Tiny library to parse Png files (or uint8Arrays containing Png data structures) into JSON human readable Objects. It extracts all the data from the IHDR chunk, and creates an array of chunks with it's relative byte blocks.
Accepts different data sources :
- HTML Input Elements : Handles automatically the File change events (async)
- File references : From HTML Input Element (async)
- Uint8Array : Typed Array containing the data (sync)
Usage example (Node):
let fileData = fs.readFileSync('./test.png');
let pngJSON = pngToJSON(fileData);
console.log(pngJSON);
pngToJSON( HTMLInputElement, callback );
HTMLInputElement
: Reference to the DOM File Input Elementcallback
: Function to receive the results of the conversion. It accepts two arguments :JSON PNG Object
andSource File Info
.
pngToJSON( FileReference, callback )
FileReference
: Reference to the file returned from the File Input Element ( InputElement.files[x] )callback
: Function to receive the results of the conversion. It accepts two arguments :JSON PNG Object
andSource File Info
.
pngToJSON( Uint8ArrayData )
Uint8ArrayData
:Uint8Array typed Array containing the png file data
Returns : JSON representation of the png file contents.
The returned JSON Object i strutured in thw following way (most keys are self explanatory)
pngJSON{
bitDepth: Integer,
chunks: Integer, // total count
colorType: Integer,
colorTypeString: String, // string representation
compressionMethod: Integer,
data: Uint8Array[...], // png file data
filterMethod: Integer,
height: Integer, // image height
interlaceMethod: Integer,
signature: Uint8Array[...], // firat 8 bytes
width: Integer, // image width
chunk: [
[0] : {
data: Uint8Array[...], // chunk data block
raw : Uint8Array[...], // full chunk block
id: Integer,
isCritical: Boolean,
isPublic: Boolean,
chunkLength: Integer, // chunk length
dataLength: Integer, // chunk data length
offset: Integer, // chunk start offset
type: String, // chunk type
},
[1] : {...},
[2] : {...},
(...)
]
}
Note : The resulting <pngJSON>.chunk[x].data
byteArrays, do not contain the the full chunk data. The bytes corresponding to the length block
, type block
, and CRC block
, are stripped from it. It only contains the data block
from each chunk.
CHUNK BYTE STRUCTURE :
<------------------------ chunk length -------------------------->
Chunk Offset Chunk Offset+8
├──────────────────┬───────────┼──────────────────────┬──────────┐
│ *DataLength (4b) │ Type (4b) │ Data (*DataLength b) │ CRC (4b) │
└──────────────────┴───────────┴──────────────────────┴──────────┘
<--------- stripped -----------> <-stripped->
Use <pngJSON>.chunk[x].raw
instead, to get all the chunk data.
In browser enviroment you can include this library using the jsdelivr CDN ( window.pngToJSON() is created autmatically ) ...
<script src='https://cdn.jsdelivr.net/gh/colxi/png-to-json@latest/src/png-to-json.min.js'></script>
If you are in the NodeJs enviroment, can install the package via:
$ npm install png-to-json --save
Git repository available too :
$ git clone https://github.com/colxi/png-to-json.git
The package can be imported using ES6 import
in Browser
import '<libraryPath>/src/png-to-json.js';
// window.pngToJSON() is created autmatically
GPL 3