opentelemetry-instrumentation-openai-v2: format test data for readability

[codefromthecrypt]

Description

Before this change, test data was hard to compare visually with assertions due to format. It is easier to do pull request reviews if you can easily read the data. For example, changing to YAML block scalar format.

   response:
     body:
-      string: "{\n  \"id\": \"chatcmpl-AMTlCEj20ZcsgWKZt8EizFMDItWNf\",\n  \"object\":
-        \"chat.completion\",\n  \"created\": 1729920978,\n  \"model\": \"gpt-4o-mini-2024-07-18\",\n
-        \ \"choices\": [\n    {\n      \"index\": 0,\n      \"message\": {\n        \"role\":
-        \"assistant\",\n        \"content\": \"This is a test. How can I assist you
-        further?\",\n        \"refusal\": null\n      },\n      \"logprobs\": null,\n
-        \     \"finish_reason\": \"stop\"\n    }\n  ],\n  \"usage\": {\n    \"prompt_tokens\":
-        12,\n    \"completion_tokens\": 12,\n    \"total_tokens\": 24,\n    \"prompt_tokens_details\":
-        {\n      \"cached_tokens\": 0\n    },\n    \"completion_tokens_details\":
-        {\n      \"reasoning_tokens\": 0\n    }\n  },\n  \"service_tier\": \"default\",\n
-        \ \"system_fingerprint\": \"fp_f59a81427f\"\n}\n"
+      string: |-
+        {
+          "id": "chatcmpl-APfFNvBVQx43PNOIf1dWnEUT5u5fA",
+          "object": "chat.completion",
+          "created": 1730680117,
+          "model": "gpt-4o-mini-2024-07-18",
+          "choices": [
+            {
+              "index": 0,
+              "message": {
+                "role": "assistant",
+                "content": "This is a test. How can I assist you further?",
+                "refusal": null
+              },
+              "logprobs": null,
+              "finish_reason": "stop"
+            }
+          ],
+          "usage": {
+            "prompt_tokens": 12,
+            "completion_tokens": 12,
+            "total_tokens": 24,
+            "prompt_tokens_details": {
+              "cached_tokens": 0
+            },
+            "completion_tokens_details": {
+              "reasoning_tokens": 0
+            }
+          },
+          "service_tier": "default",
+          "system_fingerprint": "fp_0ba0d124f1"
+        }

The upstream library has decided against custom serializers here. Instead, I used the same change I contributed to goose to solve this issue, which means fixture setup. While there is code visible to achieve this, I think it is worthwhile, and a change like this is better to do now vs later as test data accumulates.

Type of change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

How Has This Been Tested?

I deleted the existing test data, exported my OPENAI_API_KEY and ran the following:

tox -e py312-test-instrumentation-openai-v2

Doing so recreated the data. Then, I unexported my OPENAI_API_KEY and re-ran to ensure the data was used.

• py312-test-instrumentation-openai-v2

Does This PR Require a Core Repo Change?

• Yes. - Link to PR:
• No.

Checklist:

See contributing.md for styleguide, changelog guidelines, and more.

• Followed the style guidelines of this project
• Changelogs have been updated
• Unit tests have been added
• Documentation has been updated

[codefromthecrypt]

Notes:

• I changed failure assertions to be more precise as I had a test fail for the wrong reason.
• I intentionally didn't update the CHANGELOG as I thought it isn't worthwhile to end users.
• I noticed conventionally less doc strings and comments, but I retained the ones merged into the other project I did this to, as in developing I felt them helpful. That said, I'm happy to remove comments if that is more to style.

[codefromthecrypt]

ps I can't do this, so let me know if you want this in the CHANGELOG Please add a CHANGELOG entry, or add the "Skip Changelog" label if not required.

[codefromthecrypt]

p.s. is there any way to run ruff for instrumentation-openai-v2 as a part of tox -e py312-lint-instrumentation-openai-v2 as it seems both are lint so one should do the other?

[karthikscale3]

Only changelog needs to be added. everything else looks good

[codefromthecrypt]

Thanks for the look @karthikscale3!

Only changelog needs to be added. everything else looks good

So the current changelog has adding the initial openai impl, do we think we should elaborate things like polish to that initial impl? If so, happy to update. If not, maybe we can add the label to skip this check (I don't have karma to)

https://github.com/open-telemetry/opentelemetry-python-contrib/blob/main/CHANGELOG.md#unreleased

[xrmx]

Other than a small nit this LGTM, recordings are way more readable.

[lzchen]

@codefromthecrypt

Due to this refactor I've moved all genai related instrumentations under instrumentation-genai/ folder. Please rebase and the conflicts should be resolved :)

[AliWaleed97]

thanks @codefromthecrypt for doing this, neat PR!

[codefromthecrypt]

thanks for all the quick looks. rebased!