[api-minor] Add a jpx decoder based on OpenJPEG 2.5.2

The decoder is compiled in WASM:
https://github.com/mozilla/pdf.js.openjpeg

Fixes mozilla#17289, mozilla#17061, mozilla#16485, mozilla#13051, mozilla#6365, mozilla#4648, mozilla#12213.

.eslintignore

@@ -6,6 +6,7 @@ external/bcmaps/
 external/builder/fixtures/
 external/builder/fixtures_esprima/
 external/quickjs/
+external/openjpeg/
 test/tmp/
 test/pdfs/
 web/locale/

external/openjpeg/README.md

@@ -0,0 +1,12 @@
+## Build
+
+In order to generate the file `openjpeg.js`:
+* git clone https://github.com/mozilla/pdf.js.openjpeg/
+* the build requires to have a [Docker](https://www.docker.com/) setup and then:
+  * `node build.js -C` to build the Docker image
+  * `node build.js -co /pdf.js/external/openjpeg/` to compile the decoder
+
+## Licensing
+
+[OpenJPEG](https://www.openjpeg.org/) is under [BSD 2-clause "Simplified" License](https://github.com/uclouvain/openjpeg/blob/master/LICENSE)
+and [pdf.js.openjpeg](https://github.com/mozilla/pdf.js.openjpeg/) is released under [Apache 2](https://github.com/mozilla/pdf.js.openjpeg/blob/main/LICENSE) license so `openjpeg.js` is released under [Apache 2](https://github.com/mozilla/pdf.js.openjpeg/blob/main/LICENSE) license too.

gulpfile.mjs

@@ -1603,6 +1603,7 @@ function buildLib(defines, dir) {
     ),
     gulp.src(["web/*.js", "!web/{pdfjs,viewer}.js"], { base: "." }),
     gulp.src("test/unit/*.js", { base: "." }),
+    gulp.src("external/openjpeg/*.js", { base: "openjpeg/" }),
   ]);
 
   return buildLibHelper(bundleDefines, inputStream, dir);

src/core/cleanup_helper.js

@@ -16,11 +16,16 @@
 import { clearPatternCaches } from "./pattern.js";
 import { clearPrimitiveCaches } from "./primitives.js";
 import { clearUnicodeCaches } from "./unicode.js";
+import { JpxImage } from "./jpx.js";
 
 function clearGlobalCaches() {
   clearPatternCaches();
   clearPrimitiveCaches();
   clearUnicodeCaches();
+
+  // Remove the global `JpxImage` instance, since it may hold a reference to
+  // the WebAssembly module.
+  JpxImage.cleanup();
 }
 
 export { clearGlobalCaches };

src/core/decode_stream.js

@@ -73,7 +73,7 @@ class DecodeStream extends BaseStream {
     return this.buffer[this.pos++];
   }
 
-  getBytes(length) {
+  getBytes(length, ignoreColorSpace = false) {
     const pos = this.pos;
     let end;
 
@@ -82,15 +82,15 @@ class DecodeStream extends BaseStream {
       end = pos + length;
 
       while (!this.eof && this.bufferLength < end) {
-        this.readBlock();
+        this.readBlock(ignoreColorSpace);
       }
       const bufEnd = this.bufferLength;
       if (end > bufEnd) {
         end = bufEnd;
       }
     } else {
       while (!this.eof) {
-        this.readBlock();
+        this.readBlock(ignoreColorSpace);
       }
       end = this.bufferLength;
     }

src/core/image.js

@@ -104,6 +104,7 @@ class PDFImage {
     localColorSpaceCache,
   }) {
     this.image = image;
+    let jpxDecode = false;
     const dict = image.dict;
 
     const filter = dict.get("F", "Filter");
@@ -118,14 +119,14 @@ class PDFImage {
     }
     switch (filterName) {
       case "JPXDecode":
-        const jpxImage = new JpxImage();
-        jpxImage.parseImageProperties(image.stream);
+        ({
+          width: image.width,
+          height: image.height,
+          componentsCount: image.numComps,
+          bitsPerComponent: image.bitsPerComponent,
+        } = JpxImage.parseImageProperties(image.stream));
         image.stream.reset();
-
-        image.width = jpxImage.width;
-        image.height = jpxImage.height;
-        image.bitsPerComponent = jpxImage.bitsPerComponent;
-        image.numComps = jpxImage.componentsCount;
+        jpxDecode = true;
         break;
       case "JBIG2Decode":
         image.bitsPerComponent = 1;
@@ -197,6 +198,7 @@ class PDFImage {
             );
         }
       }
+
       this.colorSpace = ColorSpace.parse({
         cs: colorSpace,
         xref,
@@ -205,6 +207,10 @@ class PDFImage {
         localColorSpaceCache,
       });
       this.numComps = this.colorSpace.numComps;
+
+      // If the jpx image has a color space then it musn't be used in order to
+      // be able to use the color space that comes from the pdf.
+      this.ignoreColorSpace = jpxDecode && this.colorSpace.name === "Indexed";
     }
 
     this.decode = dict.getArray("D", "Decode");
@@ -984,7 +990,7 @@ class PDFImage {
     this.image.drawHeight = drawHeight || this.height;
     this.image.forceRGBA = !!forceRGBA;
     this.image.forceRGB = !!forceRGB;
-    const imageBytes = this.image.getBytes(length);
+    const imageBytes = this.image.getBytes(length, this.ignoreColorSpace);
 
     // If imageBytes came from a DecodeStream, we're safe to transfer it
     // (and thus detach its underlying buffer) because it will constitute