read and store return-progress output on a dedicated output file

[burnout87]

No description provided.

[burnout87]

in relation to what discussed in oda-hub/mmoda-frontend-module#250 (comment)

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 61.34%. Comparing base (7331d28) to head (53dd71a).

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #670   +/-   ##
=======================================
  Coverage   61.34%   61.34%           
=======================================
  Files          49       49           
  Lines        8731     8732    +1     
=======================================
+ Hits         5356     5357    +1     
  Misses       3375     3375           

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

[burnout87]

With this, the idea is to write the outputs of the return-progress call, in dedicated files, namely:

• query_output.json.return-progress-query-output to store the output (eg html notebook output)
• query_output.json.return-progress-job-monitor to store the resulting job_monitor object

In addition to that, the "main" job object won't be overwritten (that is the cause of the reset of the progress bar), and consequently the content of the job_monitor.json file. The job output of the return-progress call will be stored in the mentioned above.

[burnout87]

tests on the nb2workflow plugin fail, but are addressed with oda-hub/dispatcher-plugin-nb2workflow#100

[volodymyrss]

Can it be instead that return_progress just adds some information to query output, not creating an entirely new output structure?

[volodymyrss]

Can it be instead that return_progress just adds some information to query output, not creating an entirely new output structure?

[burnout87]

With this, the idea is to write the outputs of the return-progress call, in dedicated files, namely:

* `query_output.json.return-progress-query-output` to store the output (eg html  notebook output)

* `query_output.json.return-progress-job-monitor` to store the resulting job_monitor object

In addition to that, the "main" job object won't be overwritten (that is the cause of the reset of the progress bar), and consequently the content of the job_monitor.json file. The job output of the return-progress call will be stored in the mentioned above.

EDIT

The query output of 'return-progress' is not in a separate file, and the previous behavior has been re-established.

The only difference is that the job-monitor will not be overwritten, thus preventing the progress bar from being reset.

[burnout87]

the nb2workflow plugin is now failing in test for the ontology, any ideas why ?

[dsavchenko]

the nb2workflow plugin is now failing in test for the ontology, any ideas why ?

Here is a test adaptation oda-hub/dispatcher-plugin-nb2workflow#101