Merge pull request #244 from jeromewu/master

Use jsDelivr as cdn

README.md

@@ -251,19 +251,19 @@ Because of this we recommend loading `tesseract.js` from a CDN. But if you reall
 ```javascript
 window.Tesseract = Tesseract.create({
     workerPath: '/path/to/worker.js',
-    langPath: 'https://rawcdn.githack.com/naptha/tessdata/gh-pages/3.02/',
+    langPath: 'https://cdn.jsdelivr.net/gh/naptha/tessdata@gh-pages/3.02/',
-    corePath: 'https://rawcdn.githack.com/naptha/tesseract.js-core/0.1.0/index.js',
+    corePath: 'https://cdn.jsdelivr.net/gh/naptha/tesseract.js-core@0.1.0/index.js',
 })
 ```
 
 ### corePath
-A string specifying the location of the [tesseract.js-core library](https://github.com/naptha/tesseract.js-core), with default value 'https://rawcdn.githack.com/naptha/tesseract.js-core/master/index.js'. Set this string before calling `Tesseract.recognize` and `Tesseract.detect` if you want Tesseract.js to use a different file.
+A string specifying the location of the [tesseract.js-core library](https://github.com/naptha/tesseract.js-core), with default value 'https://cdn.jsdelivr.net/gh/naptha/tesseract.js-core@0.1.0/index.js'. Set this string before calling `Tesseract.recognize` and `Tesseract.detect` if you want Tesseract.js to use a different file.
 
 ### workerPath
 A string specifying the location of the [worker.js](./dist/worker.js) file. Set this string before calling `Tesseract.recognize` and `Tesseract.detect` if you want Tesseract.js to use a different file.
 
 ### langPath
-A string specifying the location of the tesseract language files, with default value 'https://rawcdn.githack.com/naptha/tessdata/gh-pages/3.02/'. Language file URLs are calculated according to the formula `langPath + langCode + '.traineddata.gz'`. Set this string before calling `Tesseract.recognize` and `Tesseract.detect` if you want Tesseract.js to use different language files.
+A string specifying the location of the tesseract language files, with default value 'https://cdn.jsdelivr.net/gh/naptha/tessdata@gh-pages/3.02/'. Language file URLs are calculated according to the formula `langPath + langCode + '.traineddata.gz'`. Set this string before calling `Tesseract.recognize` and `Tesseract.detect` if you want Tesseract.js to use different language files.
 
 
 ## Contributing

dist/tesseract.js

@@ -187,7 +187,7 @@ process.umask = function() { return 0; };
 },{}],2:[function(require,module,exports){
 module.exports={
   "name": "tesseract.js",
-  "version": "1.0.11",
+  "version": "1.0.12",
   "description": "Pure Javascript Multilingual OCR",
   "main": "src/index.js",
   "scripts": {
@@ -236,9 +236,9 @@ module.exports={
 'use strict';
 
 var defaultOptions = {
-    // workerPath: 'https://rawcdn.githack.com/naptha/tesseract.js/0.2.0/dist/worker.js',
+    // workerPath: 'https://cdn.jsdelivr.net/gh/naptha/tesseract.js@0.2.0/dist/worker.js',
-    corePath: 'https://rawcdn.githack.com/naptha/tesseract.js-core/0.1.0/index.js',
+    corePath: 'https://cdn.jsdelivr.net/gh/naptha/tesseract.js-core@0.1.0/index.js',
-    langPath: 'https://rawcdn.githack.com/naptha/tessdata/gh-pages/3.02/'
+    langPath: 'https://cdn.jsdelivr.net/gh/naptha/tessdata@gh-pages/3.02/'
 };
 
 if (process.env.NODE_ENV === "development") {
@@ -246,7 +246,7 @@ if (process.env.NODE_ENV === "development") {
     defaultOptions.workerPath = location.protocol + '//' + location.host + '/dist/worker.dev.js?nocache=' + Math.random().toString(36).slice(3);
 } else {
     var version = require('../../package.json').version;
-    defaultOptions.workerPath = 'https://rawcdn.githack.com/naptha/tesseract.js/' + version + '/dist/worker.js';
+    defaultOptions.workerPath = 'https://cdn.jsdelivr.net/gh/naptha/tesseract.js@' + version + '/dist/worker.js';
 }
 
 exports.defaultOptions = defaultOptions;

package-lock.json

@@ -1,6 +1,6 @@
 {
   "name": "tesseract.js",
-  "version": "1.0.11",
+  "version": "1.0.12",
   "lockfileVersion": 1,
   "requires": true,
   "dependencies": {

package.json

@@ -1,6 +1,6 @@
 {
   "name": "tesseract.js",
-  "version": "1.0.11",
+  "version": "1.0.12",
   "description": "Pure Javascript Multilingual OCR",
   "main": "src/index.js",
   "scripts": {

src/browser/index.js

@@ -1,7 +1,7 @@
 var defaultOptions = {
-    // workerPath: 'https://rawcdn.githack.com/naptha/tesseract.js/0.2.0/dist/worker.js',
+    // workerPath: 'https://cdn.jsdelivr.net/gh/naptha/tesseract.js@0.2.0/dist/worker.js',
-    corePath: 'https://rawcdn.githack.com/naptha/tesseract.js-core/0.1.0/index.js',    
+    corePath: 'https://cdn.jsdelivr.net/gh/naptha/tesseract.js-core@0.1.0/index.js',    
-    langPath: 'https://rawcdn.githack.com/naptha/tessdata/gh-pages/3.02/',
+    langPath: 'https://cdn.jsdelivr.net/gh/naptha/tessdata@gh-pages/3.02/',
 }
 
 if (process.env.NODE_ENV === "development") {
@@ -9,7 +9,7 @@ if (process.env.NODE_ENV === "development") {
     defaultOptions.workerPath = location.protocol + '//' + location.host + '/dist/worker.dev.js?nocache=' + Math.random().toString(36).slice(3)
 }else{
     var version = require('../../package.json').version;
-    defaultOptions.workerPath = 'https://rawcdn.githack.com/naptha/tesseract.js/' + version + '/dist/worker.js'
+    defaultOptions.workerPath = 'https://cdn.jsdelivr.net/gh/naptha/tesseract.js@' + version + '/dist/worker.js'
 }
 
 exports.defaultOptions = defaultOptions;

src/node/index.js

@@ -5,7 +5,7 @@ const fetch = require('isomorphic-fetch'),
 
 exports.defaultOptions = {
     workerPath: require('path').join(__dirname, 'worker.js'),
-    langPath: 'http://rawcdn.githack.com/naptha/tessdata/gh-pages/3.02/',
+    langPath: 'https://cdn.jsdelivr.net/gh/naptha/tessdata@gh-pages/3.02/',
 }
 
 exports.spawnWorker = function spawnWorker(instance, workerOptions){

src/node/lang.js

@@ -1,4 +1,5 @@
-const http = require("http"),
+const https = require("https"),
+      http = require("http"),
       zlib = require("zlib"),
       fs   = require("fs"),
       path = require("path"),
@@ -16,10 +17,12 @@ function getLanguageData(req, res, cb){
         lang + '.traineddata' : 
         path.join(req.workerOptions.langPath, lang + '.traineddata');
 
+    var fetchProtocol = req.workerOptions.langPath.startsWith('http://') ? http : https;
+
     fs.readFile(localPath, function (err, data) {
         if(!err) return cb(new Uint8Array(data));
 
-        http.get(req.workerOptions.langPath + langfile, stream => {
+        fetchProtocol.get(req.workerOptions.langPath + langfile, stream => {
             var received_bytes = 0;
             stream.on('data', function(chunk) {
                 received_bytes += chunk.length;