Use the ccittfax decoder from pdfium

The decoder is a dependency of the jbig2 one and is already
included in pdf.js, so we just need to wire it up.
It improves the performance of documents using ccittfax images.

Author: calixteman

external/jbig2/jbig2.js

@@ -465,20 +465,33 @@ const blackTable3 = [
  * @param {Object} [options] - Decoding options.
  */
 class CCITTFaxDecoder {
-  constructor(source, options = {}) {
+  constructor(
+    source,
+    options = {
+      K: 0,
+      EndOfLine: false,
+      EncodedByteAlign: false,
+      Columns: 1728,
+      Rows: 0,
+      EndOfBlock: true,
+      BlackIs1: false,
+    }
+  ) {
     if (typeof source?.next !== "function") {
       throw new Error('CCITTFaxDecoder - invalid "source" parameter.');
     }
     this.source = source;
     this.eof = false;
 
-    this.encoding = options.K || 0;
-    this.eoline = options.EndOfLine || false;
-    this.byteAlign = options.EncodedByteAlign || false;
-    this.columns = options.Columns || 1728;
-    this.rows = options.Rows || 0;
-    this.eoblock = options.EndOfBlock ?? true;
-    this.black = options.BlackIs1 || false;
+    ({
+      K: this.encoding,
+      EndOfLine: this.eoline,
+      EncodedByteAlign: this.byteAlign,
+      Columns: this.columns,
+      Rows: this.rows,
+      EndOfBlock: this.eoblock,
+      BlackIs1: this.black,
+    } = options);
 
     this.codingLine = new Uint32Array(this.columns + 1);
     this.refLine = new Uint32Array(this.columns + 2);

external/jbig2/jbig2.wasm

@@ -13,51 +13,114 @@
  * limitations under the License.
  */
 
+import { shadow, warn } from "../shared/util.js";
 import { CCITTFaxDecoder } from "./ccitt.js";
 import { DecodeStream } from "./decode_stream.js";
 import { Dict } from "./primitives.js";
+import { JBig2CCITTFaxWasmImage } from "./jbig2_ccittFax_wasm.js";
 
 class CCITTFaxStream extends DecodeStream {
   constructor(str, maybeLength, params) {
     super(maybeLength);
 
     this.stream = str;
+    this.maybeLength = maybeLength;
     this.dict = str.dict;
 
     if (!(params instanceof Dict)) {
       params = Dict.empty;
     }
 
-    const source = {
-      next() {
-        return str.getByte();
-      },
+    this.params = {
+      K: params.get("K") || 0,
+      EndOfLine: !!params.get("EndOfLine"),
+      EncodedByteAlign: !!params.get("EncodedByteAlign"),
+      Columns: params.get("Columns") || 1728,
+      Rows: params.get("Rows") || 0,
+      EndOfBlock: !!(params.get("EndOfBlock") ?? true),
+      BlackIs1: !!params.get("BlackIs1"),
     };
-    this.ccittFaxDecoder = new CCITTFaxDecoder(source, {
-      K: params.get("K"),
-      EndOfLine: params.get("EndOfLine"),
-      EncodedByteAlign: params.get("EncodedByteAlign"),
-      Columns: params.get("Columns"),
-      Rows: params.get("Rows"),
-      EndOfBlock: params.get("EndOfBlock"),
-      BlackIs1: params.get("BlackIs1"),
-    });
+  }
+
+  get bytes() {
+    // If `this.maybeLength` is null, we'll get the entire stream.
+    return shadow(this, "bytes", this.stream.getBytes(this.maybeLength));
   }
 
   readBlock() {
+    this.decodeImageFallback();
+  }
+
+  get isImageStream() {
+    return true;
+  }
+
+  get isAsyncDecoder() {
+    return true;
+  }
+
+  async decodeImage(bytes, length, _decoderOptions) {
+    if (this.eof) {
+      return this.buffer;
+    }
+    if (!bytes) {
+      bytes = this.stream.isAsync
+        ? (await this.stream.asyncGetBytes()) || this.bytes
+        : this.bytes;
+    }
+
+    try {
+      this.buffer = await JBig2CCITTFaxWasmImage.decode(
+        bytes,
+        this.dict.get("W", "Width"),
+        this.dict.get("H", "Height"),
+        null,
+        this.params
+      );
+    } catch {
+      warn("CCITTFaxStream: Falling back to JS CCITTFax decoder.");
+      return this.decodeImageFallback(bytes, length);
+    }
+    this.bufferLength = this.buffer.length;
+    this.eof = true;
+
+    return this.buffer;
+  }
+
+  decodeImageFallback(bytes, length) {
+    if (this.eof) {
+      return this.buffer;
+    }
+    const { params } = this;
+    if (!bytes) {
+      this.stream.reset();
+      bytes = this.bytes;
+    }
+    let pos = 0;
+    const source = {
+      next() {
+        return bytes[pos++] ?? -1;
+      },
+    };
+    if (length && this.buffer.byteLength < length) {
+      this.buffer = new Uint8Array(length);
+    }
+    this.ccittFaxDecoder = new CCITTFaxDecoder(source, params);
+    let outPos = 0;
     while (!this.eof) {
       const c = this.ccittFaxDecoder.readNextChar();
       if (c === -1) {
         this.eof = true;
-        return;
+        break;
       }
-      this.ensureBuffer(this.bufferLength + 1);
-      this.buffer[this.bufferLength++] = c;
+      if (!length) {
+        this.ensureBuffer(outPos + 1);
+      }
+      this.buffer[outPos++] = c;
     }
-  }
 
-  get isImageStream() {
-    return true;
+    this.bufferLength = this.buffer.length;
+    return this.buffer.subarray(0, length || this.bufferLength);
   }
 }
 

src/core/ccitt.js

@@ -102,12 +102,12 @@ class DecodeStream extends BaseStream {
   async getImageData(length, decoderOptions) {
     if (!this.canAsyncDecodeImageFromBuffer) {
       if (this.isAsyncDecoder) {
-        return this.decodeImage(null, decoderOptions);
+        return this.decodeImage(null, length, decoderOptions);
       }
       return this.getBytes(length, decoderOptions);
     }
     const data = await this.stream.asyncGetBytes();
-    return this.decodeImage(data, decoderOptions);
+    return this.decodeImage(data, length, decoderOptions);
   }
 
   reset() {

src/core/ccitt_stream.js

@@ -23,7 +23,7 @@ class JBig2Error extends BaseException {
   }
 }
 
-class JBig2WasmImage {
+class JBig2CCITTFaxWasmImage {
   static #buffer = null;
 
   static #handler = null;
@@ -69,7 +69,7 @@ class JBig2WasmImage {
     }
   }
 
-  static async decode(bytes, width, height, globals) {
+  static async decode(bytes, width, height, globals, CCITTOptions) {
     if (!this.#modulePromise) {
       const { promise, resolve } = Promise.withResolvers();
       const promises = [promise];
@@ -95,13 +95,28 @@ class JBig2WasmImage {
       const size = bytes.length;
       ptr = module._malloc(size);
       module.writeArrayToMemory(bytes, ptr);
-      const globalsSize = globals ? globals.length : 0;
-      if (globalsSize > 0) {
-        globalsPtr = module._malloc(globalsSize);
-        module.writeArrayToMemory(globals, globalsPtr);
-      }
 
-      module._jbig2_decode(ptr, size, width, height, globalsPtr, globalsSize);
+      if (CCITTOptions) {
+        module._ccitt_decode(
+          ptr,
+          size,
+          width,
+          height,
+          CCITTOptions.K,
+          CCITTOptions.EndOfLine ? 1 : 0,
+          CCITTOptions.EncodedByteAlign ? 1 : 0,
+          CCITTOptions.BlackIs1 ? 1 : 0,
+          CCITTOptions.Columns,
+          CCITTOptions.Rows
+        );
+      } else {
+        const globalsSize = globals ? globals.length : 0;
+        if (globalsSize > 0) {
+          globalsPtr = module._malloc(globalsSize);
+          module.writeArrayToMemory(globals, globalsPtr);
+        }
+        module._jbig2_decode(ptr, size, width, height, globalsPtr, globalsSize);
+      }
       if (!module.imageData) {
         throw new JBig2Error("Unknown error");
       }
@@ -124,4 +139,4 @@ class JBig2WasmImage {
   }
 }
 
-export { JBig2Error, JBig2WasmImage };
+export { JBig2CCITTFaxWasmImage, JBig2Error };

src/core/decode_stream.js

@@ -17,8 +17,8 @@ import { shadow, warn } from "../shared/util.js";
 import { BaseStream } from "./base_stream.js";
 import { DecodeStream } from "./decode_stream.js";
 import { Dict } from "./primitives.js";
+import { JBig2CCITTFaxWasmImage } from "./jbig2_ccittFax_wasm.js";
 import { Jbig2Image } from "./jbig2.js";
-import { JBig2WasmImage } from "./jbig2_wasm.js";
 
 /**
  * For JBIG2's we use a library to decode these images and
@@ -45,7 +45,7 @@ class Jbig2Stream extends DecodeStream {
   }
 
   readBlock() {
-    this.decodeImage();
+    this.decodeImageFallback();
   }
 
   get isAsyncDecoder() {
@@ -56,7 +56,7 @@ class Jbig2Stream extends DecodeStream {
     return true;
   }
 
-  async decodeImage(bytes, _decoderOptions) {
+  async decodeImage(bytes, length, _decoderOptions) {
     if (this.eof) {
       return this.buffer;
     }
@@ -69,23 +69,23 @@ class Jbig2Stream extends DecodeStream {
           globals = globalsStream.getBytes();
         }
       }
-      this.buffer = await JBig2WasmImage.decode(
+      this.buffer = await JBig2CCITTFaxWasmImage.decode(
         bytes,
         this.dict.get("Width"),
         this.dict.get("Height"),
         globals
       );
     } catch {
       warn("Jbig2Stream: Falling back to JS JBIG2 decoder.");
-      return this.decodeImageFallback(bytes);
+      return this.decodeImageFallback(bytes, length);
     }
     this.bufferLength = this.buffer.length;
     this.eof = true;
 
     return this.buffer;
   }
 
-  async decodeImageFallback(bytes) {
+  decodeImageFallback(bytes, _length) {
     if (this.eof) {
       return this.buffer;
     }

src/core/jbig2_wasm.js src/core/jbig2_ccittFax_wasm.jssrc/core/jbig2_wasm.js renamed to src/core/jbig2_ccittFax_wasm.js

@@ -49,7 +49,7 @@ class JpxStream extends DecodeStream {
     return true;
   }
 
-  async decodeImage(bytes, decoderOptions) {
+  async decodeImage(bytes, _length, decoderOptions) {
     if (this.eof) {
       return this.buffer;
     }

src/core/jbig2_stream.js

@@ -22,7 +22,7 @@ import {
 } from "../shared/util.js";
 import { ChunkedStreamManager } from "./chunked_stream.js";
 import { ImageResizer } from "./image_resizer.js";
-import { JBig2WasmImage } from "./jbig2_wasm.js";
+import { JBig2CCITTFaxWasmImage } from "./jbig2_ccittFax_wasm.js";
 import { JpegStream } from "./jpeg_stream.js";
 import { JpxImage } from "./jpx.js";
 import { MissingDataException } from "./core_utils.js";
@@ -82,7 +82,7 @@ class BasePdfManager {
     JpxImage.setOptions(options);
     IccColorSpace.setOptions(options);
     CmykICCBasedCS.setOptions(options);
-    JBig2WasmImage.setOptions(options);
+    JBig2CCITTFaxWasmImage.setOptions(options);
   }
 
   get docId() {