Display beginning of non-text files as text instead of triggering a download

[SergeyYakubov]

display the first 100K of a binary file or a file whose type is unknown to Galaxy as ASCII text instead of downloading this file.

If a user clicks on the "eye" button for a binary dataset (or a dataset with an unknown type), Galaxy downloads the whole file. This is probably not what a user wants, especially for large files. Even worse, when using an object store, the file will be fetched from that store and then sent to the web client. So, we implemented displaying a warning and first 100K as an ASCII text, which in many cases still gives an idea of what is inside the file.

How to test the changes?

(Select all options that apply)

• I've included appropriate automated tests.
• This is a refactoring of components with existing test coverage.
• Instructions for manual testing are as follows:
  1. Upload a binary file to Galaxy
  2. Click on the "eye" button to display the file.

License

• I agree to license these and all my past contributions to the core galaxy codebase under the MIT license.

[jdavcs]

The test failures were, likely, relevant (can't verify as logs have expired). Also, there's a conflict that needs to be resolved.
Bumping to 23.2 for now. We can re-tag it with 23.1, if needed, if these are resolved tomorrow.

[jdavcs]

Thank you for addressing these!
Can you, please, do a rebase instead of a merge? That way we get a clean commit history. I'm happy to help if needed (feel free to ping me).

[SergeyYakubov]

Thank you for addressing these! Can you, please, do a rebase instead of a merge? That way we get a clean commit history. I'm happy to help if needed (feel free to ping me).

I tried to rebase first, but I got much more conflicts this way :(. Apparently, merge and rebase use different algorithms. But if you know how to rebase without conflicts, please let me know.

[jdavcs]

I tried to rebase first, but I got much more conflicts this way ...

I've looked at it; there is one conflict, which seems not too complicated, unless I'm missing something. Still, I'd verify that your logic is still intact and all works as expected after the conflict is fixed. I'll post the specific steps for fixing it to the dev channel.

[jdavcs]

Thank you, @SergeyYakubov, this should be quite useful!

[martenson]

This is great, thanks @SergeyYakubov !