JS -- support doc.info.authors

- get the authors from dc:creator in xmp metadata - doc.info.metadata is the raw string with xml code
mozilla · Jan 8, 2021 · e630956 · e630956
1 parent 35845d1
commit e630956
Show file tree

Hide file tree

Showing 8 changed files with 70 additions and 11 deletions.
diff --git a/src/display/api.js b/src/display/api.js
@@ -2705,6 +2705,7 @@ class WorkerTransport {
       .then(results => {
         return {
           info: results[0],
+          rawMetadata: results[1],
           metadata: results[1] ? new Metadata(results[1]) : null,
           contentDispositionFilename: this._fullReader?.filename ?? null,
           contentLength: this._fullReader?.contentLength ?? null,

diff --git a/src/display/metadata.js b/src/display/metadata.js
@@ -103,11 +103,33 @@ class Metadata {
       }
 
       for (let j = 0, jj = desc.childNodes.length; j < jj; j++) {
-        if (desc.childNodes[j].nodeName.toLowerCase() !== "#text") {
-          const entry = desc.childNodes[j];
+        const entry = desc.childNodes[j];
+        if (entry.nodeName.toLowerCase() !== "#text") {
           const name = entry.nodeName.toLowerCase();
-
-          this._metadataMap.set(name, entry.textContent.trim());
+          if (name === "dc:creator") {
+            if (!entry.childNodes.length) {
+              continue;
+            }
+            // Child must be a Bag (unordered array) or a Seq.
+            const authors = [];
+            const seqNode = entry.childNodes[0];
+            if (
+              seqNode.nodeName === "rdf:Bag" ||
+              seqNode.nodeName === "rdf:Seq"
+            ) {
+              for (let k = 0, kk = seqNode.childNodes.length; k < kk; k++) {
+                const authorNode = seqNode.childNodes[k];
+                if (authorNode.nodeName === "rdf:li") {
+                  authors.push(authorNode.textContent.trim());
+                }
+              }
+            }
+            if (authors.length > 0) {
+              this._metadataMap.set(name, authors);
+            }
+          } else {
+            this._metadataMap.set(name, entry.textContent.trim());
+          }
         }
       }
     }

diff --git a/src/scripting_api/doc.js b/src/scripting_api/doc.js
@@ -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",

diff --git a/test/integration/scripting_spec.js b/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");
+        })
+      );
+    });
+  });
 });
diff --git a/test/pdfs/.gitignore b/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

diff --git a/test/pdfs/js-authors.pdf b/test/pdfs/js-authors.pdf
diff --git a/test/test_manifest.json b/test/test_manifest.json
@@ -3213,6 +3213,12 @@
       "rounds": 1,
       "type": "eq"
     },
+    {  "id": "js-authors",
+       "file": "pdfs/js-authors.pdf",
+       "md5": "e892f290ae209286a29d3c17dfeab226",
+       "rounds": 1,
+       "type": "eq"
+    },
     {  "id": "issue1810",
       "file": "pdfs/issue1810.pdf",
       "md5": "b173a9dfb7bf00e1a298c6e8cb95c03e",

diff --git a/web/app.js b/web/app.js
@@ -1655,7 +1655,8 @@ const PDFViewerApplication = {
           baseURL: this.baseUrl,
           filesize: this._contentLength,
           filename: this._docFilename,
-          metadata: this.metadata,
+          metadata: this.rawMetadata,
+          authors: this.metadata.get("dc:creator") || null,
           numPages: pdfDocument.numPages,
           URL: this.url,
           actions: docActions,
@@ -1756,6 +1757,7 @@ const PDFViewerApplication = {
   async _initializeMetadata(pdfDocument) {
     const {
       info,
+      rawMetadata,
       metadata,
       contentDispositionFilename,
       contentLength,
@@ -1765,6 +1767,7 @@ const PDFViewerApplication = {
       return; // The document was closed while the metadata resolved.
     }
     this.documentInfo = info;
+    this.rawMetadata = rawMetadata;
     this.metadata = metadata;
     this._contentDispositionFilename = contentDispositionFilename;
     this._contentLength ??= contentLength; // See `getDownloadInfo`-call above.