Handle missing fields as nulls in get_json_object()

[SrikarVanavasam]

Addresses: #10196

Previously, get_json_object() ignored fields in a JsonPath expression that are missing in the json string. This PR adds the option to return these missing fields as null instead.

[GPUtester]

Can one of the admins verify this patch?

[raydouglass]

okay to test

[codecov]

Codecov Report

❗ No coverage uploaded for pull request base (branch-22.08@aec9007). Click here to learn what that means.
The diff coverage is n/a.

❗ Current head 34d8bfe differs from pull request most recent head 30ee5d9. Consider uploading reports for the commit 30ee5d9 to get more accurate results

@@               Coverage Diff               @@
##             branch-22.08   #10970   +/-   ##
===============================================
  Coverage                ?   86.34%           
===============================================
  Files                   ?      144           
  Lines                   ?    22738           
  Branches                ?        0           
===============================================
  Hits                    ?    19632           
  Misses                  ?     3106           
  Partials                ?        0           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update aec9007...30ee5d9. Read the comment docs.

[nvdbaranec]

@mlahir1 Do you guys have some time to try this out and see if it solves your issue?

[nvdbaranec]

rerun tests

[ttnghia]

I also don't see any tests for set_missing_fields_as_nulls = false.

[SrikarVanavasam]

I also don't see any tests for set_missing_fields_as_nulls = false.

set_missing_fields_as_nulls = false is set by default so I figured the existing tests would cover that case. Should I also have cases that explicitly set the condition?

[ttnghia]

set_missing_fields_as_nulls = false is set by default so I figured the existing tests would cover that case. Should I also have cases that explicitly set the condition?

I think so. We don't know, and are not guanteed, if the existing tests have that covered or not.

[mythrocks]

set_missing_fields_as_nulls = false is set by default so I figured the existing tests would cover that case. Should I also have cases that explicitly set the condition?

I think so. We don't know, and are not guanteed, if the existing tests have that covered or not.

Agreeing with @ttnghia's assessment. I meant to post this as well: It would be good to test set_missing_fields_as_nulls = true case alongside the false. It will underline the difference in behaviour.

[mythrocks]

Thank you for addressing the review comments. LGTM.

[SrikarVanavasam]

rerun tests

[SrikarVanavasam]

@gpucibot merge

[nvdbaranec]

@gpucibot merge

[mlahir1]

@mlahir1 Do you guys have some time to try this out and see if it solves your issue?

Thank you very much for the solving this. Sure, i belv this is merged in 22.08 nightly. i will test this sometime this week and check of this solves our issue.

[mlahir1]

@SrikarVanavasam , I could find any api changes on python side. Are the missing automatically assigned as Nulls or do i need to pass in some flag in the python API side.

[nvdbaranec]

Hi @mlahir1 . Yeah, sorry we missed this. We'll get a PR up with it soon.