feat: python builder

README.md

@@ -1,137 +1,15 @@
-# mrz
+# mrz, but for python
 
-Parse MRZ (Machine Readable Zone) from identity documents.
+This is a port of https://github.com/cheminfo/mrz but for python.
 
-<h3 align="center">
+### Installation
 
-  <a href="https://www.zakodium.com">
-    <img src="https://www.zakodium.com/brand/zakodium-logo-white.svg" width="50" alt="Zakodium logo" />
-  </a>
-
-  <p>
-    Maintained by <a href="https://www.zakodium.com">Zakodium</a>
-  </p>
-
-[![NPM version][npm-image]][npm-url]
-[![build status][ci-image]][ci-url]
-[![npm download][download-image]][download-url]
-
-</h3>
-
-## Installation
-
-`$ npm install mrz`
-
-## Example
-
-```js
-const parse = require('mrz').parse;
-
-const mrz = [
-  'I<UTOD23145890<1233<<<<<<<<<<<',
-  '7408122F1204159UTO<<<<<<<<<<<6',
-  'ERIKSSON<<ANNA<MARIA<<<<<<<<<<',
-];
-
-var result = parse(mrz);
-console.log(result);
+```bash
+npm run build-py
 ```
 
-## API
-
-### `parse(mrz, [options])`
-
-Parses the provided MRZ. The argument can be an array of lines or a single string
-including line breaks. This function throws an error if the input is in an
-unsupported format. It will never throw an error when there are invalid fields
-in the MRZ. Instead, the `result.valid` value will be `false` and
-details about the invalid fields can be found in `result.details`.
-
-#### Options
-
-##### `options.autocorrect`
-
-If set to `true`, some ambiguous characters will be automatically corrected by the parser if the field is supposed to
-only contain numeric or alphabetic characters.
-For example, in a date field, the letter "O" will be converted to the number "0".
-
-Information about autocorrected characters will be added to the result details.
-
-Default: `false`.
-
-#### Shape of the parse result
-
-##### result.format
-
-String identifying the format of the parsed MRZ. Supported formats are:
-
-- TD1 (identity card with three MRZ lines)
-- TD2 (identity card with two MRZ lines)
-- TD3 (passport)
-- SWISS_DRIVING_LICENSE
-- FRENCH_NATIONAL_ID
-
-##### result.valid
-
-`true` if all fields are valid. `false` otherwise.
+### Usage
 
-##### result.fields
-
-Object mapping field names to their respective value. The value is set to `null`
-if it is invalid. The value may be different from the raw value. For example,
-`result.fields.sex` will be "male" when the raw value was "M".
-
-##### result.documentNumber
-
-The document number, as can be found in the visual elements of the document, outside the MRZ. For some documents, it may
-be composed of multiple parsed fields. It may also not include the MRZ field named `documentNumber`. If any of the used
-fields is invalid, this field will be set to `null`.
-
-##### result.details
-
-Array of objects describing all parsed fields. Its structure is:
-
-- label {string} - Full english term for the field.
-- field {string|null} - Name of the field in `result.fields`. Null for some fields such as separators that don't contain a value.
-- value {string} - Value of the field (if it's valid) or `null`.
-- valid {boolean} - Whether the field is valid.
-- ranges {Array} - Array of ranges that are necessary to compute this field.
-  Ranges are objects with `line`, `start`, `end` and `raw`.
-- line {number} - Index of the line where the field's value is located.
-- start {number} - Index of the start of the field's value in `line`.
-- end {number} - Index of the end of the field's value in `line`.
-- error {undefined|string} - Contains a message describing the error if the field is invalid.
-- autocorrect {array} - Contains indices of characters that were autocorrected and their original value.
-
-### `formats`
-
-Static mapping of supported formats.
-
-### `states`
-
-Static mapping of state code to state name.
-
-## Specifications
-
-### TD1, TD2 and TD3
-
-https://www.icao.int/publications/pages/publication.aspx?docnum=9303
-
-### Swiss driving license
-
-https://www.astra.admin.ch/dam/astra/fr/dokumente/dokumente-strassenverkehr/kreisschreiben/ch-fak.pdf.download.pdf/Le%20permis%20de%20conduire%20suisse%20format%20carte%20de%20cr%C3%A9dit%20(PCC).pdf
-
-### French national id
-
-https://fr.wikipedia.org/wiki/Carte_nationale_d%27identit%C3%A9_en_France#Codage_bande_%C3%A0_lecture_optique
-
-## License
-
-[MIT](./LICENSE)
-
-[npm-image]: https://img.shields.io/npm/v/mrz.svg
-[npm-url]: https://npmjs.org/package/mrz
-[ci-image]: https://github.com/cheminfo/mrz/workflows/Node.js%20CI/badge.svg?branch=main
-[ci-url]: https://github.com/cheminfo/mrz/actions?query=workflow%3A%22Node.js+CI%22
-[download-image]: https://img.shields.io/npm/dm/mrz.svg
-[download-url]: https://npmjs.org/package/mrz
+```python
+test.var.get('window').callprop('parse', ['I<UTOD23145890<1233<<<<<<<<<<<','7408122F1204159UTO<<<<<<<<<<<6','ERIKSSON<<ANNA<MARIA<<<<<<<<<<'])
+```

babel.config.js

@@ -1,4 +1,17 @@
 module.exports = {
-  presets: ['@babel/preset-typescript'],
-  plugins: ['@babel/plugin-transform-modules-commonjs'],
+  presets: [
+    '@babel/preset-typescript',
+    [
+      '@babel/preset-env',
+      {
+        targets: {
+          ie: '11',
+        },
+      },
+    ],
+  ],
+  plugins: [
+    '@babel/plugin-transform-modules-commonjs',
+    '@babel/plugin-transform-template-literals',
+  ],
 };

builder/buildPython.py

@@ -0,0 +1,2 @@
+import js2py
+js2py.translate_file('dist/mrz.js', 'dist/mrz.py')

package.json

@@ -11,6 +11,7 @@
     "lib-esm"
   ],
   "scripts": {
+    "build-py": "npm run prepack && npm run build && pip3 install js2py && python3 builder/buildPython.py",
     "build": "npm run buildStates && cheminfo-build",
     "buildStates": "node builder/createStatesJs.js",
     "check-types": "tsc --noEmit",
@@ -47,7 +48,11 @@
   },
   "homepage": "https://github.com/cheminfo/mrz#readme",
   "devDependencies": {
+    "@babel/cli": "^7.24.8",
+    "@babel/core": "^7.25.2",
     "@babel/plugin-transform-modules-commonjs": "^7.24.6",
+    "@babel/plugin-transform-template-literals": "^7.24.7",
+    "@babel/preset-env": "^7.25.2",
     "@babel/preset-typescript": "^7.24.6",
     "@types/jest": "^29.5.12",
     "cheminfo-build": "^1.2.0",

src/index.ts

@@ -2,5 +2,6 @@
 import states from './generated/states';
 import parse from './parse/parse';
 
-export { formats, states, parse };
-export * from './types';
+(window as any).formats = formats;
+(window as any).states = states;
+(window as any).parse = parse;

src/parsers/frenchDrivingLicence/parseDocumentCode.ts

@@ -9,7 +9,8 @@ export default function parseDocumentCode(source: string) {
     throw new Error(`invalid document code: ${source}. must be 2 char length`);
   }
 
-  const [first, second] = source;
+  const first = source[0];
+  const second = source[1];
 
   validateFirstChar(first, source);
   validateSecondChar(second, source);
@@ -19,12 +20,8 @@ export default function parseDocumentCode(source: string) {
 
 function validateFirstChar(char: string, source: string) {
   if (!knownPrefixCode.includes(char)) {
-    const formatter = new Intl.ListFormat('en-US', {
-      style: 'short',
-      type: 'disjunction',
-    });
     throw new Error(
-      `invalid document code: ${source}. First character must be ${formatter.format(knownPrefixCode)}`,
+      `invalid document code: ${source}. First character must be ???`,
     );
   }
 }

tsconfig.json

@@ -6,7 +6,7 @@
     "outDir": "lib",
     "sourceMap": true,
     "strict": true,
-    "target": "es2021",
+    "target": "ES5",
     "useUnknownInCatchVariables": false,
     "skipLibCheck": true
   },

webpack.config.js

@@ -0,0 +1,25 @@
+const path = require('path');
+
+module.exports = {
+  entry: './src/index.js', // Your entry file
+  output: {
+    filename: 'bundle.js',
+    path: path.resolve(__dirname, 'dist'),
+    globalObject: 'this', // Ensure global object reference
+    library: {
+      type: 'window', // Expose the library globally
+    },
+  },
+  module: {
+    rules: [
+      {
+        test: /\.js$/,
+        exclude: /node_modules/,
+        use: {
+          loader: 'babel-loader',
+        },
+      },
+    ],
+  },
+  mode: 'production',
+};