You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
First thanks for the library it's been a big help for the project I am working on.
I can see some opportunities to make the lib faster. The bottle necks are the moment (unsurprisingly) are reading and writing to disk.
Testing with the GT-i9515 the read takes between 600ms - 1200ms. An obvious way to resolve this would be to hold the original Bitmap in memory. This would remove this bottle neck completely. Is there some reason for not doing this originally?
As for writing, some clients might not need CropIwa to write to disk.
My use case is that I merge the cropped image with another image before I write to disk. So the flow at the moment is..
setUri()
CropIwaread
Crop file
CropIwa read and write
Merge file
Client read and write
If CropIwa was able to pass back the cropped bitmap I would see huge performance improvements.
Would it be feasible to provide 2 Callback listeners? 1 for the Uri and 1 for the Bitmap?
I am happy to contribute if you think either of these are good directions to take.
The text was updated successfully, but these errors were encountered:
First thanks for the library it's been a big help for the project I am working on.
I can see some opportunities to make the lib faster. The bottle necks are the moment (unsurprisingly) are reading and writing to disk.
Testing with the GT-i9515 the read takes between 600ms - 1200ms. An obvious way to resolve this would be to hold the original
Bitmap
in memory. This would remove this bottle neck completely. Is there some reason for not doing this originally?As for writing, some clients might not need
CropIwa
to write to disk.My use case is that I merge the cropped image with another image before I write to disk. So the flow at the moment is..
CropIwa
readCropIwa
read and writeClient
read and writeIf
CropIwa
was able to pass back the cropped bitmap I would see huge performance improvements.Would it be feasible to provide 2 Callback listeners? 1 for the Uri and 1 for the Bitmap?
I am happy to contribute if you think either of these are good directions to take.
The text was updated successfully, but these errors were encountered: