Merge pull request #345 from lightpanda-io/modules

browser: support for modules

.github/actions/install/action.yml

@@ -17,7 +17,7 @@ inputs:
   zig-v8:
     description: 'zig v8 version to install'
     required: false
-    default: 'v0.1.9'
+    default: 'v0.1.11'
   v8:
     description: 'v8 version to install'
     required: false

src/browser/browser.zig

@@ -29,6 +29,7 @@ const Mime = @import("mime.zig");
 const jsruntime = @import("jsruntime");
 const Loop = jsruntime.Loop;
 const Env = jsruntime.Env;
+const Module = jsruntime.Module;
 
 const apiweb = @import("../apiweb.zig");
 
@@ -125,6 +126,21 @@ pub const Session = struct {
         try self.env.load(&self.jstypes);
     }
 
+    fn fetchModule(ctx: *anyopaque, referrer: ?jsruntime.Module, specifier: []const u8) !jsruntime.Module {
+        _ = referrer;
+
+        const self: *Session = @ptrCast(@alignCast(ctx));
+
+        if (self.page == null) return error.NoPage;
+
+        log.debug("fetch module: specifier: {s}", .{specifier});
+        const alloc = self.arena.allocator();
+        const body = try self.page.?.fetchData(alloc, specifier);
+        defer alloc.free(body);
+
+        return self.env.compileModule(body, specifier);
+    }
+
     fn deinit(self: *Session) void {
         if (self.page) |*p| p.end();
 
@@ -362,6 +378,9 @@ pub const Page = struct {
         log.debug("start js env", .{});
         try self.session.env.start();
 
+        // register the module loader
+        try self.session.env.setModuleLoadFn(self.session, Session.fetchModule);
+
         // load polyfills
         try polyfill.load(alloc, self.session.env);
 
@@ -388,7 +407,7 @@ pub const Page = struct {
         // sasync stores scripts which can be run asynchronously.
         // for now they are just run after the non-async one in order to
         // dispatch DOMContentLoaded the sooner as possible.
-        var sasync = std.ArrayList(*parser.Element).init(alloc);
+        var sasync = std.ArrayList(Script).init(alloc);
         defer sasync.deinit();
 
         const root = parser.documentToNode(doc);
@@ -403,21 +422,10 @@ pub const Page = struct {
             }
 
             const e = parser.nodeToElement(next.?);
-            const tag = try parser.elementHTMLGetTagType(@as(*parser.ElementHTML, @ptrCast(e)));
-
-            // ignore non-script tags
-            if (tag != .script) continue;
 
             // ignore non-js script.
-            // > type
-            // > Attribute is not set (default), an empty string, or a JavaScript MIME
-            // > type indicates that the script is a "classic script", containing
-            // > JavaScript code.
-            // https://developer.mozilla.org/en-US/docs/Web/HTML/Element/script#attribute_is_not_set_default_an_empty_string_or_a_javascript_mime_type
-            const stype = try parser.elementGetAttribute(e, "type");
-            if (!isJS(stype)) {
-                continue;
-            }
+            const script = try Script.init(e) orelse continue;
+            if (script.kind == .unknown) continue;
 
             // Ignore the defer attribute b/c we analyze all script
             // after the document has been parsed.
@@ -431,8 +439,8 @@ pub const Page = struct {
             // > then the classic script will be fetched in parallel to
             // > parsing and evaluated as soon as it is available.
             // https://developer.mozilla.org/en-US/docs/Web/HTML/Element/script#async
-            if (try parser.elementGetAttribute(e, "async") != null) {
-                try sasync.append(e);
+            if (script.isasync) {
+                try sasync.append(script);
                 continue;
             }
 
@@ -455,7 +463,7 @@ pub const Page = struct {
             // > page.
             // https://developer.mozilla.org/en-US/docs/Web/HTML/Element/script#notes
             try parser.documentHTMLSetCurrentScript(html_doc, @ptrCast(e));
-            self.evalScript(e) catch |err| log.warn("evaljs: {any}", .{err});
+            self.evalScript(script) catch |err| log.warn("evaljs: {any}", .{err});
             try parser.documentHTMLSetCurrentScript(html_doc, null);
         }
 
@@ -472,9 +480,9 @@ pub const Page = struct {
         _ = try parser.eventTargetDispatchEvent(parser.toEventTarget(parser.DocumentHTML, html_doc), evt);
 
         // eval async scripts.
-        for (sasync.items) |e| {
-            try parser.documentHTMLSetCurrentScript(html_doc, @ptrCast(e));
-            self.evalScript(e) catch |err| log.warn("evaljs: {any}", .{err});
+        for (sasync.items) |s| {
+            try parser.documentHTMLSetCurrentScript(html_doc, @ptrCast(s.element));
+            self.evalScript(s) catch |err| log.warn("evaljs: {any}", .{err});
             try parser.documentHTMLSetCurrentScript(html_doc, null);
         }
 
@@ -496,15 +504,15 @@ pub const Page = struct {
     // evalScript evaluates the src in priority.
     // if no src is present, we evaluate the text source.
     // https://html.spec.whatwg.org/multipage/scripting.html#script-processing-model
-    fn evalScript(self: *Page, e: *parser.Element) !void {
+    fn evalScript(self: *Page, s: Script) !void {
         const alloc = self.arena.allocator();
 
         // https://html.spec.whatwg.org/multipage/webappapis.html#fetch-a-classic-script
-        const opt_src = try parser.elementGetAttribute(e, "src");
+        const opt_src = try parser.elementGetAttribute(s.element, "src");
         if (opt_src) |src| {
             log.debug("starting GET {s}", .{src});
 
-            self.fetchScript(src) catch |err| {
+            self.fetchScript(s) catch |err| {
                 switch (err) {
                     FetchError.BadStatusCode => return err,
 
@@ -523,26 +531,10 @@ pub const Page = struct {
             return;
         }
 
-        var try_catch: jsruntime.TryCatch = undefined;
-        try_catch.init(self.session.env);
-        defer try_catch.deinit();
-
-        const opt_text = try parser.nodeTextContent(parser.elementToNode(e));
-        if (opt_text) |text| {
         // TODO handle charset attribute
-            const res = self.session.env.exec(text, "") catch {
-                if (try try_catch.err(alloc, self.session.env)) |msg| {
-                    defer alloc.free(msg);
-                    log.info("eval inline {s}: {s}", .{ text, msg });
-                }
-                return;
-            };
-
-            if (builtin.mode == .Debug) {
-                const msg = try res.toString(alloc, self.session.env);
-                defer alloc.free(msg);
-                log.debug("eval inline {s}", .{msg});
-            }
+        const opt_text = try parser.nodeTextContent(parser.elementToNode(s.element));
+        if (opt_text) |text| {
+            try s.eval(alloc, self.session.env, text);
             return;
         }
 
@@ -557,12 +549,9 @@ pub const Page = struct {
         JsErr,
     };
 
-    // fetchScript senf a GET request to the src and execute the script
-    // received.
-    fn fetchScript(self: *Page, src: []const u8) !void {
-        const alloc = self.arena.allocator();
-
-        log.debug("starting fetch script {s}", .{src});
+    // the caller owns the returned string
+    fn fetchData(self: *Page, alloc: std.mem.Allocator, src: []const u8) ![]const u8 {
+        log.debug("starting fetch {s}", .{src});
 
         var buffer: [1024]u8 = undefined;
         var b: []u8 = buffer[0..];
@@ -573,34 +562,54 @@ pub const Page = struct {
 
         const resp = fetchres.req.response;
 
-        log.info("fetch script {any}: {d}", .{ u, resp.status });
+        log.info("fetch {any}: {d}", .{ u, resp.status });
 
         if (resp.status != .ok) return FetchError.BadStatusCode;
 
         // TODO check content-type
         const body = try fetchres.req.reader().readAllAlloc(alloc, 16 * 1024 * 1024);
-        defer alloc.free(body);
 
         // check no body
         if (body.len == 0) return FetchError.NoBody;
 
-        var try_catch: jsruntime.TryCatch = undefined;
-        try_catch.init(self.session.env);
-        defer try_catch.deinit();
-
-        const res = self.session.env.exec(body, src) catch {
-            if (try try_catch.err(alloc, self.session.env)) |msg| {
-                defer alloc.free(msg);
-                log.info("eval remote {s}: {s}", .{ src, msg });
+        return body;
     }
-            return FetchError.JsErr;
+
+    // fetchScript senf a GET request to the src and execute the script
+    // received.
+    fn fetchScript(self: *Page, s: Script) !void {
+        const alloc = self.arena.allocator();
+        const body = try self.fetchData(alloc, s.src);
+        defer alloc.free(body);
+
+        try s.eval(alloc, self.session.env, body);
+    }
+
+    const Script = struct {
+        element: *parser.Element,
+        kind: Kind,
+        isasync: bool,
+
+        src: []const u8,
+
+        const Kind = enum {
+            unknown,
+            javascript,
+            module,
         };
 
-        if (builtin.mode == .Debug) {
-            const msg = try res.toString(alloc, self.session.env);
-            defer alloc.free(msg);
-            log.debug("eval remote {s}: {s}", .{ src, msg });
-        }
+        fn init(e: *parser.Element) !?Script {
+            // ignore non-script tags
+            const tag = try parser.elementHTMLGetTagType(@as(*parser.ElementHTML, @ptrCast(e)));
+            if (tag != .script) return null;
+
+            return .{
+                .element = e,
+                .kind = kind(try parser.elementGetAttribute(e, "type")),
+                .isasync = try parser.elementGetAttribute(e, "async") != null,
+
+                .src = try parser.elementGetAttribute(e, "src") orelse "inline",
+            };
         }
 
         // > type
@@ -608,11 +617,36 @@ pub const Page = struct {
         // > type indicates that the script is a "classic script", containing
         // > JavaScript code.
         // https://developer.mozilla.org/en-US/docs/Web/HTML/Element/script#attribute_is_not_set_default_an_empty_string_or_a_javascript_mime_type
-    fn isJS(stype: ?[]const u8) bool {
-        if (stype == null or stype.?.len == 0) return true;
-        if (std.mem.eql(u8, stype.?, "application/javascript")) return true;
-        if (!std.mem.eql(u8, stype.?, "module")) return true;
+        fn kind(stype: ?[]const u8) Kind {
+            if (stype == null or stype.?.len == 0) return .javascript;
+            if (std.mem.eql(u8, stype.?, "application/javascript")) return .javascript;
+            if (std.mem.eql(u8, stype.?, "module")) return .module;
 
-        return false;
+            return .unknown;
+        }
+
+        fn eval(self: Script, alloc: std.mem.Allocator, env: Env, body: []const u8) !void {
+            var try_catch: jsruntime.TryCatch = undefined;
+            try_catch.init(env);
+            defer try_catch.deinit();
+
+            const res = switch (self.kind) {
+                .unknown => return error.UnknownScript,
+                .javascript => env.exec(body, self.src),
+                .module => env.module(body, self.src),
+            } catch {
+                if (try try_catch.err(alloc, env)) |msg| {
+                    defer alloc.free(msg);
+                    log.info("eval script {s}: {s}", .{ self.src, msg });
+                }
+                return FetchError.JsErr;
+            };
+
+            if (builtin.mode == .Debug) {
+                const msg = try res.toString(alloc, env);
+                defer alloc.free(msg);
+                log.debug("eval script {s}: {s}", .{ self.src, msg });
+            }
         }
     };
+};

src/html/history.zig

@@ -0,0 +1,104 @@
+// Copyright (C) 2023-2024  Lightpanda (Selecy SAS)
+//
+// Francis Bouvier <francis@lightpanda.io>
+// Pierre Tachoire <pierre@lightpanda.io>
+//
+// 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 <https://www.gnu.org/licenses/>.
+
+const std = @import("std");
+
+const builtin = @import("builtin");
+const jsruntime = @import("jsruntime");
+
+const Case = jsruntime.test_utils.Case;
+const checkCases = jsruntime.test_utils.checkCases;
+
+// https://html.spec.whatwg.org/multipage/nav-history-apis.html#the-history-interface
+pub const History = struct {
+    pub const mem_guarantied = true;
+
+    const ScrollRestaurationMode = enum {
+        auto,
+        manual,
+    };
+
+    scrollRestauration: ScrollRestaurationMode = .audio,
+
+    pub fn get_length(_: *History) u64 {
+        return 0;
+    }
+
+    pub fn get_appCodeName(_: *Navigator) []const u8 {
+        return "Mozilla";
+    }
+    pub fn get_appName(_: *Navigator) []const u8 {
+        return "Netscape";
+    }
+    pub fn get_appVersion(self: *Navigator) []const u8 {
+        return self.version;
+    }
+    pub fn get_platform(self: *Navigator) []const u8 {
+        return self.platform;
+    }
+    pub fn get_product(_: *Navigator) []const u8 {
+        return "Gecko";
+    }
+    pub fn get_productSub(_: *Navigator) []const u8 {
+        return "20030107";
+    }
+    pub fn get_vendor(self: *Navigator) []const u8 {
+        return self.vendor;
+    }
+    pub fn get_vendorSub(_: *Navigator) []const u8 {
+        return "";
+    }
+    pub fn get_language(self: *Navigator) []const u8 {
+        return self.language;
+    }
+    // TODO wait for arrays.
+    //pub fn get_languages(self: *Navigator) [][]const u8 {
+    //    return .{self.language};
+    //}
+    pub fn get_online(_: *Navigator) bool {
+        return true;
+    }
+    pub fn _registerProtocolHandler(_: *Navigator, scheme: []const u8, url: []const u8) void {
+        _ = scheme;
+        _ = url;
+    }
+    pub fn _unregisterProtocolHandler(_: *Navigator, scheme: []const u8, url: []const u8) void {
+        _ = scheme;
+        _ = url;
+    }
+
+    pub fn get_cookieEnabled(_: *Navigator) bool {
+        return true;
+    }
+};
+
+// Tests
+// -----
+
+pub fn testExecFn(
+    _: std.mem.Allocator,
+    js_env: *jsruntime.Env,
+) anyerror!void {
+    var navigator = [_]Case{
+        .{ .src = "navigator.userAgent", .ex = "Lightpanda/1.0" },
+        .{ .src = "navigator.appVersion", .ex = "1.0" },
+        .{ .src = "navigator.language", .ex = "en-US" },
+    };
+    try checkCases(js_env, &navigator);
+}
+