Add docs for KedroDataCatalog

[ElenaKhaustova]

Description

Solves #4237

Development notes

• Improved KedroDaraCatalog docstrings;
• Updated DataCatalog index page to include information about KedroDaraCatalog - https://kedro--4249.org.readthedocs.build/en/4249/data/index.html
• Added new doc page with KedroDaraCatalog usage examples - https://kedro--4249.org.readthedocs.build/en/4249/data/kedro_data_catalog.html

Developer Certificate of Origin

We need all contributions to comply with the Developer Certificate of Origin (DCO). All commits must be signed off by including a Signed-off-by line in the commit message. See our wiki for guidance.

If your PR is blocked due to unsigned commits, then you must follow the instructions under "Rebase the branch" on the GitHub Checks page for your PR. This will retroactively add the sign-off to all unsigned commits and allow the DCO check to pass.

Checklist

• Read the contributing guidelines
• Signed off each commit with a Developer Certificate of Origin (DCO)
• Opened this PR as a 'Draft Pull Request' if it is work-in-progress
• Updated the documentation to reflect the code changes
• Added a description of this change in the RELEASE.md file
• Added tests to cover my changes
• Checked if this change will affect Kedro-Viz, and if so, communicated that with the Viz team

[merelcht]

This is looking good! I added some small nit comments. I'll do another full review later!

[ankatiyar]

Small suggestions but looks good to me!

[noklam]

Unrelated to docs. I find this not very intuitive, is there a symmetry between the get/setter? i.e.

catalog["dataset] = some_dataset # a real dataset class
catalog["a_list"] = [1,2,3] # a list

catalog["dataset] # return a dataset
catalog["a_list"] # Does this  return a dataset or list?

[ElenaKhaustova]

The setter acts similar to the add_feed_dict(), so it allows you to add either datasets or raw data. There's a note below explaining that.

When you add raw data, it is automatically wrapped in a `MemoryDataset` under the hood.

[noklam]

I find this is not very useful since this is more about educating users printing is not needed in a terminal? The most common use case is Notebook, but I don't think we need to explains how this work in IPython/Notebook etc.

I would like to see what catalog and catalog["reviews"] print differently instead (with an example), maybe something with this format:

In [1]: %%capture my_print_output
    ...: print('test')
    ...:

In [2]: my_print_output
Out[2]: <IPython.utils.capture.CapturedIO at 0x7f2efa2c12d0>

[ElenaKhaustova]

I've added an example of the output instead

[noklam]

Not strictly related to docs, do we expect Config Resolver as something that user will interact directly? I expect it's more of a refactoring and not a very user facing component (like KedroContext).

Would it be bad if we wrap it under KedroDataCatalog? i.e.

class KedroDataCatalog:
   ...
   def resolve_pattern(self, ds_name):
       return self.config_resolver(ds_name)

[ElenaKhaustova]

We do not expect users to interact with it much, though they can. It's mostly needed for catalog CLI commands and maybe some other advanced usage. But we do not want to extend catalog API with these public methods as then it will have to be part of the Protocol.

[noklam]

Why is del not supported, is it an implementation issue? What's the correct way to remove a dataset from catalog?

[ElenaKhaustova]

It's not an issue, it's done intentionally as we don't want datasets to be removed by users manually. Of course, one can still remove it from the private _datasets dictionary but we do not provide an API for that.

[merelcht]

Some minor comments, but otherwise this looks great! 🔥