src/core/annotation.js

@@ -49,8 +49,6 @@ import {
   lookupNormalRect,
   lookupRect,
   numberToString,
-  RESOURCES_KEYS_OPERATOR_LIST,
-  RESOURCES_KEYS_TEXT_CONTENT,
   stringToAsciiOrUTF16BE,
   stringToUTF16String,
 } from "./core_utils.js";
@@ -1198,7 +1196,7 @@ class Annotation {
 
     const appearanceDict = appearance.dict;
     const resources = await this.loadResources(
-      RESOURCES_KEYS_OPERATOR_LIST,
+      ["ExtGState", "ColorSpace", "Pattern", "Shading", "XObject", "Font"],
       appearance
     );
     const bbox = lookupRect(appearanceDict.getArray("BBox"), [0, 0, 1, 1]);
@@ -1259,7 +1257,7 @@ class Annotation {
     }
 
     const resources = await this.loadResources(
-      RESOURCES_KEYS_TEXT_CONTENT,
+      ["ExtGState", "Font", "Properties", "XObject"],
       this.appearance
     );
 

src/core/core_utils.js

@@ -32,23 +32,6 @@ const MIN_INT_32 = -(2 ** 31);
 
 const IDENTITY_MATRIX = [1, 0, 0, 1, 0, 0];
 
-const RESOURCES_KEYS_OPERATOR_LIST = [
-  "ColorSpace",
-  "ExtGState",
-  "Font",
-  "Pattern",
-  "Properties",
-  "Shading",
-  "XObject",
-];
-
-const RESOURCES_KEYS_TEXT_CONTENT = [
-  "ExtGState",
-  "Font",
-  "Properties",
-  "XObject",
-];
-
 function getLookupTableFactory(initializer) {
   let lookup;
   return function () {
@@ -762,8 +745,6 @@ export {
   readUint16,
   readUint32,
   recoverJsURL,
-  RESOURCES_KEYS_OPERATOR_LIST,
-  RESOURCES_KEYS_TEXT_CONTENT,
   stringToAsciiOrUTF16BE,
   stringToUTF16HexString,
   stringToUTF16String,

src/core/document.js

@@ -45,8 +45,6 @@ import {
   lookupNormalRect,
   MissingDataException,
   PDF_VERSION_REGEXP,
-  RESOURCES_KEYS_OPERATOR_LIST,
-  RESOURCES_KEYS_TEXT_CONTENT,
   validateCSSFont,
   XRefEntryException,
   XRefParseException,
@@ -421,25 +419,6 @@ class Page {
     await objectLoader.load();
   }
 
-  async #getMergedResources(streamDict, keys) {
-    // In rare cases /Resources are also found in the /Contents stream-dict,
-    // in addition to in the /Page dict, hence we need to prefer those when
-    // available (see issue18894.pdf).
-    const localResources = streamDict?.get("Resources");
-
-    if (!(localResources instanceof Dict)) {
-      return this.resources;
-    }
-    const objectLoader = new ObjectLoader(localResources, keys, this.xref);
-    await objectLoader.load();
-
-    return Dict.merge({
-      xref: this.xref,
-      dictArray: [localResources, this.resources],
-      mergeSubDicts: true,
-    });
-  }
-
   async getOperatorList({
     handler,
     sink,
@@ -450,7 +429,15 @@ class Page {
     modifiedIds = null,
   }) {
     const contentStreamPromise = this.getContentStream();
-    const resourcesPromise = this.loadResources(RESOURCES_KEYS_OPERATOR_LIST);
+    const resourcesPromise = this.loadResources([
+      "ColorSpace",
+      "ExtGState",
+      "Font",
+      "Pattern",
+      "Properties",
+      "Shading",
+      "XObject",
+    ]);
 
     const partialEvaluator = new PartialEvaluator({
       xref: this.xref,
@@ -538,15 +525,11 @@ class Page {
       contentStreamPromise,
       resourcesPromise,
     ]).then(async ([contentStream]) => {
-      const resources = await this.#getMergedResources(
-        contentStream.dict,
-        RESOURCES_KEYS_OPERATOR_LIST
-      );
       const opList = new OperatorList(intent, sink);
 
       handler.send("StartRenderPage", {
         transparency: partialEvaluator.hasBlendModes(
-          resources,
+          this.resources,
           this.nonBlendModesSet
         ),
         pageIndex: this.pageIndex,
@@ -556,7 +539,7 @@ class Page {
       await partialEvaluator.getOperatorList({
         stream: contentStream,
         task,
-        resources,
+        resources: this.resources,
         operatorList: opList,
       });
       return opList;
@@ -659,7 +642,12 @@ class Page {
     sink,
   }) {
     const contentStreamPromise = this.getContentStream();
-    const resourcesPromise = this.loadResources(RESOURCES_KEYS_TEXT_CONTENT);
+    const resourcesPromise = this.loadResources([
+      "ExtGState",
+      "Font",
+      "Properties",
+      "XObject",
+    ]);
     const langPromise = this.pdfManager.ensureCatalog("lang");
 
     const [contentStream, , lang] = await Promise.all([
@@ -667,11 +655,6 @@ class Page {
       resourcesPromise,
       langPromise,
     ]);
-    const resources = await this.#getMergedResources(
-      contentStream.dict,
-      RESOURCES_KEYS_TEXT_CONTENT
-    );
-
     const partialEvaluator = new PartialEvaluator({
       xref: this.xref,
       handler,
@@ -689,7 +672,7 @@ class Page {
     return partialEvaluator.getTextContent({
       stream: contentStream,
       task,
-      resources,
+      resources: this.resources,
       includeMarkedContent,
       disableNormalization,
       sink,

src/core/evaluator.js

@@ -462,8 +462,7 @@ class PartialEvaluator {
     operatorList,
     task,
     initialState,
-    localColorSpaceCache,
-    seenRefs
+    localColorSpaceCache
   ) {
     const dict = xobj.dict;
     const matrix = lookupMatrix(dict.getArray("Matrix"), null);
@@ -527,7 +526,6 @@ class PartialEvaluator {
       resources: dict.get("Resources") || resources,
       operatorList,
       initialState,
-      prevRefs: seenRefs,
     });
     operatorList.addOp(OPS.paintFormXObjectEnd, []);
 
@@ -852,8 +850,7 @@ class PartialEvaluator {
     operatorList,
     task,
     stateManager,
-    localColorSpaceCache,
-    seenRefs
+    localColorSpaceCache
   ) {
     const smaskContent = smask.get("G");
     const smaskOptions = {
@@ -883,8 +880,7 @@ class PartialEvaluator {
       operatorList,
       task,
       stateManager.state.clone({ newPath: true }),
-      localColorSpaceCache,
-      seenRefs
+      localColorSpaceCache
     );
   }
 
@@ -1069,7 +1065,6 @@ class PartialEvaluator {
     stateManager,
     localGStateCache,
     localColorSpaceCache,
-    seenRefs,
   }) {
     const gStateRef = gState.objId;
     let isSimpleGState = true;
@@ -1132,8 +1127,7 @@ class PartialEvaluator {
                 operatorList,
                 task,
                 stateManager,
-                localColorSpaceCache,
-                seenRefs
+                localColorSpaceCache
               )
             );
             gStateObj.push([key, true]);
@@ -1702,19 +1696,7 @@ class PartialEvaluator {
     operatorList,
     initialState = null,
     fallbackFontDict = null,
-    prevRefs = null,
   }) {
-    const objId = stream.dict?.objId;
-    const seenRefs = new RefSet(prevRefs);
-
-    if (objId) {
-      if (prevRefs?.has(objId)) {
-        throw new Error(
-          `getOperatorList - ignoring circular reference: ${objId}`
-        );
-      }
-      seenRefs.put(objId);
-    }
     // Ensure that `resources`/`initialState` is correctly initialized,
     // even if the provided parameter is e.g. `null`.
     resources ||= Dict.empty;
@@ -1826,8 +1808,7 @@ class PartialEvaluator {
                       operatorList,
                       task,
                       stateManager.state.clone({ newPath: true }),
-                      localColorSpaceCache,
-                      seenRefs
+                      localColorSpaceCache
                     )
                     .then(function () {
                       stateManager.restore();
@@ -2177,7 +2158,6 @@ class PartialEvaluator {
                     stateManager,
                     localGStateCache,
                     localColorSpaceCache,
-                    seenRefs,
                   })
                   .then(resolveGState, rejectGState);
               }).catch(function (reason) {
@@ -2359,19 +2339,7 @@ class PartialEvaluator {
     markedContentData = null,
     disableNormalization = false,
     keepWhiteSpace = false,
-    prevRefs = null,
   }) {
-    const objId = stream.dict?.objId;
-    const seenRefs = new RefSet(prevRefs);
-
-    if (objId) {
-      if (prevRefs?.has(objId)) {
-        throw new Error(
-          `getTextContent - ignoring circular reference: ${objId}`
-        );
-      }
-      seenRefs.put(objId);
-    }
     // Ensure that `resources`/`stateManager` is correctly initialized,
     // even if the provided parameter is e.g. `null`.
     resources ||= Dict.empty;
@@ -3358,7 +3326,6 @@ class PartialEvaluator {
                     markedContentData,
                     disableNormalization,
                     keepWhiteSpace,
-                    prevRefs: seenRefs,
                   })
                   .then(function () {
                     if (!sinkWrapper.enqueueInvoked) {

src/core/fonts.js

@@ -1108,10 +1108,9 @@ class Font {
           // Repair the TrueType file. It is can be damaged in the point of
           // view of the sanitizer
           data = this.checkAndRepair(name, file, properties);
-
-          adjustWidths(properties);
-
           if (this.isOpenType) {
+            adjustWidths(properties);
+
             type = "OpenType";
           }
           break;
@@ -2646,6 +2645,8 @@ class Font {
         cffFile = new Stream(tables["CFF "].data);
         cff = new CFFFont(cffFile, properties);
 
+        adjustWidths(properties);
+
         return this.convert(name, cff, properties);
       }
 

test/pdfs/.gitignore

@@ -28,7 +28,6 @@
 !checkbox-bad-appearance.pdf
 !issue4684.pdf
 !issue8092.pdf
-!issue19802.pdf
 !issue5256.pdf
 !issue5801.pdf
 !issue5946.pdf
@@ -206,7 +205,6 @@
 !issue3928.pdf
 !issue8565.pdf
 !clippath.pdf
-!issue19800.pdf
 !issue8795_reduced.pdf
 !bug1755507.pdf
 !close-path-bug.pdf

test/pdfs/issue19800.pdf

@@ -1,90 +0,0 @@
-%PDF-1.4
-1 0 obj
-<<
- /Type /Catalog
- /Outlines 2 0 R
- /Pages 3 0 R
->>
-endobj
-2 0 obj
-<<
- /Type /Outlines
- /Count 0
->>
-endobj
-3 0 obj
-<<
- /Type /Pages
- /Kids [ 4 0 R ]
- /Count 1
->>
-endobj
-4 0 obj
-<<
- /Type /Page
- /Parent 3 0 R
- /MediaBox [ 0 0 500 300 ]
- /Contents 5 0 R
- /Resources <<
-  /ProcSet [ /PDF /Text ]
-  /XObject << /X1 6 0 R >>
- >>
->>
-endobj
-5 0 obj
-<< /Length 24 >>
-stream
-1 0 0 1 25 25 cm /X1 Do
-endstream
-endobj
-6 0 obj
-<< /Subtype /Form
-   /BBox [0 0 1000 1000]
-   /Length 61
-   /Resources <<
-     /ProcSet [ /PDF /Text ]
-     /Font << /F1 7 0 R >>
-     /XObject << /X0 8 0 R >>
-   >>
->>
-stream
-BT
-/F1 24 Tf
-(Hello world) Tj
-ET
-0.5 0 0 0.5 25 25 cm /X0 Do
-endstream
-endobj
-7 0 obj
-<<
- /Type /Font
- /Subtype /Type1
- /Name /F1
- /BaseFont /Helvetica
- /Encoding /MacRomanEncoding
->>
-endobj
-8 0 obj
-<< /Subtype /Form
-   /BBox [0 0 1000 1000]
-   /Length 61
-   /Resources <<
-     /ProcSet [ /PDF /Text ]
-     /Font << /F1 7 0 R >>
-     /XObject << /X1 6 0 R >>
-   >>
->>
-stream
-BT
-/F1 24 Tf
-(Hello world) Tj
-ET
-0.5 0 0 0.5 25 25 cm /X1 Do
-endstream
-endobj
-trailer 
-<<
- /Size 8
- /Root 1 0 R
->>
-%%EOF

test/test_manifest.json

@@ -3871,13 +3871,6 @@
     "annotations": true,
     "about": "LinkAnnotation with a relative link, and a /Catalog Base-URI."
   },
-  {
-    "id": "issue19802",
-    "file": "pdfs/issue19802.pdf",
-    "md5": "7a431fb7951ad605f676c80636bbe5d3",
-    "rounds": 1,
-    "type": "eq"
-  },
   {
     "id": "issue14824",
     "file": "pdfs/issue14824.pdf",
@@ -6897,20 +6890,6 @@
     "lastPage": 2,
     "type": "eq"
   },
-  {
-    "id": "issue19800-eq",
-    "file": "pdfs/issue19800.pdf",
-    "md5": "92825d3178196bdd01096c4081609efd",
-    "rounds": 1,
-    "type": "eq"
-  },
-  {
-    "id": "issue19800-text",
-    "file": "pdfs/issue19800.pdf",
-    "md5": "92825d3178196bdd01096c4081609efd",
-    "rounds": 1,
-    "type": "text"
-  },
   {
     "id": "issue3438",
     "file": "pdfs/issue3438.pdf",