feat: add DOM Parser

package.json

@@ -28,6 +28,20 @@
     "main": "dist/main/js/main.js",
     "types": "dist/main/js/main.d.ts",
     "unpkg": "dist/imsc.min.js",
+    "exports": {
+        ".": {
+            "types": "./dist/index.d.ts",
+            "import": "./dist/index.js"
+        },
+        "./*.js": {
+            "types": "./dist/*.d.ts",
+            "import": "./dist/*.js"
+        },
+        "./*": {
+            "types": "./dist/*.d.ts",
+            "import": "./dist/*.js"
+        }
+    },
     "scripts": {
         "prepublishOnly": "grunt build:release",
         "dev": "npx http-server build/public_html",
@@ -37,7 +51,7 @@
         "test": "node --test ./src/test/js/*Test.js"
     },
     "dependencies": {
-        "sax": "1.2.1"
+        "sax": "^1.4.1"
     },
     "devDependencies": {
         "@eslint/js": "^9.1.1",

src/main/js/doc.js

@@ -24,9 +24,9 @@
  * POSSIBILITY OF SUCH DAMAGE.
  */
 
-import sax from "sax";
 import { reportError, reportFatal, reportWarning } from "./error.js";
 import { ns_ebutts, ns_ittp, ns_itts, ns_tt, ns_ttp, ns_tts } from "./names.js";
+import { createDOMParser } from "./parser.js";
 import { byName, byQName } from "./styles.js";
 import { ComputedLength, hasOwnProperty, parseLength } from "./utils.js";
 
@@ -36,10 +36,8 @@ import { ComputedLength, hasOwnProperty, parseLength } from "./utils.js";
 
 /**
  * @typedef {import("./error").ErrorHandler} ErrorHandler
- */
-
-/**
- * @typedef {sax.Tag | sax.QualifiedTag} Node
+ * @typedef {import("./parser").Node} Node
+ * @typedef {import("./parser").Parser} Parser
  */
 
 /**
@@ -80,18 +78,18 @@ import { ComputedLength, hasOwnProperty, parseLength } from "./utils.js";
  * @param {string} xmlstring XML document
  * @param {ErrorHandler} errorHandler Error callback
  * @param {?MetadataHandler} metadataHandler Callback for <Metadata> elements
+ * @param {?Parser} parser XML parser
  * @returns {?TT} Opaque in-memory representation of an IMSC1 document
  */
 
-export function fromXML(xmlstring, errorHandler, metadataHandler) {
-    const p = sax.parser(true, { xmlns: true });
+export function fromXML(xmlstring, errorHandler, metadataHandler, parser = createDOMParser()) {
     const estack = [];
     const xmllangstack = [];
     const xmlspacestack = [];
     let metadata_depth = 0;
     let doc = null;
 
-    p.onclosetag = function () {
+    parser.onclosetag = function () {
 
         if (estack[0] instanceof Region) {
 
@@ -191,7 +189,7 @@ export function fromXML(xmlstring, errorHandler, metadataHandler) {
         estack.shift();
     };
 
-    p.ontext = function (str) {
+    parser.ontext = function (str) {
 
         if (estack[0] === undefined) {
 
@@ -234,7 +232,7 @@ export function fromXML(xmlstring, errorHandler, metadataHandler) {
 
     };
 
-    p.onopentag = function (node) {
+    parser.onopentag = function (node) {
 
         // maintain the xml:space stack
 
@@ -604,7 +602,7 @@ export function fromXML(xmlstring, errorHandler, metadataHandler) {
 
     // parse the document
 
-    p.write(xmlstring).close();
+    parser.write(xmlstring).close();
 
     // all referential styling has been flatten, so delete styles
 

src/main/js/main.js

@@ -27,3 +27,4 @@
 export { fromXML } from "./doc.js";
 export { renderHTML } from "./html.js";
 export { generateISD } from "./isd.js";
+export { createDOMParser, createSAXParser } from "./parser.js";

src/main/js/parser.js

@@ -0,0 +1,98 @@
+import sax from "sax";
+
+/**
+ * @typedef {sax.Tag | sax.QualifiedTag} Node
+ */
+
+/**
+ * @typedef {Object} Parser
+ * @property {(xml: string) => Parser} write
+ * @property {() => Parser} close
+ * @property {(node: Node) => void} onopentag
+ * @property {(text: string) => void} ontext
+ * @property {() => void} onclosetag
+ */
+
+export class XMLParser {
+  /**
+   * @param {Element} element
+   * @returns {SAX}
+   */
+  static toNode(element) {
+    const attrs = element.attributes;
+    const node = XMLParser.toNS(element);
+    node.attributes = {};
+
+    for (let i = 0, len = attrs.length; i < len; i++) {
+      const attr = attrs[i];
+      node.attributes[attr.name] = XMLParser.toNS(attr);
+    }
+
+    return node;
+  }
+
+  static toNS(node) {
+    return {
+      name: node.nodeName,
+      prefix: node.prefix,
+      local: node.localName,
+      uri: node.namespaceURI,
+      value: node.value,
+    };
+  }
+
+  onopentag = (node) => { console.log(node); }
+  ontext = (str) => { console.log(str); }
+  onclosetag = () => { }
+
+  write(xmlstring) {
+    const parser = new DOMParser();
+    const doc = parser.parseFromString(xmlstring, "application/xml");
+    const errorNode = doc.querySelector("parsererror");
+
+    if (errorNode) {
+      throw new Error("XML parsing error: " + errorNode.textContent);
+    }
+
+    this.process(doc.firstChild);
+
+    return this;
+  }
+
+  process(element) {
+    const node = XMLParser.toNode(element);
+    this.onopentag(node);
+
+    const children = element.childNodes;
+
+    for (let i = 0, len = children.length; i < len; i++) {
+      const child = children[i];
+
+      if (child.nodeType === Node.TEXT_NODE) {
+        this.ontext(child.textContent);
+      } else if (child.nodeType === Node.ELEMENT_NODE) {
+        this.process(child);
+      }
+    }
+
+    this.onclosetag();
+  }
+
+  close() {
+    return this;
+  }
+}
+
+/**
+ * @returns {Parser}
+ */
+export function createDOMParser() {
+  return new XMLParser();
+}
+
+/**
+ * @returns {Parser}
+ */
+export function createSAXParser() {
+  return sax.parser(true, { xmlns: true });
+}

src/test/js/utils/getIMSC1Document.js

@@ -1,5 +1,6 @@
 import fs from "node:fs/promises";
 import { fromXML } from "../../../main/js/doc.js";
+import { createSAXParser } from "../../../main/js/parser.js";
 
 const errorHandler = {
   info: function (msg) {
@@ -18,5 +19,5 @@ const errorHandler = {
 
 export async function getIMSC1Document(url, metadataHandler) {
   const contents = await fs.readFile(url, "utf8");
-  return fromXML(contents, errorHandler, metadataHandler);
+  return fromXML(contents, errorHandler, metadataHandler, createSAXParser());
 }

src/test/webapp/gen-renders.html

@@ -20,10 +20,14 @@
     <div id="main">
 
         <section id="start">
-            <button id="start-rendering" onclick="generateRenders('imsc-tests/imsc1')">Generate IMSC 1 reference
-                renders</button>
-            <button id="start-rendering" onclick="generateRenders('imsc-tests/imsc1_1')">Generate IMSC 1.1 reference
-                renders</button>
+            <button onclick="generateRenders('imsc-tests/imsc1', true)">Generate IMSC 1 reference
+                renders - SAX</button>
+            <button onclick="generateRenders('imsc-tests/imsc1_1', true)">Generate IMSC 1.1 reference
+                renders - SAX</button>
+            <button onclick="generateRenders('imsc-tests/imsc1', false)">Generate IMSC 1 reference
+                renders - DOM</button>
+            <button onclick="generateRenders('imsc-tests/imsc1_1', false)">Generate IMSC 1.1 reference
+                renders- DOM</button>
         </section>
 
         <section id="visual">

src/test/webapp/js/gen-renders.js

@@ -46,7 +46,7 @@ var errorHandler = {
 
 /* */
 
-function generateRenders(reffiles_root) {
+function generateRenders(reffiles_root, use_sax) {
 
     var zip = new JSZip();
 
@@ -61,7 +61,7 @@ function generateRenders(reffiles_root) {
 
             for (var i in finfos) {
 
-                p.push(asyncProcessRefFile(reffiles_root, renders_dir, pngs_dir, finfos[i]));
+                p.push(asyncProcessRefFile(reffiles_root, renders_dir, pngs_dir, finfos[i], use_sax));
 
             }
 
@@ -85,7 +85,7 @@ function generateRenders(reffiles_root) {
 
 }
 
-function asyncProcessRefFile(reffiles_root, renders_dir, pngs_dir, finfo) {
+function asyncProcessRefFile(reffiles_root, renders_dir, pngs_dir, finfo, use_sax) {
 
     var test_name = finfo.name || getTestName(finfo.path, finfo.params || {});
 
@@ -94,7 +94,8 @@ function asyncProcessRefFile(reffiles_root, renders_dir, pngs_dir, finfo) {
 
     return asyncLoadFile(getReferenceFilePath(reffiles_root, finfo.path))
         .then(function (contents) {
-            var doc = imsc.fromXML(contents.replace(/\r\n/g, '\n'), errorHandler);
+            var parser = use_sax ? imsc.createSAXParser() : imsc.createDOMParser();
+            var doc = imsc.fromXML(contents.replace(/\r\n/g, '\n'), errorHandler, parser);
 
             test_renders_dir.file("doc.json",
                 JSON.stringify(