[MNT] Changed line endings of ElectricDevices.csv and GunPoint.csv from CRLF to LF

[yarnabrina]

Reference Issues/PRs

See also #4446

What does this implement/fix? Explain your changes.

Most datasets in sktime have LF line endings, except ElectricDevices.csv and GunPoint.csv. This triggered auto-fix if configured in global gitattributes or global gitconfig. This PR attempts to convert those into LF line endings.

Before

❯ git ls-files --eol | grep -E "*.csv"
i/lf    w/lf    attr/                   sktime/datasets/data/Airline/Airline.cs
i/lf    w/lf    attr/                   sktime/datasets/data/Longley/Longley.cs
i/lf    w/lf    attr/                   sktime/datasets/data/Lynx/Lynx.csv
i/lf    w/lf    attr/                   sktime/datasets/data/PBS_dataset/PBS_dataset.csv
i/lf    w/lf    attr/                   sktime/datasets/data/ShampooSales/ShampooSales.csv
i/lf    w/lf    attr/                   sktime/datasets/data/Uschange/Uschange.csv
i/mixed w/mixed attr/                   sktime/datasets/data/segmentation/ElectricDevices.csv
i/mixed w/mixed attr/                   sktime/datasets/data/segmentation/GunPoint.csv
i/lf    w/lf    attr/                   sktime/datasets/data/solar/solar.csv

After

❯ git ls-files --eol | grep -E "*.csv"
i/lf    w/lf    attr/text               sktime/datasets/data/Airline/Airline.csv
i/lf    w/lf    attr/text               sktime/datasets/data/Longley/Longley.csv
i/lf    w/lf    attr/text               sktime/datasets/data/Lynx/Lynx.csv
i/lf    w/lf    attr/text               sktime/datasets/data/PBS_dataset/PBS_dataset.csv
i/lf    w/lf    attr/text               sktime/datasets/data/ShampooSales/ShampooSales.csv
i/lf    w/lf    attr/text               sktime/datasets/data/Uschange/Uschange.csv
i/lf    w/lf    attr/text               sktime/datasets/data/segmentation/ElectricDevices.csv
i/lf    w/lf    attr/text               sktime/datasets/data/segmentation/GunPoint.csv
i/lf    w/lf    attr/text               sktime/datasets/data/solar/solar.csv

Does your contribution introduce a new dependency? If yes, which one?

No

What should a reviewer concentrate their feedback on?

• Does any of the functionality break that is not covered by unit tests?

Did you add any tests for the change?

No

Any other comments?

PR checklist

For all contributions

• I've added myself to the list of contributors.
• Optionally, I've updated sktime's CODEOWNERS to receive notifications about future changes to these files.
• The PR title starts with either [ENH], [MNT], [DOC], or [BUG] indicating whether the PR topic is related to enhancement, maintenance, documentation, or bug.

For new estimators

• I've added the estimator to the online documentation.
• I've updated the existing example notebooks or provided a new one to showcase how my estimator works.

[fkiraly]

Looks simple enough, although the diff says there is a change in 13405 lines.
Can I ask for the clarification: the problem is in every single line, not just in the final line?

[yarnabrina]

Can I ask for the clarification: the problem is in every single line, not just in the final line?

I would guess so, not certain. To be specific, this is all I did:

1. open old file VS Code
2. change from CRLF to LF in bottom right
3. stage + commit

[fkiraly]

hm, I see.

Can I ask what is, to your knowledge, the easiest way to compare the content to be equal pre and post PR?
These are moderatly important benchmarking data sets, so we should avoid changing the content by accident (and I'm not sure how to best check that)

[yarnabrina]

Can I ask what is, to your knowledge, the easiest way to compare the content to be equal pre and post PR?

Something like this?

git difftool --tool=vimdiff origin/fix-line-endings origin/fix-line-endings~1

It shows me this (it's in LESS mode)

And ^M is carriage return character only, so not unexpected to see that difference from CRLF -> LF.

(There are probably better ways, I just don't know.)

[fkiraly]

Yes, these have the same content except for line endings changed.
Line endings are also consiste across data files post-PR.

[fkiraly]

Does any of the functionality break that is not covered by unit tests?

I don't think so - loading these is actually also tested, with some dependent functionality.