Fix issues in text selection

  - PR mozilla#13257 fixed a lot of issues but not all and this patch aims to fix almost all remaining issues.
  - the idea in this new patch is to compare position of new glyph with the last position where a glyph has been drawn;
    - no space are "drawn": it just moves the cursor but they aren't added in the chunk;
    - so this way a space followed by a cursor move can be treated as only one space: it helps to merge all spaces into one.
  - to make difference between real spaces and tracking ones, we used a factor of the space width (from the font)
    - it was a pretty good idea in general but it fails with some fonts where space was too big:
    - in Poppler, they're using a factor of the font size: this is an excellent idea (<= 0.1 * fontSize implies tracking space).

Author: calixteman

src/core/evaluator.js

@@ -185,10 +185,11 @@ function appendText(task, geom, styles, ctx) {
   let shouldScaleText = false;
   if (
     geom.str.length > 1 ||
-    (task._enhanceTextSelection && AllWhitespaceRegexp.test(geom.str))
+    (task._enhanceTextSelection &&
+      (geom.isOneSpace || AllWhitespaceRegexp.test(geom.str)))
   ) {
     shouldScaleText = true;
-  } else if (geom.transform[0] !== geom.transform[3]) {
+  } else if (!geom.isOneSpace && geom.transform[0] !== geom.transform[3]) {
     const absScaleX = Math.abs(geom.transform[0]),
       absScaleY = Math.abs(geom.transform[3]);
     // When the horizontal/vertical scaling differs significantly, also scale

src/display/text_layer.js

@@ -13,6 +13,7 @@
 !issue1155r.pdf
 !issue2017r.pdf
 !bug1727053.pdf
+!issue11913.pdf
 !issue2391-1.pdf
 !issue2391-2.pdf
 !issue14046.pdf
@@ -182,6 +183,7 @@
 !issue11931.pdf
 !issue1655r.pdf
 !issue6541.pdf
+!issue10640.pdf
 !issue2948.pdf
 !issue6231_1.pdf
 !issue10402.pdf
@@ -285,6 +287,7 @@
 !issue2840.pdf
 !issue4061.pdf
 !issue4668.pdf
+!issue13226.pdf
 !PDFJS-7562-reduced.pdf
 !issue11768_reduced.pdf
 !issue5039.pdf
@@ -440,6 +443,7 @@
 !annotation-fileattachment.pdf
 !annotation-text-widget.pdf
 !annotation-choice-widget.pdf
+!issue10900.pdf
 !annotation-button-widget.pdf
 !annotation-polyline-polygon.pdf
 !annotation-polyline-polygon-without-appearance.pdf
@@ -462,6 +466,7 @@
 !issue9972-3.pdf
 !tiling-pattern-box.pdf
 !tiling-pattern-large-steps.pdf
+!issue13201.pdf
 !issue11555.pdf
 !issue12337.pdf
 !pr12564.pdf

test/pdfs/.gitignore

@@ -73,6 +73,10 @@ describe("api", function () {
     }, WAIT_TIMEOUT);
   }
 
+  function mergeText(items) {
+    return items.map(chunk => chunk.str + (chunk.hasEOL ? "\n" : "")).join("");
+  }
+
   describe("getDocument", function () {
     it("creates pdf doc from URL-string", async function () {
       const urlStr = TEST_PDFS_PATH + basicApiFileName;
@@ -1604,11 +1608,17 @@ describe("api", function () {
       const data = await Promise.all([defaultPromise, parametersPromise]);
 
       expect(!!data[0].items).toEqual(true);
-      expect(data[0].items.length).toEqual(12);
+      expect(data[0].items.length).toEqual(11);
       expect(!!data[0].styles).toEqual(true);
 
+      const page1 = mergeText(data[0].items);
+      expect(page1).toEqual(`Table Of Content
+Chapter 1 .......................................................... 2
+Paragraph 1.1 ...................................................... 3
+page 1 / 3`);
+
       expect(!!data[1].items).toEqual(true);
-      expect(data[1].items.length).toEqual(7);
+      expect(data[1].items.length).toEqual(6);
       expect(!!data[1].styles).toEqual(true);
     });
 
@@ -1632,6 +1642,7 @@ describe("api", function () {
         transform: [18, 0, 0, 18, 441.81, 708.4499999999999],
         width: 77.49,
         hasEOL: false,
+        isOneSpace: false,
       });
       expect(styles[fontName]).toEqual({
         fontFamily: "serif",
@@ -1643,6 +1654,107 @@ describe("api", function () {
       await loadingTask.destroy();
     });
 
+    it("gets text content, with no extra spaces (issue 13226)", async function () {
+      const loadingTask = getDocument(buildGetDocumentParams("issue13226.pdf"));
+      const pdfDoc = await loadingTask.promise;
+      const pdfPage = await pdfDoc.getPage(1);
+      const { items } = await pdfPage.getTextContent();
+      const text = mergeText(items);
+
+      expect(text).toEqual(
+        "Mitarbeiterinnen und Mitarbeiter arbeiten in über 100 Ländern engagiert im Dienste"
+      );
+
+      await loadingTask.destroy();
+    });
+
+    it("gets text content, with merged spaces (issue 13201)", async function () {
+      const loadingTask = getDocument(buildGetDocumentParams("issue13201.pdf"));
+      const pdfDoc = await loadingTask.promise;
+      const pdfPage = await pdfDoc.getPage(1);
+      const { items } = await pdfPage.getTextContent();
+      const text = mergeText(items);
+
+      expect(
+        text.includes(
+          "Abstract. A purely peer-to-peer version of electronic cash would allow online"
+        )
+      ).toEqual(true);
+      expect(
+        text.includes(
+          "avoid mediating disputes. The cost of mediation increases transaction costs, limiting the"
+        )
+      ).toEqual(true);
+      expect(
+        text.includes(
+          "system is secure as long as honest nodes collectively control more CPU power than any"
+        )
+      ).toEqual(true);
+
+      await loadingTask.destroy();
+    });
+
+    it("gets text content, with no spaces between letters of words (issue 11913)", async function () {
+      const loadingTask = getDocument(buildGetDocumentParams("issue11913.pdf"));
+      const pdfDoc = await loadingTask.promise;
+      const pdfPage = await pdfDoc.getPage(1);
+      const { items } = await pdfPage.getTextContent();
+      const text = mergeText(items);
+
+      expect(
+        text.includes(
+          "1. The first of these cases arises from the tragic handicap which has blighted the life of the Plaintiff, and from the response of the"
+        )
+      ).toEqual(true);
+      expect(
+        text.includes(
+          "argued in this Court the appeal raises narrower, but important, issues which may be summarised as follows:-"
+        )
+      ).toEqual(true);
+      await loadingTask.destroy();
+    });
+
+    it("gets text content, with merged spaces (issue 10900)", async function () {
+      const loadingTask = getDocument(buildGetDocumentParams("issue10900.pdf"));
+      const pdfDoc = await loadingTask.promise;
+      const pdfPage = await pdfDoc.getPage(1);
+      const { items } = await pdfPage.getTextContent();
+      const text = mergeText(items);
+
+      expect(
+        text.includes(`3 3 3 3
+851.5 854.9 839.3 837.5
+633.6 727.8 789.9 796.2
+1,485.1 1,582.7 1,629.2 1,633.7
+114.2 121.7 125.3 130.7
+13.0x 13.0x 13.0x 12.5x`)
+      ).toEqual(true);
+
+      await loadingTask.destroy();
+    });
+
+    it("gets text content, with spaces (issue 10640)", async function () {
+      const loadingTask = getDocument(buildGetDocumentParams("issue10640.pdf"));
+      const pdfDoc = await loadingTask.promise;
+      const pdfPage = await pdfDoc.getPage(1);
+      const { items } = await pdfPage.getTextContent();
+      const text = mergeText(items);
+
+      expect(
+        text.includes(`Open Sans is a humanist sans serif typeface designed by Steve Matteson.
+Open Sans was designed with an upright stress, open forms and a neu-
+tral, yet friendly appearance. It was optimized for print, web, and mobile
+interfaces, and has excellent legibility characteristics in its letterforms (see
+figure \x81 on the following page). This font is available from the Google Font
+Directory [\x81] as TrueType files licensed under the Apache License version \x82.\x80.
+This package provides support for this font in LATEX. It includes Type \x81
+versions of the fonts, converted for this package using FontForge from its
+sources, for full support with Dvips.`)
+      ).toEqual(true);
+
+      await loadingTask.destroy();
+    });
+
     it("gets empty structure tree", async function () {
       const tree = await page.getStructTree();
 

test/pdfs/issue10640.pdf

@@ -268,7 +268,7 @@ describe("pdf_find_controller", function () {
         pageIndex: 0,
         matchIndex: 0,
       },
-      pageMatches: [[19, 48, 66]],
+      pageMatches: [[19, 46, 62]],
       pageMatchesLength: [[8, 8, 8]],
     });
   });