fix output for query cli

[jperez999]

Description

cleans output of query sub command to only display source, page_number and text.

Checklist

• I am familiar with the Contributing Guidelines.
• New or existing tests cover these changes.
• The documentation is up to date with these changes.
• If adjusting docker-compose.yaml environment variables have you ensured those are mimicked in the Helm values.yaml file.

[greptile-apps]

Greptile Summary

This PR filters the output of the query CLI sub-command so that only text, source, and page_number are returned to the user, dropping internal VDB fields such as _distance. The change is isolated to query_documents in sdk_workflow.py and is covered by an updated test that distinguishes raw retriever output from the public-facing projection.

• sdk_workflow.py: A list comprehension projects each hit to the three public fields using .get() with safe defaults, preventing KeyError on documents missing metadata.
• test_root_cli_workflow.py: Test fixtures split into raw_hits (full VDB response) and public_hits (expected CLI output), with assertions on both the parsed JSON and the exact formatted string.

Confidence Score: 4/5

Safe to merge after fixing the line-length violation that will cause the Black/Flake8 pre-commit hook to fail.

The logic change is correct and the test covers the new filtering behaviour. The single list comprehension in sdk_workflow.py is ~130 characters and will be rejected by the enforced Black + Flake8 pre-commit hooks, blocking CI on the changed file.

nemo_retriever/src/nemo_retriever/adapters/cli/sdk_workflow.py — line 85 needs reformatting to pass Black/Flake8 checks.

Important Files Changed

Filename	Overview
nemo_retriever/src/nemo_retriever/adapters/cli/sdk_workflow.py	Adds a projection step in query_documents to strip VDB-internal fields (_distance, etc.) and expose only text, source, and page_number. Uses .get() with safe defaults. One list comprehension line exceeds the 120-char limit enforced by Black/Flake8.
nemo_retriever/tests/test_root_cli_workflow.py	Test updated to distinguish raw_hits (full VDB response with _distance) from public_hits (filtered output), correctly asserting the CLI prints only the three public fields.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[CLI: retriever query] --> B[query_documents]
    B --> C[Retriever.query]
    C --> D[LanceDB VDB]
    D --> E["raw hits\n{text, source, page_number,\n_distance, ...}"]
    E --> F["project hits\n{text, source, page_number}"]
    F --> G[JSON output to stdout]

Loading

Prompt To Fix All With AI

Fix the following 1 code review issue. Work through them one at a time, proposing concise fixes.

---

### Issue 1 of 1
nemo_retriever/src/nemo_retriever/adapters/cli/sdk_workflow.py:85
The list comprehension on this line is ~130 characters, exceeding the project's enforced 120-character limit (Black + Flake8). The pre-commit hook will reject the file as-is, blocking CI. Black would reformat it to a multi-line style.

```suggestion
    hits = [
        {
            "text": hit.get("text", ""),
            "source": hit.get("source", ""),
            "page_number": hit.get("page_number"),
        }
        for hit in hits
    ]
```

_{Reviews (3): Last reviewed commit: "Merge branch 'clean-cli-skills' of https..." | Re-trigger Greptile}

[greptile-apps]

The list comprehension on this line is ~130 characters, exceeding the project's enforced 120-character limit (Black + Flake8). The pre-commit hook will reject the file as-is, blocking CI. Black would reformat it to a multi-line style.

Suggested change

	hits = [{"text": hit.get("text", ""), "source": hit.get("source", ""), "page_number": hit.get("page_number")} for hit in hits]
	hits = [
	{
	"text": hit.get("text", ""),
	"source": hit.get("source", ""),
	"page_number": hit.get("page_number"),
	}
	for hit in hits
	]

Prompt To Fix With AI

This is a comment left during a code review.
Path: nemo_retriever/src/nemo_retriever/adapters/cli/sdk_workflow.py
Line: 85

Comment:
The list comprehension on this line is ~130 characters, exceeding the project's enforced 120-character limit (Black + Flake8). The pre-commit hook will reject the file as-is, blocking CI. Black would reformat it to a multi-line style.

```suggestion
    hits = [
        {
            "text": hit.get("text", ""),
            "source": hit.get("source", ""),
            "page_number": hit.get("page_number"),
        }
        for hit in hits
    ]
```

How can I resolve this? If you propose a fix, please make it concise.

[randerzander]

returns are much cleaner, thanks!