BUG: Creating series from DataFrame does not serialize Timestamp object to MILLISECONDS from Unix epoch #74
Description
Problem
When creating a Highcharts Series object from a pandas Dataframe(), any column with pandas Timestamp() object does not get serialized as milliseconds from the Unix epoch when using the to_dict() and to_json() methods. to_js_literal() does the serialization correctly but it's not ideal when I want to send data through an API. The string returned by the to_js_literal() cannot be parsed with Javascript's JSON.parse() method.
Example
# Sample DataFrame saved in df
timestamp value
2023-04-01 13669.0
2023-04-02 21534.0
2023-04-03 23207.0
After creating Highcharts series using series = LineSeries.from_pandas(df,property_map={"x": "timestamp", "y": "value"})
# series.to_js_literal() ouputs
"{
data: [{
x: 1680307200000.0,
y: 13669.0
},
{
x: 1680393600000.0,
y: 21534.0
},
{
x: 1680480000000.0,
y: 23207.0
}],
type: 'line'
}"
# series.to_dict() ouputs
{'data': [{'x': Timestamp('2023-04-01 00:00:00'), 'y': 13669.0}, {'x': Timestamp('2023-04-02 00:00:00'), 'y': 21534.0}, {'x': Timestamp('2023-04-03 00:00:00'), 'y': 23207.0}], 'type': 'line'}
# series.to_json() just breaks with the following error
TypeError: Object of type Timestamp is not JSON serializable
While the to_js_literal() method gives me what I want, it's difficult to work with when sending data over an API.
Expected behavior
I expected to_json() to have the same output as to_js_literal(), but in JSON format and to_dict() to have serialized the Timestamp() object to milliseconds from Unix epoch.
Environment:
- OS: Linux
- Python Version: 3.10.8
- Highcharts Python Version: 1.2.3