_HAS_ATTRIBUTE (diagnose_if) -# define _GL_ATTRIBUTE_ERROR(msg) __attribute__ ((__diagnose_if__ (1, msg, "error"))) -# define _GL_ATTRIBUTE_WARNING(msg) __attribute__ ((__diagnose_if__ (1, msg, "warning"))) -# else -# define _GL_ATTRIBUTE_ERROR(msg) -# define _GL_ATTRIBUTE_WARNING(msg) -# endif -#endif - -/* _GL_ATTRIBUTE_EXTERNALLY_VISIBLE declares that the entity should remain - visible to debuggers etc., even with '-fwhole-program'. */ -/* Applies to: functions, variables. */ -#ifndef _GL_ATTRIBUTE_EXTERNALLY_VISIBLE -# if _GL_HAS_ATTRIBUTE (externally_visible) -# define _GL_ATTRIBUTE_EXTERNALLY_VISIBLE __attribute__ ((externally_visible)) -# else -# define _GL_ATTRIBUTE_EXTERNALLY_VISIBLE -# endif -#endif - -/* _GL_ATTRIBUTE_FALLTHROUGH declares that it is not a programming mistake if - the control flow falls through to the immediately following 'case' or - 'default' label. The compiler should not warn in this case. */ -/* Applies to: Empty statement (;), inside a 'switch' statement. */ -/* Always expands to something. */ -#ifndef _GL_ATTRIBUTE_FALLTHROUGH -# if _GL_HAVE___HAS_C_ATTRIBUTE -# if __has_c_attribute (__fallthrough__) -# define _GL_ATTRIBUTE_FALLTHROUGH [[__fallthrough__]] -# endif -# endif -# if !defined _GL_ATTRIBUTE_FALLTHROUGH && _GL_HAS_ATTRIBUTE (fallthrough) -# define _GL_ATTRIBUTE_FALLTHROUGH __attribute__ ((__fallthrough__)) -# endif -# ifndef _GL_ATTRIBUTE_FALLTHROUGH -# define _GL_ATTRIBUTE_FALLTHROUGH ((void) 0) -# endif -#endif - -/* _GL_ATTRIBUTE_FORMAT ((ARCHETYPE, STRING-INDEX, FIRST-TO-CHECK)) - declares that the STRING-INDEXth function argument is a format string of - style ARCHETYPE, which is one of: - printf, gnu_printf - scanf, gnu_scanf, - strftime, gnu_strftime, - strfmon, - or the same thing prefixed and suffixed with '__'. - If FIRST-TO-CHECK is not 0, arguments starting at FIRST-TO_CHECK - are suitable for the format string. */ -/* Applies to: functions. */ -#ifndef _GL_ATTRIBUTE_FORMAT -# if _GL_HAS_ATTRIBUTE (format) -# define _GL_ATTRIBUTE_FORMAT(spec) __attribute__ ((__format__ spec)) -# else -# define _GL_ATTRIBUTE_FORMAT(spec) -# endif -#endif - -/* _GL_ATTRIBUTE_LEAF declares that if the function is called from some other - compilation unit, it executes code from that unit only by return or by - exception handling. This declaration lets the compiler optimize that unit - more aggressively. */ -/* Applies to: functions. */ -#ifndef _GL_ATTRIBUTE_LEAF -# if _GL_HAS_ATTRIBUTE (leaf) -# define _GL_ATTRIBUTE_LEAF __attribute__ ((__leaf__)) -# else -# define _GL_ATTRIBUTE_LEAF -# endif -#endif - -/* _GL_ATTRIBUTE_MALLOC declares that the function returns a pointer to freshly - allocated memory. */ -/* Applies to: functions. */ -#ifndef _GL_ATTRIBUTE_MALLOC -# if _GL_HAS_ATTRIBUTE (malloc) -# define _GL_ATTRIBUTE_MALLOC __attribute__ ((__malloc__)) -# else -# define _GL_ATTRIBUTE_MALLOC -# endif -#endif - -/* _GL_ATTRIBUTE_MAY_ALIAS declares that pointers to the type may point to the - same storage as pointers to other types. Thus this declaration disables - strict aliasing optimization. */ -/* Applies to: types. */ -/* Oracle Studio 12.6 mishandles may_alias despite __has_attribute OK. */ -#ifndef _GL_ATTRIBUTE_MAY_ALIAS -# if _GL_HAS_ATTRIBUTE (may_alias) && !defined __SUNPRO_C -# define _GL_ATTRIBUTE_MAY_ALIAS __attribute__ ((__may_alias__)) -# else -# define _GL_ATTRIBUTE_MAY_ALIAS -# endif -#endif - -/* _GL_ATTRIBUTE_MAYBE_UNUSED declares that it is not a programming mistake if - the entity is not used. The compiler should not warn if the entity is not - used. */ -/* Applies to: - - function, variable, - - struct, union, struct/union member, - - enumeration, enumeration item, - - typedef, - in C++ also: class. */ -/* In C++ and C23, this is spelled [[__maybe_unused__]]. - GCC's syntax is __attribute__ ((__unused__)). - clang supports both syntaxes. Except that with clang ≥ 6, < 10, in C++ mode, - __has_c_attribute (__maybe_unused__) yields true but the use of - [[__maybe_unused__]] nevertheless produces a warning. */ -#ifndef _GL_ATTRIBUTE_MAYBE_UNUSED -# ifndef _GL_BRACKET_BEFORE_ATTRIBUTE -# if defined __clang__ && defined __cplusplus -# if !defined __apple_build_version__ && __clang_major__ >= 10 -# define _GL_ATTRIBUTE_MAYBE_UNUSED [[__maybe_unused__]] -# endif -# elif _GL_HAVE___HAS_C_ATTRIBUTE -# if __has_c_attribute (__maybe_unused__) -# define _GL_ATTRIBUTE_MAYBE_UNUSED [[__maybe_unused__]] -# endif -# endif -# endif -# ifndef _GL_ATTRIBUTE_MAYBE_UNUSED -# define _GL_ATTRIBUTE_MAYBE_UNUSED _GL_ATTRIBUTE_UNUSED -# endif -#endif -/* Alternative spelling of this macro, for convenience and for - compatibility with glibc/include/libc-symbols.h. */ -#define _GL_UNUSED _GL_ATTRIBUTE_MAYBE_UNUSED -/* Earlier spellings of this macro. */ -#define _UNUSED_PARAMETER_ _GL_ATTRIBUTE_MAYBE_UNUSED - -/* _GL_ATTRIBUTE_NODISCARD declares that the caller of the function should not - discard the return value. The compiler may warn if the caller does not use - the return value, unless the caller uses something like ignore_value. */ -/* Applies to: function, enumeration, class. */ -#ifndef _GL_ATTRIBUTE_NODISCARD -# ifndef _GL_BRACKET_BEFORE_ATTRIBUTE -# if defined __clang__ && defined __cplusplus - /* With clang up to 15.0.6 (at least), in C++ mode, [[__nodiscard__]] produces - a warning. - The 1000 below means a yet unknown threshold. When clang++ version X - starts supporting [[__nodiscard__]] without warning about it, you can - replace the 1000 with X. */ -# if __clang_major__ >= 1000 -# define _GL_ATTRIBUTE_NODISCARD [[__nodiscard__]] -# endif -# elif _GL_HAVE___HAS_C_ATTRIBUTE -# if __has_c_attribute (__nodiscard__) -# define _GL_ATTRIBUTE_NODISCARD [[__nodiscard__]] -# endif -# endif -# endif -# if !defined _GL_ATTRIBUTE_NODISCARD && _GL_HAS_ATTRIBUTE (warn_unused_result) -# define _GL_ATTRIBUTE_NODISCARD __attribute__ ((__warn_unused_result__)) -# endif -# ifndef _GL_ATTRIBUTE_NODISCARD -# define _GL_ATTRIBUTE_NODISCARD -# endif -#endif - -/* _GL_ATTRIBUTE_NOINLINE tells that the compiler should not inline the - function. */ -/* Applies to: functions. */ -#ifndef _GL_ATTRIBUTE_NOINLINE -# if _GL_HAS_ATTRIBUTE (noinline) -# define _GL_ATTRIBUTE_NOINLINE __attribute__ ((__noinline__)) -# else -# define _GL_ATTRIBUTE_NOINLINE -# endif -#endif - -/* _GL_ATTRIBUTE_NONNULL ((N1, N2,...)) declares that the arguments N1, N2,... - must not be NULL. - _GL_ATTRIBUTE_NONNULL () declares that all pointer arguments must not be - null. */ -/* Applies to: functions. */ -#ifndef _GL_ATTRIBUTE_NONNULL -# if _GL_HAS_ATTRIBUTE (nonnull) -# define _GL_ATTRIBUTE_NONNULL(args) __attribute__ ((__nonnull__ args)) -# else -# define _GL_ATTRIBUTE_NONNULL(args) -# endif -#endif - -/* _GL_ATTRIBUTE_NONSTRING declares that the contents of a character array is - not meant to be NUL-terminated. */ -/* Applies to: struct/union members and variables that are arrays of element - type '[[un]signed] char'. */ -#ifndef _GL_ATTRIBUTE_NONSTRING -# if _GL_HAS_ATTRIBUTE (nonstring) -# define _GL_ATTRIBUTE_NONSTRING __attribute__ ((__nonstring__)) -# else -# define _GL_ATTRIBUTE_NONSTRING -# endif -#endif - -/* There is no _GL_ATTRIBUTE_NORETURN; use _Noreturn instead. */ - -/* _GL_ATTRIBUTE_NOTHROW declares that the function does not throw exceptions. - */ -/* Applies to: functions. */ -/* After a function's parameter list, this attribute must come first, before - other attributes. */ -#ifndef _GL_ATTRIBUTE_NOTHROW -# if defined __cplusplus -# if _GL_GNUC_PREREQ (2, 8) || __clang_major__ >= 4 -# if __cplusplus >= 201103L -# define _GL_ATTRIBUTE_NOTHROW noexcept (true) -# else -# define _GL_ATTRIBUTE_NOTHROW throw () -# endif -# else -# define _GL_ATTRIBUTE_NOTHROW -# endif -# else -# if _GL_HAS_ATTRIBUTE (nothrow) -# define _GL_ATTRIBUTE_NOTHROW __attribute__ ((__nothrow__)) -# else -# define _GL_ATTRIBUTE_NOTHROW -# endif -# endif -#endif - -/* _GL_ATTRIBUTE_PACKED declares: - For struct members: The member has the smallest possible alignment. - For struct, union, class: All members have the smallest possible alignment, - minimizing the memory required. */ -/* Applies to: struct members, struct, union, - in C++ also: class. */ -#ifndef _GL_ATTRIBUTE_PACKED -/* Oracle Studio 12.6 miscompiles code with __attribute__ ((__packed__)) despite - __has_attribute OK. */ -# if _GL_HAS_ATTRIBUTE (packed) && !defined __SUNPRO_C -# define _GL_ATTRIBUTE_PACKED __attribute__ ((__packed__)) -# else -# define _GL_ATTRIBUTE_PACKED -# endif -#endif - -/* _GL_ATTRIBUTE_PURE declares: - It is OK for a compiler to move calls to the function and to omit - calls to the function if another call has the same arguments or the - result is not used, and if observable state is the same. - This attribute is safe for a function that does not affect observable state - and always returns exactly once. - (This attribute is looser than _GL_ATTRIBUTE_CONST because the function - can depend on observable state. It is stricter than - _GL_ATTRIBUTE_REPRODUCIBLE because the function must return exactly - once and cannot affect state addressed by its arguments.) */ -/* Applies to: functions. */ -#ifndef _GL_ATTRIBUTE_PURE -# if _GL_HAS_ATTRIBUTE (pure) -# define _GL_ATTRIBUTE_PURE __attribute__ ((__pure__)) -# else -# define _GL_ATTRIBUTE_PURE _GL_ATTRIBUTE_REPRODUCIBLE -# endif -#endif - -/* _GL_ATTRIBUTE_REPRODUCIBLE declares: - It is OK for a compiler to move calls to the function and to omit duplicate - calls to the function with the same arguments, so long as the state - addressed by its arguments is the same and is updated in time for - the rest of the program. - This attribute is safe for a function that is effectless and idempotent; see - ISO C 23 § 6.7.12.7 for a definition of these terms. - (This attribute is looser than _GL_ATTRIBUTE_UNSEQUENCED because - the function need not be stateless and idempotent. It is looser - than _GL_ATTRIBUTE_PURE because the function need not return - exactly once and can affect state addressed by its arguments.) - See also and - . - ATTENTION! Efforts are underway to change the meaning of this attribute. - See . */ -/* Applies to: functions, pointer to functions, function types. */ -#ifndef _GL_ATTRIBUTE_REPRODUCIBLE -/* This may be revisited when gcc and clang support [[reproducible]] or possibly - __attribute__ ((__reproducible__)). */ -# ifndef _GL_BRACKET_BEFORE_ATTRIBUTE -# if _GL_HAS_ATTRIBUTE (reproducible) -# define _GL_ATTRIBUTE_REPRODUCIBLE [[reproducible]] -# endif -# endif -# ifndef _GL_ATTRIBUTE_REPRODUCIBLE -# define _GL_ATTRIBUTE_REPRODUCIBLE -# endif -#endif - -/* _GL_ATTRIBUTE_RETURNS_NONNULL declares that the function's return value is - a non-NULL pointer. */ -/* Applies to: functions. */ -#ifndef _GL_ATTRIBUTE_RETURNS_NONNULL -# if _GL_HAS_ATTRIBUTE (returns_nonnull) -# define _GL_ATTRIBUTE_RETURNS_NONNULL __attribute__ ((__returns_nonnull__)) -# else -# define _GL_ATTRIBUTE_RETURNS_NONNULL -# endif -#endif - -/* _GL_ATTRIBUTE_SENTINEL(pos) declares that the variadic function expects a - trailing NULL argument. - _GL_ATTRIBUTE_SENTINEL () - The last argument is NULL (requires C99). - _GL_ATTRIBUTE_SENTINEL ((N)) - The (N+1)st argument from the end is NULL. */ -/* Applies to: functions. */ -#ifndef _GL_ATTRIBUTE_SENTINEL -# if _GL_HAS_ATTRIBUTE (sentinel) -# define _GL_ATTRIBUTE_SENTINEL(pos) __attribute__ ((__sentinel__ pos)) -# else -# define _GL_ATTRIBUTE_SENTINEL(pos) -# endif -#endif - -/* _GL_ATTRIBUTE_UNSEQUENCED declares: - It is OK for a compiler to move calls to the function and to omit duplicate - calls to the function with the same arguments, so long as the state - addressed by its arguments is the same. - This attribute is safe for a function that is effectless, idempotent, - stateless, and independent; see ISO C 23 § 6.7.12.7 for a definition of - these terms. - (This attribute is stricter than _GL_ATTRIBUTE_REPRODUCIBLE because - the function must be stateless and independent. It is looser than - _GL_ATTRIBUTE_CONST because the function need not return exactly - once and can depend on state addressed by its arguments.) - See also and - . - ATTENTION! Efforts are underway to change the meaning of this attribute. - See . */ -/* Applies to: functions, pointer to functions, function types. */ -#ifndef _GL_ATTRIBUTE_UNSEQUENCED -/* This may be revisited when gcc and clang support [[unsequenced]] or possibly - __attribute__ ((__unsequenced__)). */ -# ifndef _GL_BRACKET_BEFORE_ATTRIBUTE -# if _GL_HAS_ATTRIBUTE (unsequenced) -# define _GL_ATTRIBUTE_UNSEQUENCED [[unsequenced]] -# endif -# endif -# ifndef _GL_ATTRIBUTE_UNSEQUENCED -# define _GL_ATTRIBUTE_UNSEQUENCED -# endif -#endif - -/* A helper macro. Don't use it directly. */ -#ifndef _GL_ATTRIBUTE_UNUSED -# if _GL_HAS_ATTRIBUTE (unused) -# define _GL_ATTRIBUTE_UNUSED __attribute__ ((__unused__)) -# else -# define _GL_ATTRIBUTE_UNUSED -# endif -#endif - - -/* _GL_UNUSED_LABEL; declares that it is not a programming mistake if the - immediately preceding label is not used. The compiler should not warn - if the label is not used. */ -/* Applies to: label (both in C and C++). */ -/* Note that g++ < 4.5 does not support the '__attribute__ ((__unused__)) ;' - syntax. But clang does. */ -#ifndef _GL_UNUSED_LABEL -# if !(defined __cplusplus && !_GL_GNUC_PREREQ (4, 5)) || defined __clang__ -# define _GL_UNUSED_LABEL _GL_ATTRIBUTE_UNUSED -# else -# define _GL_UNUSED_LABEL -# endif -#endif - -/* The following attributes enable detection of multithread-safety problems - and resource leaks at compile-time, by clang ≥ 15, when the warning option - -Wthread-safety is enabled. For usage, see - . */ -#ifndef _GL_ATTRIBUTE_CAPABILITY_TYPE -# if __clang_major__ >= 15 -# define _GL_ATTRIBUTE_CAPABILITY_TYPE(concept) \ - __attribute__ ((__capability__ (concept))) -# else -# define _GL_ATTRIBUTE_CAPABILITY_TYPE(concept) -# endif -#endif -#ifndef _GL_ATTRIBUTE_ACQUIRE_CAPABILITY -# if __clang_major__ >= 15 -# define _GL_ATTRIBUTE_ACQUIRE_CAPABILITY(resource) \ - __attribute__ ((__acquire_capability__ (resource))) -# else -# define _GL_ATTRIBUTE_ACQUIRE_CAPABILITY(resource) -# endif -#endif -#ifndef _GL_ATTRIBUTE_RELEASE_CAPABILITY -# if __clang_major__ >= 15 -# define _GL_ATTRIBUTE_RELEASE_CAPABILITY(resource) \ - __attribute__ ((__release_capability__ (resource))) -# else -# define _GL_ATTRIBUTE_RELEASE_CAPABILITY(resource) -# endif -#endif - - -/* In C++, there is the concept of "language linkage", that encompasses - name mangling and function calling conventions. - The following macros start and end a block of "C" linkage. */ -#ifdef __cplusplus -# define _GL_BEGIN_C_LINKAGE extern "C" { -# define _GL_END_C_LINKAGE } -#else -# define _GL_BEGIN_C_LINKAGE -# define _GL_END_C_LINKAGE -#endif - - -/* A replacement for va_copy, if needed. */ -#define gl_va_copy(a,b) ((a) = (b)) - -/* Define to `__inline__' or `__inline' if that's what the C compiler - calls it, or to nothing if 'inline' is not supported under any name. */ -#ifndef __cplusplus -/* #undef inline */ -#endif - -/* Work around a bug in Apple GCC 4.0.1 build 5465: In C99 mode, it supports - the ISO C 99 semantics of 'extern inline' (unlike the GNU C semantics of - earlier versions), but does not display it by setting __GNUC_STDC_INLINE__. - __APPLE__ && __MACH__ test for Mac OS X. - __APPLE_CC__ tests for the Apple compiler and its version. - __STDC_VERSION__ tests for the C99 mode. */ -#if defined __APPLE__ && defined __MACH__ && __APPLE_CC__ >= 5465 && !defined __cplusplus && __STDC_VERSION__ >= 199901L && !defined __GNUC_STDC_INLINE__ -# define __GNUC_STDC_INLINE__ 1 -#endif - -/* _GL_CMP (n1, n2) performs a three-valued comparison on n1 vs. n2, where - n1 and n2 are expressions without side effects, that evaluate to real - numbers (excluding NaN). - It returns - 1 if n1 > n2 - 0 if n1 == n2 - -1 if n1 < n2 - The naïve code (n1 > n2 ? 1 : n1 < n2 ? -1 : 0) produces a conditional - jump with nearly all GCC versions up to GCC 10. - This variant (n1 < n2 ? -1 : n1 > n2) produces a conditional with many - GCC versions up to GCC 9. - The better code (n1 > n2) - (n1 < n2) from Hacker's Delight § 2-9 - avoids conditional jumps in all GCC versions >= 3.4. */ -#define _GL_CMP(n1, n2) (((n1) > (n2)) - ((n1) < (n2))) - - -/* Define to `int' if does not define. */ -/* #undef mode_t */ - -/* Define to the type of st_nlink in struct stat, or a supertype. */ -/* #undef nlink_t */ - -/* Define as a signed integer type capable of holding a process identifier. */ -/* #undef pid_t */ - -/* Define to the equivalent of the C99 'restrict' keyword, or to - nothing if this is not supported. Do not define if restrict is - supported only directly. */ -#define restrict __restrict__ -/* Work around a bug in older versions of Sun C++, which did not - #define __restrict__ or support _Restrict or __restrict__ - even though the corresponding Sun C compiler ended up with - "#define restrict _Restrict" or "#define restrict __restrict__" - in the previous line. This workaround can be removed once - we assume Oracle Developer Studio 12.5 (2016) or later. */ -#if defined __SUNPRO_CC && !defined __RESTRICT && !defined __restrict__ -# define _Restrict -# define __restrict__ -#endif - -/* Define to `unsigned int' if does not define. */ -/* #undef size_t */ - -/* Define as a signed type of the same size as size_t. */ -/* #undef ssize_t */ - - - /* This definition is a duplicate of the one in unitypes.h. - It is here so that we can cope with an older version of unitypes.h - that does not contain this definition and that is pre-installed among - the public header files. */ - # if defined __restrict \ - || 2 < __GNUC__ + (95 <= __GNUC_MINOR__) \ - || __clang_major__ >= 3 - # define _UC_RESTRICT __restrict - # elif 199901L <= __STDC_VERSION__ || defined restrict - # define _UC_RESTRICT restrict - # else - # define _UC_RESTRICT - # endif - - -/* Define as a macro for copying va_list variables. */ -/* #undef va_copy */ - -#if !(defined __cplusplus \ - ? 1 \ - : (defined __clang__ \ - ? __STDC_VERSION__ >= 202000L && __clang_major__ >= 15 \ - : (defined __GNUC__ \ - ? __STDC_VERSION__ >= 202000L && __GNUC__ >= 13 \ - : defined HAVE_C_BOOL))) -# if !defined __cplusplus && !defined __bool_true_false_are_defined -# if HAVE_STDBOOL_H -# include -# else -# if defined __SUNPRO_C -# error " is not usable with this configuration. To make it usable, add -D_STDC_C99= to $CC." -# else -# error " does not exist on this platform. Use gnulib module 'stdbool-c99' instead of gnulib module 'stdbool'." -# endif -# endif -# endif -# if !true -# define true (!false) -# endif -#endif - -#if (!(defined __clang__ \ - ? (defined __cplusplus \ - ? __cplusplus >= 201703L \ - : __STDC_VERSION__ >= 202000L && __clang_major__ >= 16 \ - && !defined __sun) \ - : (defined __GNUC__ \ - ? (defined __cplusplus \ - ? __cplusplus >= 201103L && __GNUG__ >= 6 \ - : __STDC_VERSION__ >= 202000L && __GNUC__ >= 13 \ - && !defined __sun) \ - : defined HAVE_C_STATIC_ASSERT)) \ - && !defined assert \ - && (!defined __cplusplus \ - || (__cpp_static_assert < 201411 \ - && __GNUG__ < 6 && __clang_major__ < 6))) - #include - #undef/**/assert - #ifdef __sgi - #undef/**/__ASSERT_H__ - #endif - /* Solaris 11.4 defines static_assert as a macro with 2 arguments. - We need it also to be invocable with a single argument. - Haiku 2022 does not define static_assert at all. */ - #if (__STDC_VERSION__ - 0 >= 201112L) && !defined __cplusplus - #undef/**/static_assert - #define static_assert _Static_assert - #endif -#endif - -/* lightpanda: macOS's does not declare strverscmp (a glibc - extension). gnulib normally declares it via its replacement - shim, which we don't pull in. lib/strverscmp.c provides the definition. */ -#ifndef _LIBIDN2_LP_DECLS -#define _LIBIDN2_LP_DECLS -extern int strverscmp(const char *, const char *); - -/* lightpanda: lib/lookup.c calls strchrnul() without including , - so the prototype must reach it through this header. macOS libc also - lacked the symbol entirely before 15.4 — build.zig::buildLibidn2 adds - vendor/libidn2/darwin/strchrnul.c on Darwin to provide the definition. */ -#ifdef __APPLE__ -extern char *strchrnul(const char *, int); -#endif -#endif diff --git a/vendor/libidn2/darwin/strchrnul.c b/vendor/libidn2/darwin/strchrnul.c deleted file mode 100644 index d2a5c3ca..00000000 --- a/vendor/libidn2/darwin/strchrnul.c +++ /dev/null @@ -1,20 +0,0 @@ -/* Darwin-only strchrnul shim for libidn2. - - strchrnul is a glibc extension. macOS libc lacks it before 15.4, and - libidn2's lib/lookup.c never includes — so even on newer - macOS the declaration would not reach the call site. The matching - prototype is declared next to the strverscmp shim in - vendor/libidn2/config.h (within the _LIBIDN2_LP_DECLS block, gated on - __APPLE__), so callers compile; this file provides the symbol so - they link. - - gnulib's strchrnul.c falls through to rawmemchr() when the search byte - is NUL — also a glibc extension. libidn2 only ever searches for '.', so - a straight byte scan is enough and avoids dragging in a second shim. */ - -char *strchrnul(const char *s, int c_in) { - const unsigned char c = (unsigned char) c_in; - const unsigned char *p = (const unsigned char *) s; - while (*p && *p != c) p++; - return (char *) p; -} From 5ac7b54f53e33d9fb841ebbc7b3d3163e3ad5767 Mon Sep 17 00:00:00 2001 From: Karl Seguin Date: Sat, 30 May 2026 14:12:29 +0800 Subject: [PATCH 2/9] Apply suggestions from code review Co-authored-by: Pierre Tachoire --- src/browser/Session.zig | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/src/browser/Session.zig b/src/browser/Session.zig index faaf44cf..3dbdf170 100644 --- a/src/browser/Session.zig +++ b/src/browser/Session.zig @@ -70,7 +70,7 @@ _active: ?*Page = null, // In-flight root navigation _pending: ?*Page = null, -_page_descruction_queue: std.ArrayList(*Page) = .{}, +_page_destruction_queue: std.ArrayList(*Page) = .{}, _frame_destruction_queue: std.ArrayList(*Frame) = .{}, // Loader IDs are scoped to the Session: each new BrowserContext gets a @@ -156,7 +156,7 @@ pub fn processDestroyQueues(self: *Session) void { page.deinit(); self.browser.page_pool.destroy(page); } - self._page_descruction_queue.clearRetainingCapacity(); + self._page_destruction_queue.clearRetainingCapacity(); } } } @@ -178,7 +178,7 @@ fn allocatePage(self: *Session, frame_id: u32) !*Page { // Tear down and free a Page allocated via allocatePage. fn queuePageDestruction(self: *Session, page: *Page) void { - self._page_descruction_queue.append(self.arena, page) catch @panic("OOM"); + self._page_destruction_queue.append(self.arena, page) catch @panic("OOM"); } pub fn queueFrameDestruction(self: *Session, frame: *Frame) void { From ee7e9715a468d1c4441a94ab4bb5c2b7c854f893 Mon Sep 17 00:00:00 2001 From: Karl Seguin Date: Sat, 30 May 2026 14:16:50 +0800 Subject: [PATCH 3/9] typo fix --- src/browser/Session.zig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/browser/Session.zig b/src/browser/Session.zig index 3dbdf170..10d85f83 100644 --- a/src/browser/Session.zig +++ b/src/browser/Session.zig @@ -150,7 +150,7 @@ pub fn processDestroyQueues(self: *Session) void { } { - const queue = self._page_descruction_queue.items; + const queue = self._page_destruction_queue.items; if (queue.len > 0) { for (queue) |page| { page.deinit(); From 7e0cc672be0a775115e362d04e6d9d964c6c0037 Mon Sep 17 00:00:00 2001 From: Karl Seguin Date: Sat, 30 May 2026 17:27:43 +0800 Subject: [PATCH 4/9] Remove CookieListItem Interface CookieStore get/getAll return a play JS object, not an WebAPI interface. `typeof CookieListItem === undefined` in browsers. This removes the interface and returns the CookieListItem as a plain object. Also adds various name/value validation based on WPT tests --- src/browser/tests/cookie_store.html | 22 ++- .../webapi/event/CookieChangeEvent.zig | 13 +- src/browser/webapi/storage/CookieStore.zig | 173 ++++++++++-------- 3 files changed, 122 insertions(+), 86 deletions(-) diff --git a/src/browser/tests/cookie_store.html b/src/browser/tests/cookie_store.html index 9838d7cb..bcf8b26b 100644 --- a/src/browser/tests/cookie_store.html +++ b/src/browser/tests/cookie_store.html @@ -129,8 +129,10 @@ } }; - // Empty name. - await expectReject('empty name', '', 'v'); + // Empty name only rejects when the value is empty too or carries an '='. + // A nameless cookie with an ordinary value is valid (cf. `Set-Cookie: =v`). + await expectReject('empty name and empty value', '', ''); + await expectReject('empty name with = in value', '', 'a=b'); // Forbidden chars in name. await expectReject('name with =', 'a=b', 'v'); await expectReject('name with ;', 'a;b', 'v'); @@ -142,6 +144,13 @@ await expectReject('path with ;', { name: 'k', value: 'v', path: '/;evil' }); await expectReject('domain with newline', { name: 'k', value: 'v', domain: 'bad\n.example' }); + // A nameless cookie with an ordinary value is accepted, and delete('') + // targets it without rejecting on the empty name. + await cookieStore.set('', 'nameless'); + testing.expectEqual('nameless', (await cookieStore.get('')).value); + await cookieStore.delete(''); + testing.expectEqual(null, await cookieStore.get('')); + // A clean call still works after the rejections. await cookieStore.set('after-validation', 'ok'); const item = await cookieStore.get('after-validation'); @@ -150,7 +159,7 @@ }); - diff --git a/src/browser/webapi/Navigator.zig b/src/browser/webapi/Navigator.zig index 958fd109..e1e56eb5 100644 --- a/src/browser/webapi/Navigator.zig +++ b/src/browser/webapi/Navigator.zig @@ -139,11 +139,6 @@ pub fn getModelContext(_: *const Navigator, frame: *Frame) *ModelContext { return &frame.window._model_context; } -pub fn getBattery(_: *const Navigator, frame: *Frame) !js.Promise { - log.info(.not_implemented, "navigator.getBattery", .{}); - return frame.js.local.?.rejectErrorPromise(.{ .dom_exception = .{ .err = error.NotSupported } }); -} - pub fn registerProtocolHandler(_: *const Navigator, scheme: []const u8, url: [:0]const u8, frame: *const Frame) !void { try validateProtocolHandlerScheme(scheme); try validateProtocolHandlerURL(url, frame); @@ -248,7 +243,6 @@ pub const JsApi = struct { // Methods pub const javaEnabled = bridge.function(Navigator.javaEnabled, .{}); - pub const getBattery = bridge.function(Navigator.getBattery, .{}); pub const permissions = bridge.accessor(Navigator.getPermissions, null, .{}); pub const storage = bridge.accessor(Navigator.getStorage, null, .{}); pub const userAgentData = bridge.accessor(Navigator.getUserAgentData, null, .{}); From 9eb229a963f67f4d6903bd456f37689f9ba07e2d Mon Sep 17 00:00:00 2001 From: Karl Seguin Date: Thu, 28 May 2026 16:26:09 +0800 Subject: [PATCH 8/9] Improve ImporMap This is driven by import-maps WPT tests. It generally does 3 high level additions, plus various small compliance tweaks. 1 - Entries with a trailing match are used for prefix matching 2 - Supports scopes (which are entries group into a specific route-like prefix) 3 - Because of the above, resolves in correct order with fallback Resolution is based on longest-match wins, so it doesn't require a fancy data structures. We use ordered (by length) slices, and just iterate until we find a match. Neither our parsing nor our matching is super efficient. While a page might have hundreds of scripts, it likely has only 0-1 import maps and relatively few values. ImportMap.resolve always returns the final URL. So even if a match isn't found based on the parsed JSON, it'll return the URL.resolve(base, url). Just to make WPT tests pass, we do have to track invalid entries in the ImportMap, e.g. "key": "not-a-url". In the previous version, we'd fallback to URL.resolve(base, url). Now we return null and leave it to the caller to decide. --- src/browser/ImportMap.zig | 535 +++++++++++++++++++++++ src/browser/ScriptManager.zig | 31 +- src/browser/ScriptManagerBase.zig | 32 +- src/browser/URL.zig | 62 ++- src/browser/js/Context.zig | 19 +- src/browser/tests/element/html/base.html | 27 ++ src/browser/webapi/element/html/Base.zig | 48 ++ 7 files changed, 686 insertions(+), 68 deletions(-) create mode 100644 src/browser/ImportMap.zig create mode 100644 src/browser/tests/element/html/base.html diff --git a/src/browser/ImportMap.zig b/src/browser/ImportMap.zig new file mode 100644 index 00000000..3066b95d --- /dev/null +++ b/src/browser/ImportMap.zig @@ -0,0 +1,535 @@ +// Copyright (C) 2023-2026 Lightpanda (Selecy SAS) +// +// Francis Bouvier +// Pierre Tachoire +// +// This program is free software: you can redistribute it and/or modify +// it under the terms of the GNU Affero General Public License as +// published by the Free Software Foundation, either version 3 of the +// License, or (at your option) any later version. +// +// This program is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +// GNU Affero General Public License for more details. +// +// You should have received a copy of the GNU Affero General Public License +// along with this program. If not, see . + +// Parsed + + + + + diff --git a/src/browser/webapi/element/html/Base.zig b/src/browser/webapi/element/html/Base.zig index 0c2f2e8a..bb5616c8 100644 --- a/src/browser/webapi/element/html/Base.zig +++ b/src/browser/webapi/element/html/Base.zig @@ -1,4 +1,7 @@ const js = @import("../../../js/js.zig"); +const URL = @import("../../../URL.zig"); +const Frame = @import("../../../Frame.zig"); + const Node = @import("../../Node.zig"); const Element = @import("../../Element.zig"); const HtmlElement = @import("../Html.zig"); @@ -14,6 +17,44 @@ pub fn asNode(self: *Base) *Node { return self.asElement().asNode(); } +pub fn getHref(self: *Base, frame: *Frame) ![]const u8 { + const element = self.asElement(); + const href = element.getAttributeSafe(comptime .wrap("href")) orelse return ""; + if (href.len == 0) { + return ""; + } + return URL.resolve(frame.call_arena, frame.url, href, .{}); +} + +pub fn setHref(self: *Base, value: []const u8, frame: *Frame) !void { + const element = self.asElement(); + try element.setAttributeSafe(comptime .wrap("href"), .wrap(value), frame); + + // Per HTML spec, the document's base URL is the href of the FIRST + // element in tree order that has an href attribute — not necessarily this + // one. Re-derive from scratch so that setting href on a non-authoritative + // , or clearing href on the authoritative one, both work correctly. + const node = element.asNode(); + if (!node.isConnected()) { + return; + } + + const owner = node.ownerFrame(frame); + const first = (try owner.document.querySelector(comptime .wrap("base[href]"), owner)) orelse { + owner.base_url = null; + return; + }; + const href = first.getAttributeSafe(comptime .wrap("href")) orelse { + owner.base_url = null; + return; + }; + if (href.len == 0) { + owner.base_url = null; + return; + } + owner.base_url = try URL.resolve(owner.arena, owner.url, href, .{}); +} + pub const JsApi = struct { pub const bridge = js.Bridge(Base); @@ -22,4 +63,11 @@ pub const JsApi = struct { pub const prototype_chain = bridge.prototypeChain(); pub var class_id: bridge.ClassId = undefined; }; + + pub const href = bridge.accessor(Base.getHref, Base.setHref, .{ .ce_reactions = true }); }; + +const testing = @import("../../../../testing.zig"); +test "WebApi: HTML.Base" { + try testing.htmlRunner("element/html/base.html", .{}); +} From da6d8215981c85e8488c6bbce2624b9a3ae3f2ae Mon Sep 17 00:00:00 2001 From: Karl Seguin Date: Mon, 1 Jun 2026 15:29:32 +0800 Subject: [PATCH 9/9] Coerce value to strings on string.String (SSO) target When `[]const u8` is a web api parameter, we'll coerce the value to a string, e.g. true -> "true". This is correct for almost every API. APIs can also opt to use a string.String (or string.String.Global). This is purely meant as an optimization, but its behavior is currently different than `[]const u8` as the bridge will only map actual JS strings to it. This commit makes String and []const u8 behave the same, so that the only decision as to which to use is about performance. (APIs that strictly want a string should use js.String or js.Value and do the type check) --- src/browser/js/Local.zig | 10 ++-------- src/browser/tests/element/dataset.html | 5 +++++ 2 files changed, 7 insertions(+), 8 deletions(-) diff --git a/src/browser/js/Local.zig b/src/browser/js/Local.zig index e046dcf6..ef15c4a1 100644 --- a/src/browser/js/Local.zig +++ b/src/browser/js/Local.zig @@ -767,14 +767,8 @@ fn jsValueToStruct(self: *const Local, comptime T: type, js_val: js.Value) !?T { if (!js_str.containsOnlyOneByte()) return error.InvalidCharacterError; return .{ .bytes = try js_str.toOneByteSlice(self.call_arena) }; }, - string.String => { - const js_str = js_val.isString() orelse return null; - return try js_str.toSSO(false); - }, - string.Global => { - const js_str = js_val.isString() orelse return null; - return try js_str.toSSO(true); - }, + string.String => try js_val.toSSO(false), + string.Global => try js_val.toSSO(true), else => { if (!js_val.isObject()) { return null; diff --git a/src/browser/tests/element/dataset.html b/src/browser/tests/element/dataset.html index c9178c74..72bc3fe4 100644 --- a/src/browser/tests/element/dataset.html +++ b/src/browser/tests/element/dataset.html @@ -36,6 +36,11 @@ testing.expectEqual('value', el.dataset.newAttr); testing.expectEqual('value', el.getAttribute('data-new-attr')); + + el.dataset.newAttr = true; + + testing.expectEqual('true', el.dataset.newAttr); + testing.expectEqual('true', el.getAttribute('data-new-attr')); }