Merge pull request #1 from SmallThingz/dev

Dev

Author: ItsMeSamey

DOCUMENTATION.md

@@ -83,7 +83,7 @@ All examples are verified by running `zig build examples-check`
   - `lastChild()`
   - `nextSibling()`
   - `prevSibling()`
-  - `children()` (borrowed `[]const u32` index view)
+  - `children()` (iterator of wrapped child nodes; `collect(allocator)` returns an owned `[]Node`)
 - Text:
   - `innerText(allocator)` (borrowed or allocated depending on shape)
   - `innerTextWithOptions(allocator, TextOptions)`
@@ -193,49 +193,49 @@ Source: `bench/results/latest.json` (`stable` profile).
 
 | Fixture | ours | lol-html | lexbor |
 |---|---:|---:|---:|
-| `rust-lang.html` | 1447.99 | 1474.65 | 332.72 |
-| `wiki-html.html` | 1645.45 | 1215.04 | 271.24 |
-| `mdn-html.html` | 2570.09 | 1879.00 | 404.50 |
-| `w3-html52.html` | 1064.19 | 764.62 | 199.22 |
-| `hn.html` | 1263.60 | 885.26 | 223.15 |
-| `python-org.html` | 1549.02 | 1356.21 | 284.19 |
-| `kernel-org.html` | 1440.47 | 1300.81 | 276.52 |
-| `gnu-org.html` | 1917.36 | 1482.15 | 317.74 |
-| `ziglang-org.html` | 1480.49 | 1257.62 | 291.72 |
-| `ziglang-doc-master.html` | 1122.44 | 987.16 | 214.23 |
-| `wikipedia-unicode-list.html` | 1247.00 | 1024.98 | 215.21 |
-| `whatwg-html-spec.html` | 1113.73 | 841.16 | 210.83 |
-| `synthetic-forms.html` | 1046.17 | 710.72 | 174.94 |
-| `synthetic-table-grid.html` | 768.56 | 622.31 | 152.86 |
-| `synthetic-list-nested.html` | 833.77 | 598.02 | 152.45 |
-| `synthetic-comments-doctype.html` | 1200.72 | 827.66 | 212.09 |
-| `synthetic-template-rich.html` | 628.02 | 444.34 | 134.10 |
-| `synthetic-whitespace-noise.html` | 1104.21 | 919.69 | 170.33 |
-| `synthetic-news-feed.html` | 835.27 | 577.95 | 144.46 |
-| `synthetic-ecommerce.html` | 787.72 | 556.51 | 151.95 |
-| `synthetic-forum-thread.html` | 839.48 | 579.84 | 143.06 |
+| `rust-lang.html` | 2132.70 | 1501.81 | 324.53 |
+| `wiki-html.html` | 1991.85 | 1220.28 | 267.50 |
+| `mdn-html.html` | 2939.75 | 1894.70 | 408.76 |
+| `w3-html52.html` | 956.25 | 754.98 | 196.30 |
+| `hn.html` | 1595.94 | 893.48 | 221.75 |
+| `python-org.html` | 2116.13 | 1354.79 | 280.79 |
+| `kernel-org.html` | 1979.68 | 1335.72 | 290.65 |
+| `gnu-org.html` | 2368.06 | 1490.41 | 313.42 |
+| `ziglang-org.html` | 1874.52 | 1299.04 | 289.84 |
+| `ziglang-doc-master.html` | 1431.90 | 1045.18 | 225.11 |
+| `wikipedia-unicode-list.html` | 1647.71 | 1081.56 | 226.67 |
+| `whatwg-html-spec.html` | 1344.31 | 892.66 | 219.37 |
+| `synthetic-forms.html` | 1396.29 | 781.68 | 189.28 |
+| `synthetic-table-grid.html` | 1086.37 | 724.33 | 169.46 |
+| `synthetic-list-nested.html` | 1190.11 | 652.22 | 161.58 |
+| `synthetic-comments-doctype.html` | 1862.19 | 943.51 | 223.61 |
+| `synthetic-template-rich.html` | 913.72 | 465.79 | 143.20 |
+| `synthetic-whitespace-noise.html` | 1592.74 | 1062.32 | 189.70 |
+| `synthetic-news-feed.html` | 1179.12 | 646.78 | 157.61 |
+| `synthetic-ecommerce.html` | 1134.48 | 637.67 | 163.21 |
+| `synthetic-forum-thread.html` | 1145.80 | 643.29 | 161.14 |
 
 #### Query Match Throughput (ours)
 
 | Case | ours ops/s | ours ns/op |
 |---|---:|---:|
-| `attr-heavy-button` | 1148936.76 | 870.37 |
-| `attr-heavy-nav` | 1130790.00 | 884.34 |
+| `attr-heavy-button` | 191877.12 | 5211.67 |
+| `attr-heavy-nav` | 87885.90 | 11378.39 |
 
 #### Cached Query Throughput (ours)
 
 | Case | ours ops/s | ours ns/op |
 |---|---:|---:|
-| `attr-heavy-button` | 1305257.78 | 766.13 |
-| `attr-heavy-nav` | 1347173.46 | 742.29 |
+| `attr-heavy-button` | 144352.80 | 6927.47 |
+| `attr-heavy-nav` | 120324.49 | 8310.86 |
 
 #### Query Parse Throughput (ours)
 
 | Selector case | Ops/s | ns/op |
 |---|---:|---:|
-| `simple` | 17335919.85 | 57.68 |
-| `complex` | 5836657.49 | 171.33 |
-| `grouped` | 6396371.26 | 156.34 |
+| `simple` | 10465915.34 | 95.55 |
+| `complex` | 4910511.63 | 203.64 |
+| `grouped` | 6290595.08 | 158.97 |
 
 For full per-parser, per-fixture tables and gate output:
 - `bench/results/latest.md`

README.md

@@ -25,9 +25,9 @@ Source: `bench/results/latest.json` (`stable` profile).
 ### Parse Throughput (Average Across Fixtures)
 
 ```text
-ours     │████████████████████│ 1233.61 MB/s (100.00%)
-lol-html │████████████████░░░░│ 966.94 MB/s (78.38%)
-lexbor   │████░░░░░░░░░░░░░░░░│ 222.74 MB/s (18.06%)
+ours     │████████████████████│ 1613.31 MB/s (100.00%)
+lol-html │█████████████░░░░░░░│ 1015.34 MB/s (62.94%)
+lexbor   │███░░░░░░░░░░░░░░░░░│ 229.69 MB/s (14.24%)
 ```
 
 ### Conformance Snapshot

build.zig

@@ -22,6 +22,15 @@ pub fn build(b: *std.Build) void {
         }),
     });
 
+    const parse_mode_mod = b.createModule(.{
+        .root_source_file = b.path("tools/parse_mode.zig"),
+        .target = target,
+        .optimize = optimize,
+        .imports = &.{
+            .{ .name = "htmlparser", .module = mod },
+        },
+    });
+
     const bench_exe = b.addExecutable(.{
         .name = "htmlparser-bench",
         .root_module = b.createModule(.{
@@ -30,6 +39,7 @@ pub fn build(b: *std.Build) void {
             .optimize = optimize,
             .imports = &.{
                 .{ .name = "htmlparser", .module = mod },
+                .{ .name = "parse_mode", .module = parse_mode_mod },
             },
         }),
     });
@@ -111,11 +121,13 @@ pub fn build(b: *std.Build) void {
 
     const mod_tests = b.addTest(.{
         .root_module = mod,
+        .test_runner = .{ .path = b.path("tools/test_runner.zig"), .mode = .simple },
     });
     const run_mod_tests = b.addRunArtifact(mod_tests);
 
     const exe_tests = b.addTest(.{
         .root_module = exe.root_module,
+        .test_runner = .{ .path = b.path("tools/test_runner.zig"), .mode = .simple },
     });
     const run_exe_tests = b.addRunArtifact(exe_tests);
 
@@ -126,8 +138,17 @@ pub fn build(b: *std.Build) void {
             .optimize = optimize,
             .imports = &.{
                 .{ .name = "htmlparser", .module = mod },
+                .{ .name = "examples", .module = b.createModule(.{
+                    .root_source_file = b.path("examples/examples.zig"),
+                    .target = target,
+                    .optimize = optimize,
+                    .imports = &.{
+                        .{ .name = "htmlparser", .module = mod },
+                    },
+                }) },
             },
         }),
+        .test_runner = .{ .path = b.path("tools/test_runner.zig"), .mode = .simple },
     });
     const run_examples_tests = b.addRunArtifact(examples_tests);
 
@@ -140,14 +161,41 @@ pub fn build(b: *std.Build) void {
                 .{ .name = "htmlparser", .module = mod },
             },
         }),
+        .test_runner = .{ .path = b.path("tools/test_runner.zig"), .mode = .simple },
     });
     const run_behavioral_tests = b.addRunArtifact(behavioral_tests);
 
+    const scripts_tests = b.addTest(.{
+        .root_module = b.createModule(.{
+            .root_source_file = b.path("tools/scripts.zig"),
+            .target = target,
+            .optimize = optimize,
+        }),
+        .test_runner = .{ .path = b.path("tools/test_runner.zig"), .mode = .simple },
+    });
+    const run_scripts_tests = b.addRunArtifact(scripts_tests);
+
+    const bench_tests = b.addTest(.{
+        .root_module = b.createModule(.{
+            .root_source_file = b.path("tools/bench/bench.zig"),
+            .target = target,
+            .optimize = optimize,
+            .imports = &.{
+                .{ .name = "htmlparser", .module = mod },
+                .{ .name = "parse_mode", .module = parse_mode_mod },
+            },
+        }),
+        .test_runner = .{ .path = b.path("tools/test_runner.zig"), .mode = .simple },
+    });
+    const run_bench_tests = b.addRunArtifact(bench_tests);
+
     const test_step = b.step("test", "Run tests");
     test_step.dependOn(&run_mod_tests.step);
     test_step.dependOn(&run_exe_tests.step);
     test_step.dependOn(&run_examples_tests.step);
     test_step.dependOn(&run_behavioral_tests.step);
+    test_step.dependOn(&run_scripts_tests.step);
+    test_step.dependOn(&run_bench_tests.step);
 
     const ship_check_step = b.step("ship-check", "Run release-readiness checks (test + docs + examples)");
     ship_check_step.dependOn(test_step);

examples/basic_parse_query.zig

@@ -3,7 +3,7 @@ const html = @import("htmlparser");
 const default_options: html.ParseOptions = .{};
 const Document = default_options.GetDocument();
 
-fn run() !void {
+pub fn run() !void {
     var doc = Document.init(std.testing.allocator);
     defer doc.deinit();
 

examples/cached_selector.zig

@@ -3,7 +3,7 @@ const html = @import("htmlparser");
 const default_options: html.ParseOptions = .{};
 const Document = default_options.GetDocument();
 
-fn run() !void {
+pub fn run() !void {
     var doc = Document.init(std.testing.allocator);
     defer doc.deinit();
 
@@ -20,7 +20,7 @@ fn run() !void {
     defer arena.deinit();
 
     const sel = try html.Selector.compileRuntime(arena.allocator(), "a[href^=https][class~=button]");
-    const first = doc.queryOneCached(&sel) orelse return error.TestUnexpectedResult;
+    const first = doc.queryOneCached(sel) orelse return error.TestUnexpectedResult;
     try std.testing.expectEqualStrings("a1", first.getAttributeValue("id").?);
 }
 

examples/debug_query_report.zig

@@ -3,19 +3,19 @@ const html = @import("htmlparser");
 const default_options: html.ParseOptions = .{};
 const Document = default_options.GetDocument();
 
-fn run() !void {
+pub fn run() !void {
     var doc = Document.init(std.testing.allocator);
     defer doc.deinit();
 
     var input = "<div><a id='one' class='nav'></a><a id='two'></a></div>".*;
     try doc.parse(&input, .{});
 
-    var report: html.QueryDebugReport = .{};
-    const node = try doc.queryOneRuntimeDebug("a[href^=https]", &report);
-    try std.testing.expect(node == null);
-    try std.testing.expect(report.visited_elements > 0);
-    try std.testing.expect(report.near_miss_len > 0);
-    try std.testing.expect(report.near_misses[0].reason.kind != .none);
+    const result = doc.queryOneRuntimeDebug("a[href^=https]");
+    try std.testing.expect(result.err == null);
+    try std.testing.expect(result.node == null);
+    try std.testing.expect(result.report.visited_elements > 0);
+    try std.testing.expect(result.report.near_miss_len > 0);
+    try std.testing.expect(result.report.near_misses[0].reason.kind != .none);
 }
 
 test "query debug report for selector mismatch" {

examples/examples.zig

@@ -0,0 +1,9 @@
+pub const basic_parse_query = @import("basic_parse_query.zig");
+pub const runtime_selector = @import("runtime_selector.zig");
+pub const cached_selector = @import("cached_selector.zig");
+pub const navigation_and_children = @import("navigation_and_children.zig");
+pub const inner_text_options = @import("inner_text_options.zig");
+pub const strict_vs_fastest_parse = @import("strict_vs_fastest_parse.zig");
+pub const debug_query_report = @import("debug_query_report.zig");
+pub const instrumentation_hooks = @import("instrumentation_hooks.zig");
+pub const query_time_decode = @import("query_time_decode.zig");

examples/inner_text_options.zig

@@ -3,7 +3,7 @@ const html = @import("htmlparser");
 const default_options: html.ParseOptions = .{};
 const Document = default_options.GetDocument();
 
-fn run() !void {
+pub fn run() !void {
     var doc = Document.init(std.testing.allocator);
     defer doc.deinit();
 

examples/instrumentation_hooks.zig

@@ -26,17 +26,17 @@ const Hooks = struct {
     }
 };
 
-fn run() !void {
+pub fn run() !void {
     var doc = Document.init(std.testing.allocator);
     defer doc.deinit();
 
     var hooks: Hooks = .{};
     var input = "<div><span id='x'></span></div>".*;
-    try html.parseWithHooks(&doc, &input, .{}, &hooks);
+    try html.parseWithHooks(std.testing.io, &doc, &input, .{}, &hooks);
     try std.testing.expectEqual(@as(usize, 1), hooks.parse_start_calls);
     try std.testing.expectEqual(@as(usize, 1), hooks.parse_end_calls);
 
-    _ = try html.queryOneRuntimeWithHooks(&doc, "span#x", &hooks);
+    _ = try html.queryOneRuntimeWithHooks(std.testing.io, &doc, "span#x", &hooks);
     try std.testing.expectEqual(@as(usize, 1), hooks.query_start_calls);
     try std.testing.expectEqual(@as(usize, 1), hooks.query_end_calls);
 }

examples/navigation_and_children.zig

@@ -3,7 +3,7 @@ const html = @import("htmlparser");
 const default_options: html.ParseOptions = .{};
 const Document = default_options.GetDocument();
 
-fn run() !void {
+pub fn run() !void {
     var doc = Document.init(std.testing.allocator);
     defer doc.deinit();
 
@@ -18,13 +18,10 @@ fn run() !void {
     try std.testing.expectEqualStrings("body", last.getAttributeValue("id").?);
 
     var children = main.children();
-    var child_indexes: std.ArrayListUnmanaged(u32) = .{};
-    defer child_indexes.deinit(std.testing.allocator);
-    try children.collect(std.testing.allocator, &child_indexes);
-    try std.testing.expectEqual(@as(usize, 3), child_indexes.items.len);
-    const first_idx = child_indexes.items[0];
-    const first_via_index = main.doc.nodeAt(first_idx) orelse return error.TestUnexpectedResult;
-    try std.testing.expectEqualStrings("title", first_via_index.getAttributeValue("id").?);
+    const child_nodes = try children.collect(std.testing.allocator);
+    defer std.testing.allocator.free(child_nodes);
+    try std.testing.expectEqual(@as(usize, 3), child_nodes.len);
+    try std.testing.expectEqualStrings("title", child_nodes[0].getAttributeValue("id").?);
 }
 
 test "navigation and children iterator" {