fix(data): accept datetime bounds in the PIT P operator

[Divyansh Barodiya (Divyansh151005)]

Description

P now accepts the datetime bounds passed by LocalExpressionProvider when time2idx=False.

P._load_internal builds its result length with end_index - start_index + 1 and then walks integer calendar positions. With time2idx=True those bounds are calendar positions. With time2idx=False the provider passes the query datetimes through. Timestamp - Timestamp is a Timedelta, and Timedelta + 1 raises:

TypeError: unsupported operand type(s) for +: 'Timedelta' and 'int'

Datetime bounds are aligned with Cal.locate_index. That path returns a series indexed by the aligned trading days, so series.loc[start_time:end_time] in the provider still selects the requested range. Integer queries still return a RangeIndex. A future period reference is still rejected, and missing period data still returns an empty series.

Motivation and Context

Fixes #1551

P($$roewa_q) is the supported way to read a point-in-time feature on each trading day. time2idx=False is the expression-provider mode for datetime-indexed data, and P was the operator that crashed on those bounds.

How Has This Been Tested?

• Pass the test by running: pytest qlib/tests/test_all_pipeline.py under upper directory of qlib.
• If you are adding a new feature, test on your own test scripts.

test_all_pipeline.py trains a CSI300 LightGBM workflow and needs the market dataset. These tests stub the period feature and the calendar. They do not read PIT files or market data.

$ python -m pytest tests/ops/test_pit_operator.py -q --tb=line
....                                                                     [100%]
4 passed in 1.41s

flake8 on the changed Python files, with the ignore list from the developer guide, reported no issues. black -l 120 left the files unchanged.

Screenshots of Test Results (if appropriate):

1. Pipeline test: not run. It needs the CSI300 dataset and a model fit.
2. Your own tests: tests/ops/test_pit_operator.py, 4 passed.

Types of changes

• Fix bugs
• Add new feature
• Update documentation

[Divyansh Barodiya (Divyansh151005)]

@microsoft-github-policy-service agree