CSV Dataset append mode not working with expected args

[maxschulz-COL]

Description

When using a csv dataset, the below doesn't work

model_metrics_aggregated:
  type: pandas.CSVDataset
  filepath: data/08_reporting/model_metrics_aggregated.csv
  save_args:
    mode: a

but this one does

model_metrics_aggregated:
  type: pandas.CSVDataset
  filepath: data/08_reporting/model_metrics_aggregated.csv
  fs_args:
    open_args_save:
      mode: a

Why? From the docs it seems that the top one should be the route to go...
I used these docs: https://docs.kedro.org/projects/kedro-datasets/en/kedro-datasets-2.0.0.post1/api/kedro_datasets.pandas.CSVDataset.html

Context

It annoyed me :)

Steps to Reproduce

Just use the dataset as described above

Expected Result

Append to work by using the described commands

Actual Result

No append

Your Environment

Latest kedro

[merelcht]

@maxschulz-COL what version of kedro-datasets are you using? This was fixed in version 5.0.0 with this PR kedro-org/kedro-plugins#805

[maxschulz-COL]

@maxschulz-COL what version of kedro-datasets are you using? This was fixed in version 5.0.0 with this PR kedro-org/kedro-plugins#805

kedro-datasets==6.0.0

But just to be clear (after glancing at the PR): the fs_args args route works, but the save_args doesn't.

[merelcht]

Ah righhttt yes sorry. So I guess the main point is here that I should've updated the docs too 😅

[ravi-kumar-pilla]

Hi @maxschulz-COL , Thank you for raising the issue. Would you be happy to raise a docs PR for the same ?

[maxschulz-COL]

Hi @maxschulz-COL , Thank you for raising the issue. Would you be happy to raise a docs PR for the same ?

I think I may not find the time, but if I do, I am very happy of course!