Import and Export Schema Objects to JSON

[jrasband-dev]

Description

From an operational perspective it would be nice to read and write schemas.

Use Cases:

• Documentation and Versioning
• Reuse in other scripts
• Integration with other tools

Example solution for exporting schema

import json 
import polars as pl


custom_schema = pl.Schema({'Employee ID':pl.Int64
           ,'First Name':pl.String
           ,'Last Name':pl.String
           ,'Gender':pl.String
           ,'Date of Birth':pl.Date
           ,'Department':pl.String
           ,'Position':pl.String
           ,'Salary ($)':pl.Float64
           ,'Email':pl.String
           ,'Phone':pl.String})

def write_schema(fp):
    string_values = [str(value) for value in custom_schema.dtypes()]
    schema_dict = dict(zip(custom_schema.names(),string_values))
    f = open(fp,'w')
    f.write(json.dumps(schema_dict))
    f.close()

fp = 'employees_schema.json'
write_schema(fp)

Example solution for importing schemas

import json 

fp = 'employees_schema.json'

def read_schema(fp):
    f = open(fp,'r')
    schema = json.load(f)
    f.close()
    names = list(schema.keys())
    data_types = []
    for key, value in schema.items():
        if value == 'Int64':
            value = pl.Int64
            data_types.append(value)
        if value == 'String':
            value = pl.String
            data_types.append(value)
        if value == 'Date':
            value = pl.Date
            data_types.append(value)
        else:
            pass

    schema_dict = dict(zip(names,data_types))
    schema_object = pl.Schema(schema_dict)
    return schema_object

schema = read_schema(fp)
schema

@ritchie46