src/core/chunked_stream.js

@@ -14,7 +14,7 @@
  */
 
 import { arrayBuffersToBytes, MissingDataException } from "./core_utils.js";
-import { assert, MathClamp } from "../shared/util.js";
+import { assert } from "../shared/util.js";
 import { Stream } from "./stream.js";
 
 class ChunkedStream extends Stream {
@@ -70,12 +70,6 @@ class ChunkedStream extends Stream {
       throw new Error(`Bad end offset: ${end}`);
     }
 
-    if (typeof PDFJSDev === "undefined" || PDFJSDev.test("TESTING")) {
-      assert(
-        chunk instanceof ArrayBuffer,
-        "onReceiveData - expected an ArrayBuffer."
-      );
-    }
     this.bytes.set(new Uint8Array(chunk), begin);
     const beginChunk = Math.floor(begin / chunkSize);
     const endChunk = Math.floor((end - 1) / chunkSize) + 1;
@@ -91,12 +85,6 @@ class ChunkedStream extends Stream {
     let position = this.progressiveDataLength;
     const beginChunk = Math.floor(position / this.chunkSize);
 
-    if (typeof PDFJSDev === "undefined" || PDFJSDev.test("TESTING")) {
-      assert(
-        data instanceof ArrayBuffer,
-        "onReceiveProgressiveData - expected an ArrayBuffer."
-      );
-    }
     this.bytes.set(new Uint8Array(data), position);
     position += data.byteLength;
     this.progressiveDataLength = position;
@@ -322,7 +310,7 @@ class ChunkedStreamManager {
       if (this.aborted) {
         return; // Ignoring any data after abort.
       }
-      this.onReceiveData({ chunk: data.buffer, begin });
+      this.onReceiveData({ chunk: data, begin });
     });
   }
 
@@ -523,11 +511,7 @@ class ChunkedStreamManager {
     }
 
     this.msgHandler.send("DocProgress", {
-      loaded: MathClamp(
+      loaded: stream.numChunksLoaded * chunkSize,
-        stream.numChunksLoaded * chunkSize,
-        stream.progressiveDataLength,
-        length
-      ),
       total: length,
     });
   }

src/core/evaluator.js

@@ -2538,7 +2538,7 @@ class PartialEvaluator {
 
     const preprocessor = new EvaluatorPreprocessor(stream, xref, stateManager);
 
-    let textState, currentTextState;
+    let textState;
 
     function pushWhitespace({
       width = 0,
@@ -2800,9 +2800,7 @@ class PartialEvaluator {
 
         // When the total height of the current chunk is negative
         // then we're writing from bottom to top.
-        const textOrientation = Math.sign(
+        const textOrientation = Math.sign(textContentItem.height);
-          textContentItem.height || textContentItem.totalHeight
-        );
         if (advanceY < textOrientation * textContentItem.negativeSpaceMax) {
           if (
             Math.abs(advanceX) >
@@ -2866,9 +2864,7 @@ class PartialEvaluator {
 
       // When the total width of the current chunk is negative
       // then we're writing from right to left.
-      const textOrientation = Math.sign(
+      const textOrientation = Math.sign(textContentItem.width);
-        textContentItem.width || textContentItem.totalWidth
-      );
       if (advanceX < textOrientation * textContentItem.negativeSpaceMax) {
         if (
           Math.abs(advanceY) >
@@ -2926,15 +2922,6 @@ class PartialEvaluator {
     }
 
     function buildTextContentItem({ chars, extraSpacing }) {
-      if (
-        currentTextState !== textState &&
-        (currentTextState.fontName !== textState.fontName ||
-          currentTextState.fontSize !== textState.fontSize)
-      ) {
-        flushTextContentItem();
-        currentTextState = textState.clone();
-      }
-
       const font = textState.font;
       if (!chars) {
         // Just move according to the space we have.
@@ -3190,8 +3177,8 @@ class PartialEvaluator {
           break;
         }
 
+        const previousState = textState;
         textState = stateManager.state;
-        currentTextState ||= textState.clone();
         const fn = operation.fn;
         args = operation.args;
 
@@ -3208,6 +3195,7 @@ class PartialEvaluator {
               break;
             }
 
+            flushTextContentItem();
             textState.fontName = fontNameArg;
             textState.fontSize = fontSizeArg;
             next(handleSetFont(fontNameArg, null));
@@ -3564,10 +3552,14 @@ class PartialEvaluator {
             }
             break;
           case OPS.restore:
-            stateManager.restore();
+            if (
-            break;
+              previousState &&
-          case OPS.save:
+              (previousState.font !== textState.font ||
-            stateManager.save();
+                previousState.fontSize !== textState.fontSize ||
+                previousState.fontName !== textState.fontName)
+            ) {
+              flushTextContentItem();
+            }
             break;
         } // switch
         if (textContent.items.length >= (sink?.desiredSize ?? 1)) {
@@ -5091,7 +5083,7 @@ class TextState {
   }
 
   clone() {
-    const clone = Object.assign(Object.create(this), this);
+    const clone = Object.create(this);
     clone.textMatrix = this.textMatrix.slice();
     clone.textLineMatrix = this.textLineMatrix.slice();
     clone.fontMatrix = this.fontMatrix.slice();

test/pdfs/.gitignore

@@ -871,4 +871,3 @@
 !page_with_number.pdf
 !page_with_number_and_link.pdf
 !Brotli-Prototype-FileA.pdf
-!bug2013793.pdf

test/unit/api_spec.js

@@ -4069,29 +4069,6 @@ Caron Broadcasting, Inc., an Ohio corporation (“Lessee”).`)
       await loadingTask.destroy();
     });
 
-    it("gets text content with some fake font changes (bug 2013793)", async function () {
-      const loadingTask = getDocument(buildGetDocumentParams("bug2013793.pdf"));
-      const pdfDoc = await loadingTask.promise;
-      const pdfPage = await pdfDoc.getPage(1);
-      const { items } = await pdfPage.getTextContent({
-        disableNormalization: true,
-      });
-      const text = mergeText(items);
-      expect(text)
-        .toEqual(`This is a great deal of nothing. The purpose is to help in identifying a bug when the PDF
-is read by Firefox. I want to know whether any of the two words in this paragraph run
-together. If they do, I will file a bug report. The problem seems to occur somewhere
-between the 240th and 260th character in the paragraph. I should have written that much
-by now. So, here’s to squashing bugs.
-This is a great deal of nothing. The purpose is to help in identifying a bug when the
-PDF is read by Firefox. I want to know whether any of the two words in this
-paragraph run together. If they do, I will file a bug report. The problem seems to
-occur somewhere between the 240th and 260th character in the paragraph. I should
-have written that much by now. So, here’s to squashing bugs.`);
-
-      await loadingTask.destroy();
-    });
-
     it("gets empty structure tree", async function () {
       const tree = await page.getStructTree();