Merge pull request #893 from lightpanda-io/dump_noscript

karlseguin · web-flow · commit a74f79118ffa · 2025-07-15T21:22:33.000+08:00
Add a --noscript option to "improve" --dump
diff --git a/src/browser/dom/element.zig b/src/browser/dom/element.zig
@@ -110,13 +110,13 @@ pub const Element = struct {
 
     pub fn get_innerHTML(self: *parser.Element, page: *Page) ![]const u8 {
         var buf = std.ArrayList(u8).init(page.arena);
-        try dump.writeChildren(parser.elementToNode(self), buf.writer());
+        try dump.writeChildren(parser.elementToNode(self), .{}, buf.writer());
         return buf.items;
     }
 
     pub fn get_outerHTML(self: *parser.Element, page: *Page) ![]const u8 {
         var buf = std.ArrayList(u8).init(page.arena);
-        try dump.writeNode(parser.elementToNode(self), buf.writer());
+        try dump.writeNode(parser.elementToNode(self), .{}, buf.writer());
         return buf.items;
     }
 
diff --git a/src/browser/dump.zig b/src/browser/dump.zig
@@ -21,10 +21,14 @@ const std = @import("std");
 const parser = @import("netsurf.zig");
 const Walker = @import("dom/walker.zig").WalkerChildren;
 
+pub const Opts = struct {
+    exclude_scripts: bool = false,
+};
+
 // writer must be a std.io.Writer
-pub fn writeHTML(doc: *parser.Document, writer: anytype) !void {
+pub fn writeHTML(doc: *parser.Document, opts: Opts, writer: anytype) !void {
     try writer.writeAll("<!DOCTYPE html>\n");
-    try writeChildren(parser.documentToNode(doc), writer);
+    try writeChildren(parser.documentToNode(doc), opts, writer);
     try writer.writeAll("\n");
 }
 
@@ -54,10 +58,15 @@ pub fn writeDocType(doc_type: *parser.DocumentType, writer: anytype) !void {
     try writer.writeAll(">");
 }
 
-pub fn writeNode(node: *parser.Node, writer: anytype) anyerror!void {
+pub fn writeNode(node: *parser.Node, opts: Opts, writer: anytype) anyerror!void {
     switch (try parser.nodeType(node)) {
         .element => {
             // open the tag
+            const tag_type = try parser.elementHTMLGetTagType(@ptrCast(node));
+            if (tag_type == .script and opts.exclude_scripts) {
+                return;
+            }
+
             const tag = try parser.nodeLocalName(node);
             try writer.writeAll("<");
             try writer.writeAll(tag);
@@ -82,12 +91,12 @@ pub fn writeNode(node: *parser.Node, writer: anytype) anyerror!void {
             // void elements can't have any content.
             if (try isVoid(parser.nodeToElement(node))) return;
 
-            if (try parser.elementHTMLGetTagType(@ptrCast(node)) == .script) {
+            if (tag_type == .script) {
                 try writer.writeAll(try parser.nodeTextContent(node) orelse "");
             } else {
                 // write the children
                 // TODO avoid recursion
-                try writeChildren(node, writer);
+                try writeChildren(node, opts, writer);
             }
 
             // close the tag
@@ -129,12 +138,12 @@ pub fn writeNode(node: *parser.Node, writer: anytype) anyerror!void {
 }
 
 // writer must be a std.io.Writer
-pub fn writeChildren(root: *parser.Node, writer: anytype) !void {
+pub fn writeChildren(root: *parser.Node, opts: Opts, writer: anytype) !void {
     const walker = Walker{};
     var next: ?*parser.Node = null;
     while (true) {
         next = try walker.get_next(root, next) orelse break;
-        try writeNode(next.?, writer);
+        try writeNode(next.?, opts, writer);
     }
 }
 
@@ -238,6 +247,6 @@ fn testWriteFullHTML(comptime expected: []const u8, src: []const u8) !void {
     defer parser.documentHTMLClose(doc_html) catch {};
 
     const doc = parser.documentHTMLToDocument(doc_html);
-    try writeHTML(doc, buf.writer(testing.allocator));
+    try writeHTML(doc, .{}, buf.writer(testing.allocator));
     try testing.expectEqualStrings(expected, buf.items);
 }
diff --git a/src/browser/page.zig b/src/browser/page.zig
@@ -142,15 +142,15 @@ pub const Page = struct {
     }
 
     // dump writes the page content into the given file.
-    pub fn dump(self: *const Page, out: std.fs.File) !void {
+    pub fn dump(self: *const Page, opts: Dump.Opts, out: std.fs.File) !void {
         if (self.raw_data) |raw_data| {
             // raw_data was set if the document was not HTML, dump the data content only.
             return try out.writeAll(raw_data);
         }
 
         // if the page has a pointer to a document, dumps the HTML.
         const doc = parser.documentHTMLToDocument(self.window.document);
-        try Dump.writeHTML(doc, out);
+        try Dump.writeHTML(doc, opts, out);
     }
 
     pub fn fetchModuleSource(ctx: *anyopaque, src: []const u8) !?[]const u8 {
diff --git a/src/browser/xmlserializer/xmlserializer.zig b/src/browser/xmlserializer/xmlserializer.zig
@@ -36,9 +36,9 @@ pub const XMLSerializer = struct {
     pub fn _serializeToString(_: *const XMLSerializer, root: *parser.Node, page: *Page) ![]const u8 {
         var buf = std.ArrayList(u8).init(page.arena);
         switch (try parser.nodeType(root)) {
-            .document => try dump.writeHTML(@as(*parser.Document, @ptrCast(root)), buf.writer()),
+            .document => try dump.writeHTML(@as(*parser.Document, @ptrCast(root)), .{}, buf.writer()),
             .document_type => try dump.writeDocType(@as(*parser.DocumentType, @ptrCast(root)), buf.writer()),
-            else => try dump.writeNode(root, buf.writer()),
+            else => try dump.writeNode(root, .{}, buf.writer()),
         }
         return buf.items;
     }
diff --git a/src/main.zig b/src/main.zig
@@ -134,7 +134,7 @@ fn run(alloc: Allocator) !void {
 
             // dump
             if (opts.dump) {
-                try page.dump(std.io.getStdOut());
+                try page.dump(.{ .exclude_scripts = opts.noscript }, std.io.getStdOut());
             }
         },
         else => unreachable,
@@ -212,6 +212,7 @@ const Command = struct {
         url: []const u8,
         dump: bool = false,
         common: Common,
+        noscript: bool = false,
     };
 
     const Common = struct {
@@ -275,6 +276,7 @@ const Command = struct {
             \\Options:
             \\--dump          Dumps document to stdout.
             \\                Defaults to false.
+            \\--noscript      Exclude <script> tags in dump. Defaults to false.
             \\
         ++ common_options ++
             \\
@@ -352,6 +354,9 @@ fn inferMode(opt: []const u8) ?App.RunMode {
     if (std.mem.eql(u8, opt, "--dump")) {
         return .fetch;
     }
+    if (std.mem.eql(u8, opt, "--noscript")) {
+        return .fetch;
+    }
     if (std.mem.startsWith(u8, opt, "--") == false) {
         return .fetch;
     }
@@ -437,6 +442,7 @@ fn parseFetchArgs(
     args: *std.process.ArgIterator,
 ) !Command.Fetch {
     var dump: bool = false;
+    var noscript: bool = true;
     var url: ?[]const u8 = null;
     var common: Command.Common = .{};
 
@@ -446,6 +452,11 @@ fn parseFetchArgs(
             continue;
         }
 
+        if (std.mem.eql(u8, "--noscript", opt)) {
+            noscript = true;
+            continue;
+        }
+
         if (try parseCommonArg(allocator, opt, args, &common)) {
             continue;
         }
@@ -471,6 +482,7 @@ fn parseFetchArgs(
         .url = url.?,
         .dump = dump,
         .common = common,
+        .noscript = noscript,
     };
 }
 

Original file line number	Diff line number	Diff line change
`@@ -110,13 +110,13 @@ pub const Element = struct {`
`110`	`110`
`111`	`111`	`pub fn get_innerHTML(self: parser.Element, page: Page) ![]const u8 {`
`112`	`112`	`var buf = std.ArrayList(u8).init(page.arena);`
`113`		`- try dump.writeChildren(parser.elementToNode(self), buf.writer());`
	`113`	`+ try dump.writeChildren(parser.elementToNode(self), .{}, buf.writer());`
`114`	`114`	`return buf.items;`
`115`	`115`	`}`
`116`	`116`
`117`	`117`	`pub fn get_outerHTML(self: parser.Element, page: Page) ![]const u8 {`
`118`	`118`	`var buf = std.ArrayList(u8).init(page.arena);`
`119`		`- try dump.writeNode(parser.elementToNode(self), buf.writer());`
	`119`	`+ try dump.writeNode(parser.elementToNode(self), .{}, buf.writer());`
`120`	`120`	`return buf.items;`
`121`	`121`	`}`
`122`	`122`
Original file line number	Diff line number	Diff line change
`@@ -142,15 +142,15 @@ pub const Page = struct {`
`142`	`142`	`}`
`143`	`143`
`144`	`144`	`// dump writes the page content into the given file.`
`145`		`- pub fn dump(self: *const Page, out: std.fs.File) !void {`
	`145`	`+ pub fn dump(self: *const Page, opts: Dump.Opts, out: std.fs.File) !void {`
`146`	`146`	`if (self.raw_data) \|raw_data\| {`
`147`	`147`	`// raw_data was set if the document was not HTML, dump the data content only.`
`148`	`148`	`return try out.writeAll(raw_data);`
`149`	`149`	`}`
`150`	`150`
`151`	`151`	`// if the page has a pointer to a document, dumps the HTML.`
`152`	`152`	`const doc = parser.documentHTMLToDocument(self.window.document);`
`153`		`- try Dump.writeHTML(doc, out);`
	`153`	`+ try Dump.writeHTML(doc, opts, out);`
`154`	`154`	`}`
`155`	`155`
`156`	`156`	`pub fn fetchModuleSource(ctx: *anyopaque, src: []const u8) !?[]const u8 {`
Original file line number	Diff line number	Diff line change
`@@ -36,9 +36,9 @@ pub const XMLSerializer = struct {`
`36`	`36`	`pub fn _serializeToString(_: const XMLSerializer, root: parser.Node, page: *Page) ![]const u8 {`
`37`	`37`	`var buf = std.ArrayList(u8).init(page.arena);`
`38`	`38`	`switch (try parser.nodeType(root)) {`
`39`		`- .document => try dump.writeHTML(@as(*parser.Document, @ptrCast(root)), buf.writer()),`
	`39`	`+ .document => try dump.writeHTML(@as(*parser.Document, @ptrCast(root)), .{}, buf.writer()),`
`40`	`40`	`.document_type => try dump.writeDocType(@as(*parser.DocumentType, @ptrCast(root)), buf.writer()),`
`41`		`- else => try dump.writeNode(root, buf.writer()),`
	`41`	`+ else => try dump.writeNode(root, .{}, buf.writer()),`
`42`	`42`	`}`
`43`	`43`	`return buf.items;`
`44`	`44`	`}`