tesseract.js - npm Package Compare versions

Comparing version 2.0.0-alpha.6 to 2.0.0-alpha.7

docs/examples.md

@@ -112,4 +112,6 @@ # Tesseract.js Examples
 
-### with pdf output (^2.0.0-alpha.5)
+### with pdf output (^2.0.0-alpha.7)
 
+In this example, pdf file will be downloaded in browser and write to file system in Node.js
+
 ```javascript
@@ -136,1 +138,27 @@ import Tesseract from 'tesseract.js';
 ```
+
+If you want to handle pdf file by yourself
+
+```javascript
+import Tesseract from 'tesseract.js';
+
+const { TesseractWorker } = Tesseract;
+const worker = new TesseractWorker();
+
+worker
+  .recognize(
+    'http://jeroen.github.io/images/testocr.png',
+    'eng',
+    {
+      'tessedit_create_pdf': '1',
+      'pdf_auto_download': false, // disable auto download
+      'pdf_bin': true,            // add pdf file bin array in result
+    }
+  )
+  .progress((p) => {
+    console.log('progress', p);
+  })
+  .then((result) => {
+    console.log(result.files.pdf); // You can access pdf binary array here.
+  });
+```

docs/local-installation.md

@@ -13,3 +13,3 @@ ## Local Installation
 const worker = Tesseract.TesseractWorker({
-  workerPath: 'https://unpkg.com/tesseract.js@v2.0.0-alpha.6/dist/worker.min.js',
+  workerPath: 'https://unpkg.com/tesseract.js@v2.0.0-alpha.7/dist/worker.min.js',
   langPath: 'https://tessdata.projectnaptha.com/4.0.0',
@@ -16,0 +16,0 @@ corePath: 'https://unpkg.com/tesseract.js-core@v2.0.0-beta.10/tesseract-core.wasm.js',

package.json

 {
   "name": "tesseract.js",
-  "version": "2.0.0-alpha.6",
+  "version": "2.0.0-alpha.7",
   "description": "Pure Javascript Multilingual OCR",
@@ -5,0 +5,0 @@ "main": "src/index.js",

README.md

@@ -42,3 +42,3 @@ # [Tesseract.js](http://tesseract.projectnaptha.com/)
 ```html
-<script src='https://unpkg.com/tesseract.js@v2.0.0-alpha.6/dist/tesseract.min.js'></script>
+<script src='https://unpkg.com/tesseract.js@v2.0.0-alpha.7/dist/tesseract.min.js'></script>
 ```
@@ -45,0 +45,0 @@

src/browser/index.js

@@ -133,3 +133,4 @@ /**
     } else if (data.jobId.startsWith('Download')) {
-      const { path, blob } = data;
+      const { path, data: d, type } = data;
+      const blob = new Blob([d], { type });
       downloadFile(path, blob);
@@ -136,0 +137,0 @@ }

src/browser/worker.js

@@ -47,9 +47,9 @@ /**
   writeFile: (path, data, type) => {
-    const blob = new Blob([data], { type });
     self.postMessage({
       jobId: 'Download',
       path,
-      blob,
+      data,
+      type,
     });
   },
 });

src/common/options.js

@@ -23,3 +23,5 @@ const { OEM, PSM } = require('./types');
     pdf_title: 'Tesseract.js OCR Result',
+    pdf_auto_download: true,
+    pdf_bin: false,
   },
 };

src/common/workerUtils.js

@@ -88,2 +88,3 @@ /**
 const handleOutput = (customParams) => {
+  let files = {};
   const {
@@ -94,2 +95,4 @@ tessedit_create_pdf,
     pdf_title,
+    pdf_auto_download,
+    pdf_bin,
   } = {
@@ -105,5 +108,16 @@ ...defaultParams,
     pdfRenderer.EndDocument();
-    adapter.writeFile(`${pdf_name}.pdf`, TessModule.FS.readFile(`/${pdf_name}.pdf`), 'application/pdf');
     TessModule._free(pdfRenderer);
+
+    const data = TessModule.FS.readFile(`/${pdf_name}.pdf`);
+
+    if (pdf_bin) {
+      files = { pdf: data, ...files };
+    }
+    
+    if (pdf_auto_download) {
+      adapter.writeFile(`${pdf_name}.pdf`, data, 'application/pdf');
+    }
   }
+
+  return files;
 }
@@ -193,7 +207,7 @@
           api.Recognize(null);
-          handleOutput(params);
+          const files = handleOutput(params);
           const result = dump(TessModule, api);
           api.End();
           TessModule._free(ptr);
-          res.resolve(result);
+          res.resolve({ files, ...result });
         })
@@ -200,0 +214,0 @@ ))

src/node/worker.js

@@ -39,6 +39,6 @@ /**
     const fs = require('fs');
-    fs.writeFile(path, data, () => {
-      console.log('File Write Succeeded!');
+    fs.writeFile(path, data, (err) => {
+      if (err) throw err;
     });
   },
 });

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

Improved metrics

Total package byte prevSize

1366877

increased by14.35%

Number of package files

53

increased by20.45%

Lines of code

2080

increased by0.68%

No dependency changes