tesseract.js - npm Package Compare versions

Comparing version 2.0.0-alpha.7 to 2.0.0-alpha.8

docs/examples.md

@@ -5,2 +5,7 @@ # Tesseract.js Examples
 
+Example repositories:
+
+- Offline version: https://github.com/jeromewu/tesseract.js-offline
+- With Vue (similar with React/Angular): https://github.com/jeromewu/tesseract-vue-app
+
 ### basic
@@ -7,0 +12,0 @@

docs/faq.md

@@ -9,1 +9,7 @@ FAQ
 Tesseract.js will first check if \*.traineddata already exists. (browser: [IndexedDB](https://developer.mozilla.org/en-US/docs/Web/API/IndexedDB_API), Node.js: fs, in the folder you execute the command) If the \*.traineddata doesn't exist, it will fetch \*.traineddata.gz from [tessdata](https://github.com/naptha/tessdata), ungzip and store in IndexedDB or fs, you can delete it manually and it will download again for you.
+
+## How can I train my own \*.traineddata?
+
+For tesseract.js v2, check [TrainingTesseract 4.00](https://github.com/tesseract-ocr/tesseract/wiki/TrainingTesseract-4.00)
+
+For tesseract.js v1, check [Training Tesseract 3.03–3.05](https://github.com/tesseract-ocr/tesseract/wiki/Training-Tesseract-3.03%E2%80%933.05)

package.json

 {
   "name": "tesseract.js",
-  "version": "2.0.0-alpha.7",
+  "version": "2.0.0-alpha.8",
   "description": "Pure Javascript Multilingual OCR",
@@ -15,4 +15,4 @@ "main": "src/index.js",
     "test:browser-tpl": "mocha-headless-chrome -a incognito -a no-sandbox -a disable-setuid-sandbox -t 300000",
-    "test:browser:detect": "npm run test:browser-tpl -- -f ./tests/browser/detect.test.html",
-    "test:browser:recognize": "npm run test:browser-tpl -- -f ./tests/browser/recognize.test.html",
+    "test:browser:detect": "npm run test:browser-tpl -- -f ./tests/detect.test.html",
+    "test:browser:recognize": "npm run test:browser-tpl -- -f ./tests/recognize.test.html",
     "lint": "eslint src"
@@ -29,2 +29,6 @@ },
   "devDependencies": {
+    "@babel/core": "^7.4.5",
+    "@babel/preset-env": "^7.4.5",
+    "acorn": "^6.1.1",
+    "babel-loader": "^8.0.6",
     "cors": "^2.8.5",
@@ -48,2 +52,3 @@ "eslint": "^5.9.0",
   "dependencies": {
+    "axios": "^0.18.0",
     "check-types": "^7.4.0",
@@ -54,3 +59,3 @@ "is-url": "1.2.2",
     "tesseract.js-core": "^2.0.0-beta.10",
-    "tesseract.js-utils": "^1.0.0-beta.5"
+    "tesseract.js-utils": "^1.0.0-beta.6"
   },
@@ -57,0 +62,0 @@ "repository": {

scripts/webpack.config.dev.js

 const path = require('path');
 const webpack = require('webpack');
+const common = require('./webpack.config.common');
 
@@ -7,2 +8,3 @@ const genConfig = ({
 }) => ({
+  ...common,
   mode: 'development',
@@ -9,0 +11,0 @@ entry,

scripts/webpack.config.prod.js

 const path = require('path');
+const common = require('./webpack.config.common');
 
@@ -6,2 +7,3 @@ const genConfig = ({
 }) => ({
+  ...common,
   mode: 'production',
@@ -8,0 +10,0 @@ devtool: 'source-map',

src/browser/index.js

@@ -12,2 +12,3 @@ /**
 const resolveURL = require('resolve-url');
+const axios = require('axios');
 const { defaultOptions } = require('../common/options');
@@ -43,3 +44,3 @@ const { version } = require('../../package.json');
  *   video HTMLElement: extract image source from poster attribute
- *   canvas HTMLElement: extract image data by converting to Blob 
+ *   canvas HTMLElement: extract image data by converting to Blob
  *   File instance: data from <input type="file" />
@@ -50,4 +51,6 @@ * @returns {array} binary image in array format
   if (check.string(image)) {
-    return fetch(resolveURL(image))
-      .then(resp => resp.arrayBuffer());
+    return axios.get(resolveURL(image), {
+      responseType: 'arraybuffer',
+    })
+      .then(resp => resp.data);
   }
@@ -94,3 +97,3 @@ if (check.instance(image, HTMLElement)) {
   }
-}
+};
 
@@ -97,0 +100,0 @@ /*

src/browser/worker.js

@@ -36,3 +36,3 @@ /**
         global.TesseractCore = global.TesseractCoreWASM;
-      } else if (check.not.undefined(global.TesseractCoreASM)){
+      } else if (check.not.undefined(global.TesseractCoreASM)) {
         global.TesseractCore = global.TesseractCoreASM;
@@ -48,3 +48,3 @@ } else {
   writeFile: (path, data, type) => {
-    self.postMessage({
+    postMessage({
       jobId: 'Download',
@@ -51,0 +51,0 @@ path,

src/common/workerUtils.js

@@ -11,3 +11,2 @@ /**
 const { readImage, loadLang } = require('tesseract.js-utils');
-const check = require('check-types');
 const pdfTTF = require('./pdf-ttf');
@@ -63,3 +62,3 @@ const dump = require('./dump');
  * @access private
- * @param {string} lang - lang string for Init() 
+ * @param {string} lang - lang string for Init()
  * @param {object} customParams - an object of params
@@ -115,3 +114,3 @@ */
     }
-    
+
     if (pdf_auto_download) {
@@ -123,3 +122,3 @@ adapter.writeFile(`${pdf_name}.pdf`, data, 'application/pdf');
   return files;
-}
+};
 
@@ -198,2 +197,13 @@ /**
       loadLanguage({ lang, options }, res)
+        .catch((e) => {
+          if (e 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 {
+            throw e;
+          }
+        })
         .then(() => {
@@ -200,0 +210,0 @@ const progressUpdate = (progress) => {

src/node/index.js

@@ -12,3 +12,3 @@ /**
 const fs = require('fs');
-const fetch = require('node-fetch');
+const axios = require('axios');
 const isURL = require('is-url');
@@ -33,4 +33,6 @@ const { fork } = require('child_process');
   if (isURL(image)) {
-    return fetch(image)
-      .then(resp => resp.arrayBuffer());
+    return axios.get(image, {
+      responseType: 'arraybuffer',
+    })
+      .then(resp => resp.data);
   }
@@ -37,0 +39,0 @@ return readFile(image);

New alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Fixed alerts

License Policy Violation

License

This package is not allowed per your license policy. Review the package's license to ensure compliance.

Found 1 instance in 1 package

Network access

Supply chain risk

This module accesses the network.

Found 1 instance in 1 package

Improved metrics

Lines of code

2284

increased by9.81%

Number of medium supply chain risk alerts

1

decreased by-83.33%

Worsened metrics

Total package byte prevSize

1356882

decreased by-0.73%

Dependency count

7

increased by16.67%

Dev dependency count

20

increased by25%

Number of package files

44

decreased by-16.98%

Dependency changes

• + Addedaxios@^0.18.0

• Updatedtesseract.js-utils@^1.0.0-beta.6