Implement cancel job operation for CLI

[amritghimire]

This introduces a cancel job operation for CLI that can be used to
cancel a job running in studio.

Command:
datachain studio cancel <JOB_ID

[cloudflare-workers-and-pages]

Deploying datachain-documentation with    Cloudflare Pages

Latest commit:	6fce628
Status:	✅  Deploy successful!
Preview URL:	https://6b4cda96.datachain-documentation.pages.dev
Branch Preview URL:	https://amrit-cancel-job.datachain-documentation.pages.dev

View logs

[codecov]

Codecov Report

Attention: Patch coverage is 90.00000% with 2 lines in your changes missing coverage. Please review.

Project coverage is 87.32%. Comparing base (b67d599) to head (6fce628).
Report is 3 commits behind head on main.

Files with missing lines	Patch %	Lines
src/datachain/studio.py	83.33%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #695   +/-   ##
=======================================
  Coverage   87.32%   87.32%           
=======================================
  Files         113      113           
  Lines       10798    10817   +19     
  Branches     1480     1483    +3     
=======================================
+ Hits         9429     9446   +17     
- Misses        991      992    +1     
- Partials      378      379    +1     

Flag	Coverage Δ
datachain	87.26% <90.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[dreadatour]

Looks good to me!

I am not sure datachain studio cancel is a good command name, it is too general and not verbose enough IMO. But I have no better options.

[shcheklein]

thanks @amritghimire !

datachain studio cancel <JOB_ID

let's rename it please though.

It should be:

datachain job cancel --studio (similar to ds)

same with run, same with API - it should be consistent + we should be using PUT, POST, DELETE, etc

[amritghimire]

thanks @amritghimire !

datachain studio cancel <JOB_ID

let's rename it please though.

It should be:

datachain job cancel --studio (similar to ds)

I think that it may be confusing to users since we don't have non studio version of job in local. The reason I included the cancel inside studio is that the operation is a concept totally located in Studio.

[amritghimire]

same with run, same with API - it should be consistent + we should be using PUT, POST, DELETE, etc

https://github.com/iterative/studio/pull/11077#discussion_r1885529578

[shcheklein]

I think that it may be confusing to users since we don't have non studio version of job in local. The reason I included the cancel inside studio is that the operation is a concept totally located in Studio.

I see the point.

But, I don't think it justifies the inconsistency tbh (datasets, jobs, etc - all should be the same in API and in CLI to my mind). Also the name "datachain studio cancel" is strange a bit anymways. It should be then datachain studio job cancel

or let's find some other options please ...

[shcheklein]

https://github.com/iterative/studio/pull/11077#discussion_r1885529578

is it PUT / POST though then?

[amritghimire]

I think that it may be confusing to users since we don't have non studio version of job in local. The reason I included the cancel inside studio is that the operation is a concept totally located in Studio.

I see the point.

But, I don't think it justifies the inconsistency tbh (datasets, jobs, etc - all should be the same in API and in CLI to my mind). Also the name "datachain studio cancel" is strange a bit anymways. It should be then datachain studio job cancel

or let's find some other options please ...

WDYT about datachain studio cancel-job?

is it PUT / POST though then?

It is POST currently. We already have POST request for update-job.

[shcheklein]

WDYT about datachain studio cancel-job?

I think it is still not good, since it's not consistent with datasets. We should be using the same general schema for all types of entities. This is quite important I think.