Skip to content

Commit

Permalink
chore: update changelog
Browse files Browse the repository at this point in the history
  • Loading branch information
@Sec-ant
Sec-ant committed Jan 5, 2025
1 parent e2461ad commit 75d80cf
Showing 1 changed file with 43 additions and 50 deletions.
93 changes: 43 additions & 50 deletions CHANGELOG.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -2,86 +2,79 @@

## 2.0.0

### Major Changes
This release introduces a major refactoring of the underlying Embind APIs and read / write functions. Key changes include a new set of default reader options, enhanced writer capabilities backed by [`zint`](https://sourceforge.net/projects/zint/), and updated APIs for reading and writing barcodes. These changes break backward compatibility, so we are upgrading to the next major version.

- 1a77296: # V2: Breaking Release - Next Major Version
### Breaking Changes

This release introduces a major refactoring of the underlying Embind APIs and read / write functions. Key changes include transitioning away from [Embind Enums](https://emscripten.org/docs/porting/connecting_cpp_and_javascript/embind.html#enums) toward numeric encoding and decoding, a new set of default reader options, enhanced writer capabilities backed by [`zint`](https://sourceforge.net/projects/zint/), and updated APIs for reading and writing barcodes. These changes break backward compatibility, so we are upgrading to the next major version.
#### Consolidated Reader Function

## Breaking Changes
`readBarcodes(...)` replaces both `readBarcodesFromImageFile(...)` and `readBarcodesFromImageData(...)`. The new function can accept either a `Blob` or an `ImageData` as its input.

### Renamed & Consolidated Reader Function
> [!NOTE]
>
> The v1 reader functions `readBarcodesFromImageFile` and `readBarcodesFromImageData` are still kept for a smooth migration experience, but marked as deprecated.
`readBarcodes(...)` replaces both `readBarcodesFromImageFile(...)` and `readBarcodesFromImageData(...)`. The new function unifies code paths for `Blob` and `ImageData` inputs.
#### Updated Reader Options

> [!NOTE]
>
> The v1 reader functions `readBarcodesFromImageFile` and `readBarcodesFromImageData` are still kept for a smooth migration experience, but marked as deprecated.
A few reader options have changed their default values. This change is to align with the latest ZXing C++ library and provide a more consistent experience across different platforms:

### Updated Reader Options
1. `tryCode39ExtendedMode` is now `true` by default. It was previously `false`.
2. `eanAddOnSymbol` is now `"Ignore"` by default. It was previously `"Read"`.
3. `textMode` is now `"HRI"` by default. It was previously `"Plain"`.

A few reader options have changed their default values. This change is to align with the latest ZXing C++ library and provide a more consistent experience across different platforms:
Some deprecated options have been removed, see [zxing-cpp#704](https://github.com/zxing-cpp/zxing-cpp/discussions/704) for more details:

1. `tryCode39ExtendedMode` is now `true` by default. It was previously `false`.
2. `eanAddOnSymbol` is now `"Ignore"` by default. It was previously `"Read"`.
3. `textMode` is now `"HRI"` by default. It was previously `"Plain"`.
1. `validateCode39CheckSum` is now removed. The Code39 symbol has a valid checksum if the third character of the `symbologyIdentifier` is an odd digit.
2. `validateITFCheckSum` is now removed. The ITF symbol has a valid checksum if the third character of the `symbologyIdentifier` is a `'1'`.
3. `returnCodabarStartEnd` is now removed. The detected results of Codabar symbols now always include the start and end characters.

Some deprecated options have been removed, see [zxing-cpp#704](https://github.com/zxing-cpp/zxing-cpp/discussions/704) for more details:
#### `eccLevel` in Read Result Renamed to `ecLevel`

1. `validateCode39CheckSum` is now removed. The Code39 symbol has a valid checksum if the third character of the `symbologyIdentifier` is an odd digit.
2. `validateITFCheckSum` is now removed. The ITF symbol has a valid checksum if the third character of the `symbologyIdentifier` is a `'1'`.
3. `returnCodabarStartEnd` is now removed. The detected results of Codabar symbols now always include the start and end characters.
In `ReadResult`, the `eccLevel` field has been renamed to `ecLevel`. It now holds strings like `"L"`, `"M"`, `"Q"`, or `"H"` or stringified numeric percentage values for error correction levels. An empty string indicates that the error correction level is not applicable.

### `eccLevel` in Read Result Renamed to `ecLevel`
> [!NOTE]
>
> The `eccLevel` field is still kept for a smooth migration experience, but marked as deprecated.
In `ReadResult`, the `eccLevel` field has been renamed to `ecLevel`. It now holds strings like `"L"`, `"M"`, `"Q"`, or `"H"` or stringified numeric values for error correction levels. An empty string indicates that the error correction level is not applicable.
#### Renamed & Enhanced Writer Function

> [!NOTE]
>
> The `eccLevel` field is still kept for a smooth migration experience, but marked as deprecated.
`writeBarcode(...)` replaces `writeBarcodeToImageFile(...)`. This function is now powered by the new [`zint`](https://sourceforge.net/projects/zint/) backend which supports all available formats that are currently supported by the reader. It accepts either a `string` text or an `Uint8Array` binary data as its input for barcode generation, and provides new output formats (e.g. SVG, UTF-8) in addition to the an image file blob.

### Renamed & Enhanced Writer Function
The `WriterOptions` object has also been updated completely.

`writeBarcode(...)` replaces `writeBarcodeToImageFile(...)`. The new function is powered by the new [`zint`](https://sourceforge.net/projects/zint/) writer, which supports more barcode formats, supports both `string` and `Uint8Array` inputs for generating barcodes from text or binary data, and provides new output formats (e.g. SVG, UTF-8) in addition to the binary image file output.
> [!NOTE]
>
> The final shape of the `writeBarcode` function is still under discussion. The current implementation is subject to change.
The `WriterOptions` object has also been updated completely.
#### `.wasm` Module Initialization / Caching Overhaul

> [!NOTE]
>
> The final shape of the `writeBarcode` function is still under discussion. The current implementation is subject to change.
`prepareZXingModule(...)` replaces both `setZXingModuleOverrides(...)` and `getZXingModuleOverrides(...)`. The new function provides a more flexible way to initialize the ZXing module with custom options.

### Module Initialization / Caching Overhaul

`prepareZXingModule(...)` replaces both `setZXingModuleOverrides(...)` and `getZXingModuleOverrides(...)`. The new function provides a more flexible way to initialize the ZXing module with custom options.
> [!NOTE]
>
> The v1 module initialization functions `setZXingModuleOverrides` and `getZXingModuleOverrides` are still kept for a smooth migration experience, but marked as deprecated.
> [!NOTE]
>
> The v1 module initialization functions `setZXingModuleOverrides` and `getZXingModuleOverrides` are still kept for a smooth migration experience, but marked as deprecated.
`purgeZXingModule` now only clears the relevant module cache from where it is imported. It no longer resets the global module cache.

`purgeZXingModule` now only clears the relevant module cache from where it is imported. It no longer resets the global module cache.
#### Redefined `BarcodeFormat`-Family Types

## New Features & Enhancements
`None` is removed from the `BarcodeFormat` union type. New types like `LinearBarcodeFormat`, `MatrixBarcodeFormat` and `LooseBarcodeFormat` are introduced. See [`barcodeFormat.ts`](https://github.com/Sec-ant/zxing-wasm/blob/main/src/bindings/barcodeFormat.ts) for more details.

### More Barcode Formats Supported in Writer
### New Features & Enhancements

The new `writeBarcode` function supports more barcode formats than the previous `writeBarcodeToImageFile`. All barcode formats supported by the reader except for `DXFilmEdge` are now supported by the writer.
#### More Barcode Formats Supported in Writer

### New `tryDenoise` Option for Reading Barcodes
The new `writeBarcode` function supports more barcode formats than the previous `writeBarcodeToImageFile`. All barcode formats supported by the reader are now supported by the writer.

The new `tryDenoise` option in `ReaderOptions` allows you to enable or disable the denoising algorithm when reading barcodes. This is an experimental feature and by default, it is set to `false`.
#### New `tryDenoise` Option for Reading Barcodes

### Minor Changes
The new `tryDenoise` option in `ReaderOptions` allows you to enable or disable the denoising algorithm when reading barcodes. This is an experimental feature. By default, it is set to `false`.

- 615a321: - Add `DXFilmEdge` writing support.
- Fix subpath exports TS compatibility with types-versions-wildcards strategy. Check [this](https://github.com/andrewbranch/example-subpath-exports-ts-compat/tree/main/examples/node_modules/types-versions-wildcards) for more information.
- Add types to `.wasm` subpath exports.
- Add `ImageData` ambient type export.
### Bug Fixes

### Patch Changes
#### Fix TS `moduleResolution: node` Subpath Exports Resolution

- b856d58: Add `typesVersions` field for `moduleResolution: node`
- f8c33b2: Fix the `zxing_writer.wasm` size issue. See [#190](https://github.com/Sec-ant/zxing-wasm/discussions/190).
- a10ffcc: Bump `zxing-cpp` to `0dfa36b` to fix DataBarExpanded decoder error and ITF quiet zone detection heuristic
The subpath export types are now compatible with TypeScript's `moduleResolution: node` strategy by using the [types-versions-wildcards strategy](https://github.com/andrewbranch/example-subpath-exports-ts-compat/tree/main/examples/node_modules/types-versions-wildcards). This package now passes all the [`arethetypeswrong` checks](https://arethetypeswrong.github.io/?p=zxing-wasm%402.0.0).

## 2.0.0-beta.4

Expand Down

0 comments on commit 75d80cf

Please sign in to comment.