Fix langs TS typing (#570)

* Added webp test case per #548

* Added gif test per #499

README.md

@@ -12,13 +12,6 @@
 [![Downloads Total](https://img.shields.io/npm/dt/tesseract.js.svg)](https://www.npmjs.com/package/tesseract.js)
 [![Downloads Month](https://img.shields.io/npm/dm/tesseract.js.svg)](https://www.npmjs.com/package/tesseract.js)
 
-<h3 align="center">
-  Version 2 is now available and under development in the master branch, read a story about v2: <a href="https://jeromewu.github.io/why-i-refactor-tesseract.js-v2/">Why I refactor tesseract.js v2?</a><br>
-  Check the <a href="https://github.com/naptha/tesseract.js/tree/support/1.x">support/1.x</a> branch for version 1
-</h3>
-
-<br>
-
 Tesseract.js is a javascript library that gets words in [almost any language](./docs/tesseract_lang_list.md) out of images. ([Demo](http://tesseract.projectnaptha.com/))
 
 Image Recognition
@@ -69,6 +62,16 @@ const worker = createWorker({
 
 [Check out the docs](#documentation) for a full explanation of the API.
 
+## Major changes in v3
+- Significantly faster performance
+   - Runtime reduction of 84% for Browser and 96% for Node.js when recognizing the [example images](./examples/data)
+- Upgrade to Tesseract v5.1.0 (using emscripten 3.1.18)
+- Added SIMD-enabled build for supported devices
+- Added support:
+   - Node.js version 18
+- Removed support:
+   - ASM.js version, any other old versions of Tesseract.js-core (<3.0.0) 
+   - Node.js versions 10 and 12
 
 ## Major changes in v2
 - Upgrade to tesseract v4.1.1 (using emscripten 1.39.10 upstream)
@@ -77,7 +80,8 @@ const worker = createWorker({
 - Support WebAssembly (fallback to ASM.js when browser doesn't support)
 - Support Typescript
 
-
+Read a story about v2: <a href="https://jeromewu.github.io/why-i-refactor-tesseract.js-v2/">Why I refactor tesseract.js v2?</a><br>
+  Check the <a href="https://github.com/naptha/tesseract.js/tree/support/1.x">support/1.x</a> branch for version 1
 ## Installation
 Tesseract.js works with a `<script>` tag via local copy or CDN, with webpack via `npm` and on Node.js with `npm/yarn`.
 
@@ -95,16 +99,16 @@ After including the script the `Tesseract` variable will be globally available.
 
 ### Node.js
 
-**Tesseract.js currently requires Node.js v6.8.0 or higher**
+**Tesseract.js v3 requires Node.js v14 or higher**
 
 ```shell
-# For v2
+# For v3
 npm install tesseract.js
 yarn add tesseract.js
 
-# For v1
-npm install tesseract.js@1
-yarn add tesseract.js@1
+# For v2
+npm install tesseract.js@2
+yarn add tesseract.js@2
 ```
 
 

docs/image-format.md

@@ -1,17 +1,18 @@
 # Image Format
 
-Support Format: **bmp, jpg, png, pbm**
+The main Tesseract.js functions (ex. recognize, detect) take an `image` parameter.  The image formats and data types supported are listed below. 
 
-The main Tesseract.js functions (ex. recognize, detect) take an `image` parameter, which should be something that is like an image. What's considered "image-like" differs depending on whether it is being run from the browser or through NodeJS.
+Support Image Formats: **bmp, jpg, png, pbm, webp**
 
-On a browser, an image can be:
-- an `img`, `video`, or `canvas` element
-- a `File` object (from a file `<input>`)
-- a `Blob` object
-- a path or URL to an accessible image
-- a base64 encoded image fits `data:image\/([a-zA-Z]*);base64,([^"]*)` regexp
+For browser and Node, supported data types are:
+ - string with base64 encoded image (fits `data:image\/([a-zA-Z]*);base64,([^"]*)` regexp)
+ - buffer
 
-In Node.js, an image can be
-- a path to a local image
-- a Buffer storing binary image 
-- a base64 encoded image fits `data:image\/([a-zA-Z]*);base64,([^"]*)` regexp
+For browser only, supported data types are:
+ - `File` or `Blob` object
+ - `img` or `canvas` element
+
+For Node only, supported data types are:
+ - string containing a path to local image
+
+Note: images must be a supported image format **and** a supported data type.  For example, a buffer containing a png image is supported.  A buffer containing raw pixel data is not supported. 

examples/browser/demo.html

@@ -71,7 +71,6 @@ async function recognizeFile(file) {
 <option value='meme'     > Internet Meme                </option>
 <option value='epo'     > Esperanto             </option>
 <option value='epo_alt' > Esperanto alternative </option>
-<option value='equ'     > Math                  </option>
 <option value='est'     > Estonian              </option>
 <option value='eus'     > Basque                </option>
 <option value='fin'     > Finnish               </option>

package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "tesseract.js",
-  "version": "2.1.5",
+  "version": "3.0.3",
   "lockfileVersion": 2,
   "requires": true,
   "packages": {
     "": {
       "name": "tesseract.js",
-      "version": "2.1.5",
+      "version": "3.0.3",
       "hasInstallScript": true,
       "license": "Apache-2.0",
       "dependencies": {
@@ -20,12 +20,14 @@
         "opencollective-postinstall": "^2.0.2",
         "regenerator-runtime": "^0.13.3",
         "resolve-url": "^0.2.1",
-        "tesseract.js-core": "^3.0.1",
+        "tesseract.js-core": "^3.0.2",
+        "wasm-feature-detect": "^1.2.11",
         "zlibjs": "^0.3.1"
       },
       "devDependencies": {
         "@babel/core": "^7.18.7",
         "@babel/preset-env": "^7.18.7",
+        "@rollup/plugin-commonjs": "^22.0.2",
         "acorn": "^6.4.0",
         "babel-loader": "^8.2.0",
         "buffer": "^6.0.3",
@@ -40,6 +42,7 @@
         "npm-run-all": "^4.1.5",
         "nyc": "^15.1.0",
         "rimraf": "^2.7.1",
+        "rollup": "^2.79.0",
         "wait-on": "^3.3.0",
         "webpack": "^5.74.0",
         "webpack-bundle-analyzer": "^4.6.0",
@@ -2058,6 +2061,62 @@
       "integrity": "sha512-a5Sab1C4/icpTZVzZc5Ghpz88yQtGOyNqYXcZgOssB2uuAr+wF/MvN6bgtW32q7HHrvBki+BsZ0OuNv6EV3K9g==",
       "dev": true
     },
+    "node_modules/@rollup/plugin-commonjs": {
+      "version": "22.0.2",
+      "resolved": "https://registry.npmjs.org/@rollup/plugin-commonjs/-/plugin-commonjs-22.0.2.tgz",
+      "integrity": "sha512-//NdP6iIwPbMTcazYsiBMbJW7gfmpHom33u1beiIoHDEM0Q9clvtQB1T0efvMqHeKsGohiHo97BCPCkBXdscwg==",
+      "dev": true,
+      "dependencies": {
+        "@rollup/pluginutils": "^3.1.0",
+        "commondir": "^1.0.1",
+        "estree-walker": "^2.0.1",
+        "glob": "^7.1.6",
+        "is-reference": "^1.2.1",
+        "magic-string": "^0.25.7",
+        "resolve": "^1.17.0"
+      },
+      "engines": {
+        "node": ">= 12.0.0"
+      },
+      "peerDependencies": {
+        "rollup": "^2.68.0"
+      }
+    },
+    "node_modules/@rollup/plugin-commonjs/node_modules/estree-walker": {
+      "version": "2.0.2",
+      "resolved": "https://registry.npmjs.org/estree-walker/-/estree-walker-2.0.2.tgz",
+      "integrity": "sha512-Rfkk/Mp/DL7JVje3u18FxFujQlTNR2q6QfMSMB7AvCBx91NGj/ba3kCfza0f6dVDbw7YlRf/nDrn7pQrCCyQ/w==",
+      "dev": true
+    },
+    "node_modules/@rollup/pluginutils": {
+      "version": "3.1.0",
+      "resolved": "https://registry.npmjs.org/@rollup/pluginutils/-/pluginutils-3.1.0.tgz",
+      "integrity": "sha512-GksZ6pr6TpIjHm8h9lSQ8pi8BE9VeubNT0OMJ3B5uZJ8pz73NPiqOtCog/x2/QzM1ENChPKxMDhiQuRHsqc+lg==",
+      "dev": true,
+      "dependencies": {
+        "@types/estree": "0.0.39",
+        "estree-walker": "^1.0.1",
+        "picomatch": "^2.2.2"
+      },
+      "engines": {
+        "node": ">= 8.0.0"
+      },
+      "peerDependencies": {
+        "rollup": "^1.20.0||^2.0.0"
+      }
+    },
+    "node_modules/@rollup/pluginutils/node_modules/@types/estree": {
+      "version": "0.0.39",
+      "resolved": "https://registry.npmjs.org/@types/estree/-/estree-0.0.39.tgz",
+      "integrity": "sha512-EYNwp3bU+98cpU4lAWYYL7Zz+2gryWH1qbdDTidVd6hkiR6weksdbMadyXKXNPEkQFhXM+hVO9ZygomHXp+AIw==",
+      "dev": true
+    },
+    "node_modules/@rollup/pluginutils/node_modules/estree-walker": {
+      "version": "1.0.1",
+      "resolved": "https://registry.npmjs.org/estree-walker/-/estree-walker-1.0.1.tgz",
+      "integrity": "sha512-1fMXF3YP4pZZVozF8j/ZLfvnR8NSIljt56UhbZ5PeeDmmGHpgpdwQt7ITlGvYaQukCvuBRMLEiKiYC+oeIg4cg==",
+      "dev": true
+    },
     "node_modules/@types/eslint": {
       "version": "8.4.6",
       "resolved": "https://registry.npmjs.org/@types/eslint/-/eslint-8.4.6.tgz",
@@ -5364,6 +5423,15 @@
         "node": ">=0.10.0"
       }
     },
+    "node_modules/is-reference": {
+      "version": "1.2.1",
+      "resolved": "https://registry.npmjs.org/is-reference/-/is-reference-1.2.1.tgz",
+      "integrity": "sha512-U82MsXXiFIrjCK4otLT+o2NA2Cd2g5MLoOVXUZjIOhLurrRxpEXzI8O0KZHr3IjLvlAH1kTPYSuqer5T9ZVBKQ==",
+      "dev": true,
+      "dependencies": {
+        "@types/estree": "*"
+      }
+    },
     "node_modules/is-regex": {
       "version": "1.1.4",
       "resolved": "https://registry.npmjs.org/is-regex/-/is-regex-1.1.4.tgz",
@@ -6070,6 +6138,15 @@
         "node": ">=8"
       }
     },
+    "node_modules/magic-string": {
+      "version": "0.25.9",
+      "resolved": "https://registry.npmjs.org/magic-string/-/magic-string-0.25.9.tgz",
+      "integrity": "sha512-RmF0AsMzgt25qzqqLc1+MbHmhdx0ojF2Fvs4XnOqz2ZOBXzzkEwc/dJQZCYHAn7v1jbVOjAZfK8msRn4BxO4VQ==",
+      "dev": true,
+      "dependencies": {
+        "sourcemap-codec": "^1.4.8"
+      }
+    },
     "node_modules/make-dir": {
       "version": "3.1.0",
       "resolved": "https://registry.npmjs.org/make-dir/-/make-dir-3.1.0.tgz",
@@ -7724,6 +7801,35 @@
         "url": "https://github.com/sponsors/isaacs"
       }
     },
+    "node_modules/rollup": {
+      "version": "2.79.0",
+      "resolved": "https://registry.npmjs.org/rollup/-/rollup-2.79.0.tgz",
+      "integrity": "sha512-x4KsrCgwQ7ZJPcFA/SUu6QVcYlO7uRLfLAy0DSA4NS2eG8japdbpM50ToH7z4iObodRYOJ0soneF0iaQRJ6zhA==",
+      "dev": true,
+      "bin": {
+        "rollup": "dist/bin/rollup"
+      },
+      "engines": {
+        "node": ">=10.0.0"
+      },
+      "optionalDependencies": {
+        "fsevents": "~2.3.2"
+      }
+    },
+    "node_modules/rollup/node_modules/fsevents": {
+      "version": "2.3.2",
+      "resolved": "https://registry.npmjs.org/fsevents/-/fsevents-2.3.2.tgz",
+      "integrity": "sha512-xiqMQR4xAeHTuB9uWm+fFRcIOgKBMiOBP+eXiyT7jsgVCq1bkVygt00oASowB7EdtpOHaaPgKt812P9ab+DDKA==",
+      "dev": true,
+      "hasInstallScript": true,
+      "optional": true,
+      "os": [
+        "darwin"
+      ],
+      "engines": {
+        "node": "^8.16.0 || ^10.6.0 || >=11.0.0"
+      }
+    },
     "node_modules/rx": {
       "version": "4.1.0",
       "resolved": "https://registry.npmjs.org/rx/-/rx-4.1.0.tgz",
@@ -8002,6 +8108,12 @@
         "source-map": "^0.6.0"
       }
     },
+    "node_modules/sourcemap-codec": {
+      "version": "1.4.8",
+      "resolved": "https://registry.npmjs.org/sourcemap-codec/-/sourcemap-codec-1.4.8.tgz",
+      "integrity": "sha512-9NykojV5Uih4lgo5So5dtw+f0JgJX30KCNI8gwhz2J9A15wD0Ml6tjHKwf6fTSa6fAdVBdZeNOs9eJ71qCk8vA==",
+      "dev": true
+    },
     "node_modules/spawn-wrap": {
       "version": "2.0.0",
       "resolved": "https://registry.npmjs.org/spawn-wrap/-/spawn-wrap-2.0.0.tgz",
@@ -8388,9 +8500,9 @@
       }
     },
     "node_modules/tesseract.js-core": {
-      "version": "3.0.1",
-      "resolved": "https://registry.npmjs.org/tesseract.js-core/-/tesseract.js-core-3.0.1.tgz",
-      "integrity": "sha512-kcEGcZG4Vl8tmdsPgLacPYoXFRo8IhG9SQxTLbK/6vG+aVTwD89voIJs1KgL3x+RcWDR7sNmd2sMVcpgcLBMMg=="
+      "version": "3.0.2",
+      "resolved": "https://registry.npmjs.org/tesseract.js-core/-/tesseract.js-core-3.0.2.tgz",
+      "integrity": "sha512-2fD76ka9nO/C616R0fq+M9Zu91DA3vEfyozp0jlxaJOBmpfeprtgRP3cqVweZh2darE1kK/DazoxZ65g7WU99Q=="
     },
     "node_modules/test-exclude": {
       "version": "6.0.0",
@@ -8740,6 +8852,11 @@
         "node": ">=4.0.0"
       }
     },
+    "node_modules/wasm-feature-detect": {
+      "version": "1.2.11",
+      "resolved": "https://registry.npmjs.org/wasm-feature-detect/-/wasm-feature-detect-1.2.11.tgz",
+      "integrity": "sha512-HUqwaodrQGaZgz1lZaNioIkog9tkeEJjrM3eq4aUL04whXOVDRc/o2EGb/8kV0QX411iAYWEqq7fMBmJ6dKS6w=="
+    },
     "node_modules/watchpack": {
       "version": "2.4.0",
       "resolved": "https://registry.npmjs.org/watchpack/-/watchpack-2.4.0.tgz",
@@ -10960,6 +11077,54 @@
       "integrity": "sha512-a5Sab1C4/icpTZVzZc5Ghpz88yQtGOyNqYXcZgOssB2uuAr+wF/MvN6bgtW32q7HHrvBki+BsZ0OuNv6EV3K9g==",
       "dev": true
     },
+    "@rollup/plugin-commonjs": {
+      "version": "22.0.2",
+      "resolved": "https://registry.npmjs.org/@rollup/plugin-commonjs/-/plugin-commonjs-22.0.2.tgz",
+      "integrity": "sha512-//NdP6iIwPbMTcazYsiBMbJW7gfmpHom33u1beiIoHDEM0Q9clvtQB1T0efvMqHeKsGohiHo97BCPCkBXdscwg==",
+      "dev": true,
+      "requires": {
+        "@rollup/pluginutils": "^3.1.0",
+        "commondir": "^1.0.1",
+        "estree-walker": "^2.0.1",
+        "glob": "^7.1.6",
+        "is-reference": "^1.2.1",
+        "magic-string": "^0.25.7",
+        "resolve": "^1.17.0"
+      },
+      "dependencies": {
+        "estree-walker": {
+          "version": "2.0.2",
+          "resolved": "https://registry.npmjs.org/estree-walker/-/estree-walker-2.0.2.tgz",
+          "integrity": "sha512-Rfkk/Mp/DL7JVje3u18FxFujQlTNR2q6QfMSMB7AvCBx91NGj/ba3kCfza0f6dVDbw7YlRf/nDrn7pQrCCyQ/w==",
+          "dev": true
+        }
+      }
+    },
+    "@rollup/pluginutils": {
+      "version": "3.1.0",
+      "resolved": "https://registry.npmjs.org/@rollup/pluginutils/-/pluginutils-3.1.0.tgz",
+      "integrity": "sha512-GksZ6pr6TpIjHm8h9lSQ8pi8BE9VeubNT0OMJ3B5uZJ8pz73NPiqOtCog/x2/QzM1ENChPKxMDhiQuRHsqc+lg==",
+      "dev": true,
+      "requires": {
+        "@types/estree": "0.0.39",
+        "estree-walker": "^1.0.1",
+        "picomatch": "^2.2.2"
+      },
+      "dependencies": {
+        "@types/estree": {
+          "version": "0.0.39",
+          "resolved": "https://registry.npmjs.org/@types/estree/-/estree-0.0.39.tgz",
+          "integrity": "sha512-EYNwp3bU+98cpU4lAWYYL7Zz+2gryWH1qbdDTidVd6hkiR6weksdbMadyXKXNPEkQFhXM+hVO9ZygomHXp+AIw==",
+          "dev": true
+        },
+        "estree-walker": {
+          "version": "1.0.1",
+          "resolved": "https://registry.npmjs.org/estree-walker/-/estree-walker-1.0.1.tgz",
+          "integrity": "sha512-1fMXF3YP4pZZVozF8j/ZLfvnR8NSIljt56UhbZ5PeeDmmGHpgpdwQt7ITlGvYaQukCvuBRMLEiKiYC+oeIg4cg==",
+          "dev": true
+        }
+      }
+    },
     "@types/eslint": {
       "version": "8.4.6",
       "resolved": "https://registry.npmjs.org/@types/eslint/-/eslint-8.4.6.tgz",
@@ -13475,6 +13640,15 @@
         "isobject": "^3.0.1"
       }
     },
+    "is-reference": {
+      "version": "1.2.1",
+      "resolved": "https://registry.npmjs.org/is-reference/-/is-reference-1.2.1.tgz",
+      "integrity": "sha512-U82MsXXiFIrjCK4otLT+o2NA2Cd2g5MLoOVXUZjIOhLurrRxpEXzI8O0KZHr3IjLvlAH1kTPYSuqer5T9ZVBKQ==",
+      "dev": true,
+      "requires": {
+        "@types/estree": "*"
+      }
+    },
     "is-regex": {
       "version": "1.1.4",
       "resolved": "https://registry.npmjs.org/is-regex/-/is-regex-1.1.4.tgz",
@@ -14016,6 +14190,15 @@
         }
       }
     },
+    "magic-string": {
+      "version": "0.25.9",
+      "resolved": "https://registry.npmjs.org/magic-string/-/magic-string-0.25.9.tgz",
+      "integrity": "sha512-RmF0AsMzgt25qzqqLc1+MbHmhdx0ojF2Fvs4XnOqz2ZOBXzzkEwc/dJQZCYHAn7v1jbVOjAZfK8msRn4BxO4VQ==",
+      "dev": true,
+      "requires": {
+        "sourcemap-codec": "^1.4.8"
+      }
+    },
     "make-dir": {
       "version": "3.1.0",
       "resolved": "https://registry.npmjs.org/make-dir/-/make-dir-3.1.0.tgz",
@@ -15265,6 +15448,24 @@
         }
       }
     },
+    "rollup": {
+      "version": "2.79.0",
+      "resolved": "https://registry.npmjs.org/rollup/-/rollup-2.79.0.tgz",
+      "integrity": "sha512-x4KsrCgwQ7ZJPcFA/SUu6QVcYlO7uRLfLAy0DSA4NS2eG8japdbpM50ToH7z4iObodRYOJ0soneF0iaQRJ6zhA==",
+      "dev": true,
+      "requires": {
+        "fsevents": "~2.3.2"
+      },
+      "dependencies": {
+        "fsevents": {
+          "version": "2.3.2",
+          "resolved": "https://registry.npmjs.org/fsevents/-/fsevents-2.3.2.tgz",
+          "integrity": "sha512-xiqMQR4xAeHTuB9uWm+fFRcIOgKBMiOBP+eXiyT7jsgVCq1bkVygt00oASowB7EdtpOHaaPgKt812P9ab+DDKA==",
+          "dev": true,
+          "optional": true
+        }
+      }
+    },
     "rx": {
       "version": "4.1.0",
       "resolved": "https://registry.npmjs.org/rx/-/rx-4.1.0.tgz",
@@ -15497,6 +15698,12 @@
         "source-map": "^0.6.0"
       }
     },
+    "sourcemap-codec": {
+      "version": "1.4.8",
+      "resolved": "https://registry.npmjs.org/sourcemap-codec/-/sourcemap-codec-1.4.8.tgz",
+      "integrity": "sha512-9NykojV5Uih4lgo5So5dtw+f0JgJX30KCNI8gwhz2J9A15wD0Ml6tjHKwf6fTSa6fAdVBdZeNOs9eJ71qCk8vA==",
+      "dev": true
+    },
     "spawn-wrap": {
       "version": "2.0.0",
       "resolved": "https://registry.npmjs.org/spawn-wrap/-/spawn-wrap-2.0.0.tgz",
@@ -15782,9 +15989,9 @@
       }
     },
     "tesseract.js-core": {
-      "version": "3.0.1",
-      "resolved": "https://registry.npmjs.org/tesseract.js-core/-/tesseract.js-core-3.0.1.tgz",
-      "integrity": "sha512-kcEGcZG4Vl8tmdsPgLacPYoXFRo8IhG9SQxTLbK/6vG+aVTwD89voIJs1KgL3x+RcWDR7sNmd2sMVcpgcLBMMg=="
+      "version": "3.0.2",
+      "resolved": "https://registry.npmjs.org/tesseract.js-core/-/tesseract.js-core-3.0.2.tgz",
+      "integrity": "sha512-2fD76ka9nO/C616R0fq+M9Zu91DA3vEfyozp0jlxaJOBmpfeprtgRP3cqVweZh2darE1kK/DazoxZ65g7WU99Q=="
     },
     "test-exclude": {
       "version": "6.0.0",
@@ -16052,6 +16259,11 @@
         "rx": "^4.1.0"
       }
     },
+    "wasm-feature-detect": {
+      "version": "1.2.11",
+      "resolved": "https://registry.npmjs.org/wasm-feature-detect/-/wasm-feature-detect-1.2.11.tgz",
+      "integrity": "sha512-HUqwaodrQGaZgz1lZaNioIkog9tkeEJjrM3eq4aUL04whXOVDRc/o2EGb/8kV0QX411iAYWEqq7fMBmJ6dKS6w=="
+    },
     "watchpack": {
       "version": "2.4.0",
       "resolved": "https://registry.npmjs.org/watchpack/-/watchpack-2.4.0.tgz",

package.json

@@ -1,6 +1,6 @@
 {
   "name": "tesseract.js",
-  "version": "2.1.5",
+  "version": "3.0.3",
   "description": "Pure Javascript Multilingual OCR",
   "main": "src/index.js",
   "types": "src/index.d.ts",
@@ -8,7 +8,7 @@
   "jsdelivr": "dist/tesseract.min.js",
   "scripts": {
     "start": "node scripts/server.js",
-    "build": "rimraf dist && webpack --config scripts/webpack.config.prod.js",
+    "build": "rimraf dist && webpack --config scripts/webpack.config.prod.js && rollup -c scripts/rollup.esm.js",
     "profile:tesseract": "webpack-bundle-analyzer dist/tesseract-stats.json",
     "profile:worker": "webpack-bundle-analyzer dist/worker-stats.json",
     "prepublishOnly": "npm run build",
@@ -37,6 +37,7 @@
   "devDependencies": {
     "@babel/core": "^7.18.7",
     "@babel/preset-env": "^7.18.7",
+    "@rollup/plugin-commonjs": "^22.0.2",
     "acorn": "^6.4.0",
     "babel-loader": "^8.2.0",
     "buffer": "^6.0.3",
@@ -51,6 +52,7 @@
     "npm-run-all": "^4.1.5",
     "nyc": "^15.1.0",
     "rimraf": "^2.7.1",
+    "rollup": "^2.79.0",
     "wait-on": "^3.3.0",
     "webpack": "^5.74.0",
     "webpack-bundle-analyzer": "^4.6.0",
@@ -68,7 +70,8 @@
     "opencollective-postinstall": "^2.0.2",
     "regenerator-runtime": "^0.13.3",
     "resolve-url": "^0.2.1",
-    "tesseract.js-core": "^3.0.1",
+    "tesseract.js-core": "^3.0.2",
+    "wasm-feature-detect": "^1.2.11",
     "zlibjs": "^0.3.1"
   },
   "repository": {

scripts/rollup.esm.js

@@ -0,0 +1,13 @@
+import commonjs from "@rollup/plugin-commonjs";
+
+export default [
+  {
+    input: "dist/tesseract.min.js",
+    output: {
+      file: "dist/tesseract.esm.min.js",
+      format: "esm",
+      banner: "/* eslint-disable */",
+    },
+    plugins: [commonjs()],
+  },
+];

src/constants/PSM.js

@@ -15,4 +15,5 @@ module.exports = {
   SINGLE_CHAR: '10',
   SPARSE_TEXT: '11',
   SPARSE_TEXT_OSD: '12',
+  RAW_LINE: '13',
 };

src/index.d.ts

@@ -19,12 +19,18 @@ declare namespace Tesseract {
     readText(path: string, jobId?: string): Promise<ConfigResult>
     removeText(path: string, jobId?: string): Promise<ConfigResult>
     FS(method: string, args: any[], jobId?: string): Promise<ConfigResult>
-    loadLanguage(langs?: string, jobId?: string): Promise<ConfigResult>
-    initialize(langs?: string, oem?: OEM, jobId?: string): Promise<ConfigResult>
+    loadLanguage(langs?: string | Lang[], jobId?: string): Promise<ConfigResult>
+    initialize(langs?: string | Lang[], oem?: OEM, jobId?: string): Promise<ConfigResult>
     setParameters(params: Partial<WorkerParams>, jobId?: string): Promise<ConfigResult>
     recognize(image: ImageLike, options?: Partial<RecognizeOptions>, jobId?: string): Promise<RecognizeResult>
     detect(image: ImageLike, jobId?: string): Promise<DetectResult>
     terminate(jobId?: string): Promise<ConfigResult>
+    getPDF(title?: string, textonly?: boolean, jobId?: string):Promise<GetPDFResult>
+  }
+
+  interface Lang {
+    code: string;
+    data: unknown;
   }
 
   interface WorkerOptions {
@@ -62,6 +68,10 @@ declare namespace Tesseract {
     jobId: string
     data: Page
   }
+  interface GetPDFResult {
+    jobId: string
+    data: number[]
+  }
   interface DetectResult {
     jobId: string
     data: DetectData
@@ -79,13 +89,13 @@ declare namespace Tesseract {
     width: number
     height: number
   }
-  const enum OEM {
+  enum OEM {
     TESSERACT_ONLY,
     LSTM_ONLY,
     TESSERACT_LSTM_COMBINED,
     DEFAULT,
   }
-  const enum PSM {
+  enum PSM {
     OSD_ONLY = '0',
     AUTO_OSD = '1',
     AUTO_ONLY = '2',
@@ -99,6 +109,7 @@ declare namespace Tesseract {
     SINGLE_CHAR = '10',
     SPARSE_TEXT = '11',
     SPARSE_TEXT_OSD = '12',
+    RAW_LINE = '13'
   }
   type ImageLike = string | HTMLImageElement | HTMLCanvasElement | HTMLVideoElement
     | CanvasRenderingContext2D | File | Blob | ImageData | Buffer;

src/utils/resolvePaths.js

@@ -4,7 +4,7 @@ const resolveURL = isBrowser ? require('resolve-url') : s => s; // eslint-disabl
 module.exports = (options) => {
   const opts = { ...options };
   ['corePath', 'workerPath', 'langPath'].forEach((key) => {
-    if (typeof options[key] !== 'undefined') {
+    if (options[key]) {
       opts[key] = resolveURL(opts[key]);
     }
   });

src/worker-script/browser/getCore.js

@@ -1,15 +1,26 @@
-module.exports = (corePath, res) => {
+const { simd } = require('wasm-feature-detect');
+const { dependencies } = require('../../../package.json');
+
+module.exports = async (corePath, res) => {
   if (typeof global.TesseractCore === 'undefined') {
     res.progress({ status: 'loading tesseract core', progress: 0 });
-    global.importScripts(corePath);
-    /*
-     * Depending on whether the browser supports WebAssembly,
-     * the version of the TesseractCore will be different.
-     */
+
+    // If the user specifies a core path, we use that
+    // Otherwise, we detect the correct core based on SIMD support
+    let corePathImport = corePath;
+    if (!corePathImport) {
+      const simdSupport = await simd();
+      if (simdSupport) {
+        corePathImport = `https://unpkg.com/tesseract.js-core@v${dependencies['tesseract.js-core'].substring(1)}/tesseract-core-simd.wasm.js`;
+      } else {
+        corePathImport = `https://unpkg.com/tesseract.js-core@v${dependencies['tesseract.js-core'].substring(1)}/tesseract-core.wasm.js`;
+      }
+    }
+
+    global.importScripts(corePathImport);
+
     if (typeof global.TesseractCoreWASM !== 'undefined' && typeof WebAssembly === 'object') {
       global.TesseractCore = global.TesseractCoreWASM;
-    } else if (typeof global.TesseractCoreASM !== 'undefined') {
-      global.TesseractCore = global.TesseractCoreASM;
     } else {
       throw Error('Failed to load TesseractCore');
     }

src/worker-script/index.js

@@ -28,10 +28,10 @@ let latestJob;
 let adapter = {};
 let params = defaultParams;
 
-const load = ({ workerId, jobId, payload: { options: { corePath, logging } } }, res) => {
+const load = async ({ workerId, jobId, payload: { options: { corePath, logging } } }, res) => {
   setLogging(logging);
   if (!TessModule) {
-    const Core = adapter.getCore(corePath, res);
+    const Core = await adapter.getCore(corePath, res);
 
     res.progress({ workerId, status: 'initializing tesseract', progress: 0 });
 
@@ -144,16 +144,8 @@ res) => {
     res.progress({ workerId, status: 'loaded language traineddata', progress: 1 });
     res.resolve(langs);
   } catch (err) {
-    if (isWebWorker && err instanceof DOMException) {
-      /*
-       * For some reason google chrome throw DOMException in loadLang,
-       * while other browser is OK, for now we ignore this exception
-       * and hopefully to find the root cause one day.
-       */
-    } else {
     res.reject(err.toString());
   }
-  }
 };
 
 const setParameters = ({ payload: { params: _params } }, res) => {
@@ -185,7 +177,10 @@ const initialize = ({
       api.End();
     }
     api = new TessModule.TessBaseAPI();
-    api.Init(null, langs, oem);
+    const status = api.Init(null, langs, oem);
+    if (status === -1) {
+      res.reject('initialization failed');
+    }
     params = defaultParams;
     setParameters({ payload: { params } });
     res.progress({

src/worker-script/node/getCore.js

@@ -1,12 +1,19 @@
+const { simd } = require('wasm-feature-detect');
+
 let TesseractCore = null;
 /*
  * getCore is a sync function to load and return
  * TesseractCore.
  */
-module.exports = (_, res) => {
+module.exports = async (_, res) => {
   if (TesseractCore === null) {
+    const simdSupport = await simd();
     res.progress({ status: 'loading tesseract core', progress: 0 });
-    TesseractCore = require('tesseract.js-core');
+    if (simdSupport) {
+      TesseractCore = require('tesseract.js-core/tesseract-core-simd');
+    } else {
+      TesseractCore = require('tesseract.js-core/tesseract-core');
+    }
     res.progress({ status: 'loaded tesseract core', progress: 1 });
   }
   return TesseractCore;

src/worker-script/utils/setImage.js

@@ -20,9 +20,9 @@ module.exports = (TessModule, api, image) => {
   const exif = buf.slice(0, 500).toString().match(/\x01\x12\x00\x03\x00\x00\x00\x01\x00(.)/)?.[1]?.charCodeAt(0) || 1;
 
   /*
-   * Although leptonica should support reading bmp, there is a bug of "compressed BMP files".
-   * As there is no solution, we need to use bmp-js for now.
-   * @see https://groups.google.com/forum/#!topic/tesseract-ocr/4mPD9zTxdxE
+   * Leptonica supports uncompressed but not compressed bmp files
+   * @see https://github.com/DanBloomberg/leptonica/issues/607#issuecomment-1068802516
+   * We therefore use bmp-js to process all bmp files
    */
   if (type && type.mime === 'image/bmp') {
     const bmpBuf = bmp.decode(buf);

src/worker/browser/defaultOptions.js

@@ -1,5 +1,5 @@
 const resolveURL = require('resolve-url');
-const { version, dependencies } = require('../../../package.json');
+const { version } = require('../../../package.json');
 const defaultOptions = require('../../constants/defaultOptions');
 
 /*
@@ -14,5 +14,5 @@ module.exports = {
    * If browser doesn't support WebAssembly,
    * load ASM version instead
    */
-  corePath: `https://unpkg.com/tesseract.js-core@v${dependencies['tesseract.js-core'].substring(1)}/tesseract-core.${typeof WebAssembly === 'object' ? 'wasm' : 'asm'}.js`,
+  corePath: null,
 };