Delete resource from everywhere when force_delete is selected

[thesujai]

Summary

1. Whenever force_delete is selected, the ContentNodes with same content_id is appended to the node_ids so the content can be deleted from everywhere
2. Adds a validate method to DeleteChannelValidator so force_delete wont be ignored in the validation process

References

Fixes #12606

Reviewer guidance

Try and replicate this, if it doesn't replicate then we are good to go.

---

Testing checklist

• Contributor has fully tested the PR manually
• If there are any front-end changes, before/after screenshots are included
• Critical user journeys are covered by Gherkin stories
• Critical and brittle code paths are covered by unit tests

PR process

• PR has the correct target branch and milestone
• PR has 'needs review' or 'work-in-progress' label
• If PR is ready for review, a reviewer has been added. (Don't use 'Assignees')
• If this is an important user-facing change, PR or related issue has a 'changelog' label
• If this includes an internal dependency change, a link to the diff is provided

Reviewer checklist

• PR is fully functional
• PR has been tested for accessibility regressions
• External dependency files were updated if necessary (yarn and pip)
• Documentation is updated
• Contributor is in AUTHORS.md

[nucleogenesis]

If the issue is fixable with only the fix to the validator class, then we should avoid adding additional untested code to the command itself. I think that if the force-deletion isn't applied even when the data is successfully passed from front-end through to the creation of the task, then the additional code will be necessary.

[nucleogenesis]

I'm curious if maybe this addition isn't necessary - could you try testing it with these changes not applied?

I suspect that the validate method being fixed, which ensures that force_delete is passed to the task, then the feature should work as it did before. I mainly suspect this because it is a feature that used to work but no longer does.

I haven't looked at it closely, but I'd guess that delete_metadata uses the force_delete to the expected effect.

[thesujai]

That force_delete is behaving differently than my logic here. It deletes the entire channel when a single resource is force_deleted.
See In my Imported Channel. I have some content titled Sample Video that is present in both Topic 1 and Imported Folder(They share the same content_id):

So if we force_deleted the Sample Video, only that should have been deleted from both folders. But the entire channel gets deleted:
Screencast from 26-09-24 11:31:00 AM IST.webm

Note: This is only with the changes to DeleteChannelValidator without my change in the deletecontent command. Even with my changes, the behavior remains the same. I

[nucleogenesis]

@thesujai thank you for that and apologies for my delay in responding here.

We've stumbled upon a chain of bugs... 😅

This all started when I found that the checkbox to select "force_delete" had a bug, which then led you here to find the bug in the DeleteChannelValidator#validate method, which also led you to find that there is a bug with how we're handling granular deletion as well.

---

I was expecting the scope of this issue to be more along the lines of what the validator updates fix: the user's selection is passed along to the queuing of the deletion task.

Let's get that fix merged in this PR and fixing the deletion issue itself can be done in a follow-up issue (which you're welcome to take on if you're interested).

[thesujai]

reverted! I can try fixing this behavior in a follow-up

[github-actions]

Build Artifacts

Asset type	Download link
PEX file	kolibri-0.17.3a0.dev0_git.22.g62a70edb.pex
Windows Installer (EXE)	kolibri-0.17.3a0.dev0+git.22.g62a70edb-windows-setup-unsigned.exe
Debian Package	kolibri_0.17.3a0.dev0+git.22.g62a70edb-0ubuntu1_all.deb
Mac Installer (DMG)	kolibri-0.17.3a0.dev0+git.22.g62a70edb.dmg
Android Package (APK)	kolibri-0.17.3a0.dev0+git.22.g62a70edb-0.1.4-debug.apk
TAR file	kolibri-0.17.3a0.dev0+git.22.g62a70edb.tar.gz
WHL file	kolibri-0.17.3a0.dev0+git.22.g62a70edb-py2.py3-none-any.whl

[nucleogenesis]

Let's get the changes to deletecontent reverted here and just merge the fix to the validator - despite now knowing that there is something weird going on when things are actually deleted.

I will make a follow-up issue for that particular problem.

[nucleogenesis]

@thesujai thank you for that and apologies for my delay in responding here.

We've stumbled upon a chain of bugs... 😅

This all started when I found that the checkbox to select "force_delete" had a bug, which then led you here to find the bug in the DeleteChannelValidator#validate method, which also led you to find that there is a bug with how we're handling granular deletion as well.

---

I was expecting the scope of this issue to be more along the lines of what the validator updates fix: the user's selection is passed along to the queuing of the deletion task.

Let's get that fix merged in this PR and fixing the deletion issue itself can be done in a follow-up issue (which you're welcome to take on if you're interested).

[nucleogenesis]

This LGTM @thesujai -- I'll be making a follow-up issue next week after chatting to @rtibbles about it.

Thank you so much for all of your efforts investigating and working on this @thesujai <3

[rtibbles]

Just did a quick retarget and rebase, as this is aimed at 0.17 planned patch 2. Will merge once the build completes.