[api-minor] Change the "dc:creator" Metadata field to an Array

- add scripting support for doc.info.authors
 - doc.info.metadata is the raw string with xml code

src/core/writer.js

@@ -130,9 +130,7 @@ function updateXFA(datasetsRef, newRefs, xref) {
   }
   const datasets = xref.fetchIfRef(datasetsRef);
   const str = bytesToString(datasets.getBytes());
-  const xml = new SimpleXMLParser(/* hasAttributes */ true).parseFromString(
-    str
-  );
+  const xml = new SimpleXMLParser({ hasAttributes: true }).parseFromString(str);
 
   for (const { xfa } of newRefs) {
     if (!xfa) {

src/display/metadata.js

@@ -24,7 +24,7 @@ class Metadata {
     data = this._repair(data);
 
     // Convert the string to an XML document.
-    const parser = new SimpleXMLParser();
+    const parser = new SimpleXMLParser({ lowerCaseName: true });
     const xmlDocument = parser.parseFromString(data);
 
     this._metadataMap = new Map();
@@ -32,6 +32,7 @@ class Metadata {
     if (xmlDocument) {
       this._parse(xmlDocument);
     }
+    this._data = data;
   }
 
   _repair(data) {
@@ -79,40 +80,71 @@ class Metadata {
       });
   }
 
+  _getSequence(entry) {
+    const name = entry.nodeName;
+    if (name !== "rdf:bag" && name !== "rdf:seq" && name !== "rdf:alt") {
+      return null;
+    }
+
+    return entry.childNodes.filter(node => node.nodeName === "rdf:li");
+  }
+
+  _getCreators(entry) {
+    if (entry.nodeName !== "dc:creator") {
+      return false;
+    }
+    if (!entry.hasChildNodes()) {
+      return true;
+    }
+
+    // Child must be a Bag (unordered array) or a Seq.
+    const seqNode = entry.childNodes[0];
+    const authors = this._getSequence(seqNode) || [];
+    this._metadataMap.set(
+      entry.nodeName,
+      authors.map(node => node.textContent.trim())
+    );
+
+    return true;
+  }
+
   _parse(xmlDocument) {
     let rdf = xmlDocument.documentElement;
 
-    if (rdf.nodeName.toLowerCase() !== "rdf:rdf") {
+    if (rdf.nodeName !== "rdf:rdf") {
       // Wrapped in <xmpmeta>
       rdf = rdf.firstChild;
-      while (rdf && rdf.nodeName.toLowerCase() !== "rdf:rdf") {
+      while (rdf && rdf.nodeName !== "rdf:rdf") {
         rdf = rdf.nextSibling;
       }
     }
 
-    const nodeName = rdf ? rdf.nodeName.toLowerCase() : null;
-    if (!rdf || nodeName !== "rdf:rdf" || !rdf.hasChildNodes()) {
+    if (!rdf || rdf.nodeName !== "rdf:rdf" || !rdf.hasChildNodes()) {
       return;
     }
 
-    const children = rdf.childNodes;
-    for (let i = 0, ii = children.length; i < ii; i++) {
-      const desc = children[i];
-      if (desc.nodeName.toLowerCase() !== "rdf:description") {
+    for (const desc of rdf.childNodes) {
+      if (desc.nodeName !== "rdf:description") {
         continue;
       }
 
-      for (let j = 0, jj = desc.childNodes.length; j < jj; j++) {
-        if (desc.childNodes[j].nodeName.toLowerCase() !== "#text") {
-          const entry = desc.childNodes[j];
-          const name = entry.nodeName.toLowerCase();
-
-          this._metadataMap.set(name, entry.textContent.trim());
+      for (const entry of desc.childNodes) {
+        const name = entry.nodeName;
+        if (name === "#text") {
+          continue;
         }
+        if (this._getCreators(entry)) {
+          continue;
+        }
+        this._metadataMap.set(name, entry.textContent.trim());
       }
     }
   }
 
+  getRaw() {
+    return this._data;
+  }
+
   get(name) {
     return this._metadataMap.has(name) ? this._metadataMap.get(name) : null;
   }

src/scripting_api/doc.js

@@ -42,7 +42,7 @@ class Doc extends PDFObject {
     this._dirty = false;
     this._disclosed = false;
     this._media = undefined;
-    this._metadata = data.metadata;
+    this._metadata = data.metadata || "";
     this._noautocomplete = undefined;
     this._nocache = undefined;
     this._spellDictionaryOrder = [];
@@ -74,12 +74,13 @@ class Doc extends PDFObject {
     // and they're are read-only.
     this._info = new Proxy(
       {
-        title: this.title,
-        author: this.author,
-        subject: this.subject,
-        keywords: this.keywords,
-        creator: this.creator,
-        producer: this.producer,
+        title: this._title,
+        author: this._author,
+        authors: data.authors || [this._author],
+        subject: this._subject,
+        keywords: this._keywords,
+        creator: this._creator,
+        producer: this._producer,
         creationdate: this._creationDate,
         moddate: this._modDate,
         trapped: data.Trapped || "Unknown",

src/shared/xml_parser.js

@@ -427,12 +427,13 @@ class SimpleDOMNode {
 }
 
 class SimpleXMLParser extends XMLParserBase {
-  constructor(hasAttributes = false) {
+  constructor({ hasAttributes = false, lowerCaseName = false }) {
     super();
     this._currentFragment = null;
     this._stack = null;
     this._errorCode = XMLParserErrorCode.NoError;
     this._hasAttributes = hasAttributes;
+    this._lowerCaseName = lowerCaseName;
   }
 
   parseFromString(data) {
@@ -476,6 +477,9 @@ class SimpleXMLParser extends XMLParserBase {
   }
 
   onBeginElement(name, attributes, isEmpty) {
+    if (this._lowerCaseName) {
+      name = name.toLowerCase();
+    }
     const node = new SimpleDOMNode(name);
     node.childNodes = [];
     if (this._hasAttributes) {

test/integration/scripting_spec.js

@@ -451,4 +451,29 @@ describe("Interaction", () => {
       );
     });
   });
+
+  describe("in js-authors.pdf", () => {
+    let pages;
+
+    beforeAll(async () => {
+      pages = await loadAndWait("js-authors.pdf", "#\\32 5R");
+    });
+
+    afterAll(async () => {
+      await closePages(pages);
+    });
+
+    it("must print authors in a text field", async () => {
+      await Promise.all(
+        pages.map(async ([browserName, page]) => {
+          const text = await actAndWaitForInput(page, "#\\32 5R", async () => {
+            await page.click("[data-annotation-id='26R']");
+          });
+          expect(text)
+            .withContext(`In ${browserName}`)
+            .toEqual("author1::author2::author3::author4::author5");
+        })
+      );
+    });
+  });
 });

test/pdfs/.gitignore

@@ -324,6 +324,7 @@
 !tensor-allflags-withfunction.pdf
 !issue10084_reduced.pdf
 !issue4246.pdf
+!js-authors.pdf
 !issue4461.pdf
 !issue4573.pdf
 !issue4722.pdf

test/unit/metadata_spec.js

@@ -96,7 +96,7 @@ describe("metadata", function () {
     expect(metadata.get("dc:qux")).toEqual(null);
 
     expect(metadata.getAll()).toEqual({
-      "dc:creator": "ODIS",
+      "dc:creator": ["ODIS"],
       "dc:title": "L'Odissee thématique logo Odisséé - décembre 2008.pub",
       "xap:creatortool": "PDFCreator Version 0.9.6",
     });
@@ -168,7 +168,7 @@ describe("metadata", function () {
     expect(metadata.get("dc:qux")).toEqual(null);
 
     expect(metadata.getAll()).toEqual({
-      "dc:creator": "",
+      "dc:creator": [""],
       "dc:description": "",
       "dc:format": "application/pdf",
       "dc:subject": "",

test/unit/xml_spec.js

@@ -47,8 +47,9 @@ describe("XML", function () {
               <g a="121110"/>
           </b>
       </a>`;
-      const root = new SimpleXMLParser(true).parseFromString(xml)
-        .documentElement;
+      const root = new SimpleXMLParser({ hasAttributes: true }).parseFromString(
+        xml
+      ).documentElement;
       function getAttr(path) {
         return root.searchNode(parseXFAPath(path), 0).attributes[0].value;
       }
@@ -96,8 +97,9 @@ describe("XML", function () {
               <g a="121110"/>
           </b>
       </a>`;
-      const root = new SimpleXMLParser(true).parseFromString(xml)
-        .documentElement;
+      const root = new SimpleXMLParser({ hasAttributes: true }).parseFromString(
+        xml
+      ).documentElement;
       const buffer = [];
       root.dump(buffer);
 

web/app.js

@@ -1655,7 +1655,8 @@ const PDFViewerApplication = {
           baseURL: this.baseUrl,
           filesize: this._contentLength,
           filename: this._docFilename,
-          metadata: this.metadata,
+          metadata: this.metadata?.getRaw(),
+          authors: this.metadata?.get("dc:creator"),
           numPages: pdfDocument.numPages,
           URL: this.url,
           actions: docActions,