Daniel G. Krakowczyk

Do you mind giving me permission to run test workflows without further approval, please? I'd like to increase test coverage but currently you need to approve every single workflow run...

The failing tests are the same error as the one in the colab notebook (`CustomSeriesParam`).

@chr5tphr I have fixed the test for `MixedComposite` and now I think this PR should be ready for review.

this should be the fix: #729 can you approve and merge the changes afterwards from the mainbranch into your PR?

#715 should be resolved first, and it would solve the auto nest issues and remove a large part of the `if preprocessed` branch.

maybe instead of defining the dataset type via `datatype`, it would make more sense to define properties like `has_gaze_files` and `has_event_files` (and even `has_stimulus_files`)? Maybe this could be more flexible...

I think the signature of `Dataset.load()` is a bit different. While `DatasetDefinition.has_event_files` indicates if there are event files in the downloaded dataset to begin with, the `events` argument in `Dataset.load()`...

This is the current signature: https://github.com/aeye-lab/pymovements/blob/9eb46c56230cbe5deebbc7af6797745a46af1613/src/pymovements/utils/parsing.py#L110-L114 - [x] It should be changed to: ```python def parse_eyelink( filepath: Path | str, patterns: list | None = None, schema: dict | None...

#591 has added integration tests we now only need to cache the public datasets in a github workflow

putting it right after checking the time column during initialization seems most reasonable to me. https://github.com/aeye-lab/pymovements/blob/47e29b71f53a38e0c756f2a65e0232d53f8f9e66/src/pymovements/gaze/gaze_dataframe.py#L181-L184 creating a DataFrame `df_missing` with a single column `time` which holds all the missing...