method get_job_by_id() missing in Qiskit Function Intro

[abbycross]

Closes #2627 which was opened by @johannesgreiner. Johannes, can you provide more info on the fix please?

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[qiskit-bot]

One or more of the following people are relevant to this code:

• @abbycross
• @garrison
• @jenglick
• @johannesgreiner
• @kaelynj
• @pandasa123

[johannesgreiner]

So basically we'd want a few lines showing how to take an existing job_id and pass it to catalog.get_job_by_id() to load a job

[johannesgreiner]

e.g. we can use the existing old_jobs = catalog.jobs() and take jobid=old_jobs[0].job_id to get job_by_id = catalog.get_job_by_id(jobid)

[abbycross]

@Eric-Arellano when there is no print statement, would the assert statement still give an output of any kind? (What is here is left over from when there was a print statement, so I need to update it)

[Eric-Arellano]

No, the assert does not give output. Rather than hardcoding the output as markdown like this, it's better to update the code block to directly call print() and rely on nb-tester (i.e. run tox) to generate the output.

[abbycross]

I've been running the notebook locally to get outputs. Should I be leaving them blank and leave it to tox to supply the outputs?

[Eric-Arellano]

Running it locally is fine too

[Eric-Arellano]

This PR makes a bad revert

Suggested change

	"outputs": [],
	"outputs": [
	{
	"data": {
	"text/plain": [
	"'d76b6abf-3a16-4dd6-a9ac-a7ffa780d2f7'"
	]
	},
	"execution_count": 5,
	"metadata": {},
	"output_type": "execute_result"
	}
	],

[Eric-Arellano]

@johannesgreiner this seems wrong to me to say "by its position in the list"? I think this should instead read as something like:

If you already have the job ID for a certain job, you can retrieve it with catalog.get_job_by_id():

[abbycross]

Isn't the job id the thing you're retrieving, and rather you're trying to find the nth job id in a list?

[Eric-Arellano]

No, get_job_by_id is when you already have a particular job ID (some string) and you are getting a job (I think https://docs.quantum.ibm.com/api/qiskit-ibm-runtime/runtime-job-v2?)

and rather you're trying to find the nth job id in a list?

That's what you're doing with old_jobs[0]: you're getting the nth job where old_jobs is all the jobs that have run, sorted by date.

[Eric-Arellano]

job_by_id and latest_job should be the same thing. It might be clearer if we use job_by_id since that is what we are demoing

Suggested change

	"# Print the job_id for the latest job.\n",
	"print(latest_job)"
	"# Print the job_id for this job.\n",
	"print(job_by_id.job_id)"

[Eric-Arellano]

I think this will actually be more clear. It demonstrates that the entire Job is the same, not just the ID.

Suggested change

	"# Verify that the job ID is the same using both retrieval methods.\n",
	"assert job_by_id.job_id == latest_job.job_id\n",
	"# Verify that the job is the same using both retrieval methods.\n",
	"assert job_by_id == latest_job\n",

[Eric-Arellano]

Suggested change

	"latest_job = old_jobs[0]\n",
	"job_by_id = catalog.get_job_by_id(latest_job.job_id)\n",
	"latest_job = old_jobs[0]\n",
	"\n",
	"# We can also get that same job with get_job_by_id\n",
	"job_by_id = catalog.get_job_by_id(latest_job.job_id)\n",

[Eric-Arellano]

Suggested change

	"# Get the most recent job that has been executed.\n",
	"# First, get the most recent job that has been executed.\n",

[frankharkins]

Not sure this should be in the dictionary, I think we should either say " job ID" in a sentence, or use code stying (jobid) if referring to the method / variable.