digital_ink_library (Python extension)

This is a Python implementation of the digital-ink-library. It provides classes for pen strokes and sketches which contain multiple such strokes. Both classes allow to store additional meta-data (standard Python types only) and can be serialized to JSON which enables cross-platform data exchange, e.g., to the Java implementation of the digital-ink-library.

Stroke and Sketch Classes

Metadata

Stroke and Sketch offer a field meta for storing/transferring additional meta-data. For pen strokes, this might be currently unsupported signals of the pen like tilt, for a sketch, this might be the pen's mac address. The meta field is a dictionary and can be used as follows (sample for Sketch):

from digital_ink_library.stroke import Sketch
sketch = Sketch()

# add meta data entry
sketch.meta['some key'] = 'some value'

# read and print a meta data entry
print(sketch.meta['some key'])  # prints 'some value'

Serialization

The following sample demonstrates how a Stroke instance can be serialized and deserialized. The same pattern can be used with instances of Sketch

from digital_ink_library.stroke import Stroke
from digital_ink_library.serialization.json import JsonDataSerialization

json_serializer = JsonDataSerialization()
stroke = Stroke(x=[1], y=[1], pressure=[1], timestamp=[1])

# write a stroke object to a json file
file_name = "define_some_filename.json"
with open(file_name, "w", encoding="utf-8") as fp:
    json_serializer.dump(obj=stroke, fp=fp)
    
# load a stroke object from a json file    
with open(file_name, "r", encoding="utf-8") as fp:
    loaded_stroke = json_serializer.load(fp=fp)

The JSON data format is compatible to other implementations of the digital-ink-library. A definition of the data format can be found here: TODO

Installation

run python setup.py bdist_wheel and install the resulting whl package.