On Android, the [MLKit Vision Barcode Scanning](https://developers.google.com/ml-kit/vision/barcode-scanning) API will be used, and the model (2.2MB) needs to be included into your app first.
The Code Scanner will call your [`onCodeScanned`](/docs/api/interfaces/CodeScanner#oncodescanned) callback with all detected codes ([`Code`](/docs/api/interfaces/Code)), including their decoded string value, and their coordinates on the screen relative to the Preview.
UPC-A is a special case to handle if you need to cater for it. Android's SDK officially supports UPC-A but iOS does not, instead they handle the code as EAN-13. Since EAN-13 is a superset of UPC-A, with an extra 0 digit at the front.
This means, the `upc-a` types are reported under the `ean-13` umbrella type on iOS:
```jsx
const codeScanner = useCodeScanner({
codeTypes: ['upc-a'], // <-- ✅ We configure for 'upc-a' types
onCodeScanned: (codes) => {
for (const code of codes) {
console.log(code.type); // <-- ❌ On iOS, we receive 'ean-13'
}
}
})
```
You will need to keep this in mind and do the conversion from EAN-13 to UPC-A yourself. This can be done by removing the front `0` digit from the code to get a UPC-A code.