diff --git a/src/dom/document.zig b/src/dom/document.zig
index b889169b..f869f8e5 100644
--- a/src/dom/document.zig
+++ b/src/dom/document.zig
@@ -7,7 +7,8 @@ const Case = jsruntime.test_utils.Case;
 const checkCases = jsruntime.test_utils.checkCases;
 
 const Node = @import("node.zig").Node;
-const HTMLCollection = @import("html_collection.zig").HTMLCollection;
+
+const collection = @import("html_collection.zig");
 
 const Element = @import("element.zig").Element;
 const ElementUnion = @import("element.zig").Union;
@@ -49,12 +50,9 @@ pub const Document = struct {
     // the spec changed to return an HTMLCollection instead.
     // That's why we reimplemented getElementsByTagName by using an
     // HTMLCollection in zig here.
-    pub fn _getElementsByTagName(self: *parser.Document, tag_name: []const u8) HTMLCollection {
+    pub fn _getElementsByTagName(self: *parser.Document, allocator: std.mem.Allocator, tag_name: []const u8) !collection.HTMLCollection {
         const root = parser.documentGetDocumentElement(self);
-        return HTMLCollection{
-            .root = parser.elementToNode(root),
-            .match = tag_name,
-        };
+        return collection.HTMLCollectionByTagName(allocator, parser.elementToNode(root), tag_name);
     }
 };
 
diff --git a/src/dom/html_collection.zig b/src/dom/html_collection.zig
index 10cfb78f..c9e02b22 100644
--- a/src/dom/html_collection.zig
+++ b/src/dom/html_collection.zig
@@ -5,11 +5,59 @@ const parser = @import("../netsurf.zig");
 const jsruntime = @import("jsruntime");
 const Case = jsruntime.test_utils.Case;
 const checkCases = jsruntime.test_utils.checkCases;
+const generate = @import("../generate.zig");
 
 const utils = @import("utils.z");
 const Element = @import("element.zig").Element;
 const Union = @import("element.zig").Union;
 
+const Match = union(enum) {
+    matchByTagName: MatchByTagName,
+
+    pub fn match(self: Match, node: *parser.Node) bool {
+        switch (self) {
+            inline else => |case| return case.match(node),
+        }
+    }
+
+    pub fn deinit(self: Match, allocator: std.mem.Allocator) void {
+        switch (self) {
+            inline else => |case| case.deinit(allocator),
+        }
+    }
+};
+
+pub const MatchByTagName = struct {
+    // tag is used to select node against their name.
+    // tag comparison is case insensitive.
+    tag: []const u8,
+    is_wildcard: bool,
+
+    fn init(allocator: std.mem.Allocator, tag_name: []const u8) !MatchByTagName {
+        return MatchByTagName{
+            .tag = try std.ascii.allocUpperString(allocator, tag_name),
+            .is_wildcard = std.mem.eql(u8, tag_name, "*"),
+        };
+    }
+
+    fn deinit(self: *MatchByTagName, allocator: std.mem.Allocator) void {
+        allocator.free(self.tag);
+    }
+
+    pub fn match(self: MatchByTagName, node: *parser.Node) bool {
+        return self.is_wildcard or std.mem.eql(u8, self.tag, parser.nodeName(node));
+    }
+};
+
+pub fn HTMLCollectionByTagName(allocator: std.mem.Allocator, root: *parser.Node, tag_name: []const u8) !HTMLCollection {
+    return HTMLCollection{
+        .root = root,
+        .match = Match{
+            .matchByTagName = try MatchByTagName.init(allocator, tag_name),
+        },
+    };
+}
+
 // WEB IDL https://dom.spec.whatwg.org/#htmlcollection
 // HTMLCollection is re implemented in zig here because libdom
 // dom_html_collection expects a comparison function callback as arguement.
@@ -17,10 +65,9 @@ const Union = @import("element.zig").Union;
 pub const HTMLCollection = struct {
     pub const mem_guarantied = true;
 
+    match: Match,
+
     root: *parser.Node,
-    // match is used to select node against their name.
-    // match comparison is case insensitive.
-    match: []const u8,
 
     // save a state for the collection to improve the _item speed.
     cur_idx: ?u32 = undefined,
@@ -76,20 +123,15 @@ pub const HTMLCollection = struct {
     /// get_length computes the collection's length dynamically according to
     /// the current root structure.
     // TODO: nodes retrieved must be de-referenced.
-    pub fn get_length(self: *HTMLCollection, allocator: std.mem.Allocator) !u32 {
+    pub fn get_length(self: *HTMLCollection) u32 {
         var len: u32 = 0;
         var node: *parser.Node = self.root;
         var ntype: parser.NodeType = undefined;
 
-        const imatch = try std.ascii.allocUpperString(allocator, self.match);
-        defer allocator.free(imatch);
-
-        const is_wildcard = std.mem.eql(u8, self.match, "*");
-
         while (true) {
             ntype = parser.nodeType(node);
             if (ntype == .element) {
-                if (is_wildcard or std.mem.eql(u8, imatch, parser.nodeName(node))) {
+                if (self.match.match(node)) {
                     len += 1;
                 }
             }
@@ -100,26 +142,21 @@ pub const HTMLCollection = struct {
         return len;
     }
 
-    pub fn _item(self: *HTMLCollection, allocator: std.mem.Allocator, index: u32) !?Union {
+    pub fn _item(self: *HTMLCollection, index: u32) ?Union {
         var i: u32 = 0;
         var node: *parser.Node = self.root;
         var ntype: parser.NodeType = undefined;
 
-        const is_wildcard = std.mem.eql(u8, self.match, "*");
-
         // Use the current state to improve speed if possible.
         if (self.cur_idx != null and index >= self.cur_idx.?) {
             i = self.cur_idx.?;
             node = self.cur_node.?;
         }
 
-        const imatch = try std.ascii.allocUpperString(allocator, self.match);
-        defer allocator.free(imatch);
-
         while (true) {
             ntype = parser.nodeType(node);
             if (ntype == .element) {
-                if (is_wildcard or std.mem.eql(u8, imatch, parser.nodeName(node))) {
+                if (self.match.match(node)) {
                     // check if we found the searched element.
                     if (i == index) {
                         // save the current state
@@ -140,7 +177,7 @@ pub const HTMLCollection = struct {
         return null;
     }
 
-    pub fn _namedItem(self: *HTMLCollection, allocator: std.mem.Allocator, name: []const u8) !?Union {
+    pub fn _namedItem(self: *HTMLCollection, name: []const u8) ?Union {
         if (name.len == 0) {
             return null;
         }
@@ -148,15 +185,10 @@ pub const HTMLCollection = struct {
         var node: *parser.Node = self.root;
         var ntype: parser.NodeType = undefined;
 
-        const is_wildcard = std.mem.eql(u8, self.match, "*");
-
-        const imatch = try std.ascii.allocUpperString(allocator, self.match);
-        defer allocator.free(imatch);
-
         while (true) {
             ntype = parser.nodeType(node);
             if (ntype == .element) {
-                if (is_wildcard or std.mem.eql(u8, imatch, parser.nodeName(node))) {
+                if (self.match.match(node)) {
                     const elem = @as(*parser.Element, @ptrCast(node));
 
                     var attr = parser.elementGetAttribute(elem, "id");