feat: add word_confidence_threshold parameter to whisper

[johnyrahul]

Summary

Adds the new word_confidence_threshold parameter to the v2 client's whisper() method, in sync with llmwhisperer-docs#57.

• word_confidence_threshold (float, default 0.3): the minimum OCR confidence score a word must have to be included in the extracted text. Words below the threshold are excluded from the final output.
• Works only with form, high_quality and table modes.

Changes

• Added the parameter to the whisper() signature, docstring, and the request params dict in client_v2.py.
• Added unit tests verifying the default (0.3) and a custom value are forwarded as request params.

Testing

uv run pytest tests/unit/client_v2_test.py -k word_confidence -v
# 2 passed

🤖 Generated with Claude Code

[greptile-apps]

Greptile Summary

This PR adds a word_confidence_threshold parameter (default 0.3) to the whisper() method in the v2 client, allowing callers to filter out low-confidence OCR words from extracted text. It also patches the integration test with a pre-test webhook cleanup to prevent failures from stale state.

• client_v2.py: new word_confidence_threshold: float = 0.3 parameter appended to the signature, docstring, and params dict; the table mode is also added to the mode docstring.
• tests/unit/client_v2_test.py: two new unit tests verify the default and a custom value are forwarded correctly as URL query params.
• tests/integration/client_v2_test.py: defensive delete_webhook call added before webhook registration to clean up any leftover state from prior runs.

Confidence Score: 5/5

Safe to merge — the change is a straightforward additive parameter with a sensible default, backed by unit tests and consistent with the existing parameter-passing pattern.

The new parameter is always forwarded unconditionally in the params dict (matching every other param in the method), the default of 0.3 preserves existing server-side behaviour, and two focused unit tests confirm both the default and a custom value reach the prepared request URL. No auth, data-integrity, or breaking-change concerns are introduced.

No files require special attention.

Important Files Changed

Filename	Overview
src/unstract/llmwhisperer/client_v2.py	Adds word_confidence_threshold parameter to whisper() signature, docstring, and params dict; also adds "table" to the mode docstring. No validation is performed client-side, consistent with all other numeric params in this method.
tests/unit/client_v2_test.py	Two new unit tests check that the default (0.3) and a custom (0.75) word_confidence_threshold value are forwarded as query params; uses parse_qs on the prepared request URL, which is the correct approach.
tests/integration/client_v2_test.py	Pre-test webhook cleanup added to prevent duplicate-registration failures from prior failed runs; no functional changes to the test assertions.

Sequence Diagram

sequenceDiagram
    participant Caller
    participant LLMWhispererClientV2
    participant LLMWhispererAPI

    Caller->>LLMWhispererClientV2: "whisper(url, word_confidence_threshold=0.3, ...)"
    Note over LLMWhispererClientV2: Build params dict including<br/>word_confidence_threshold
    LLMWhispererClientV2->>LLMWhispererAPI: "POST /whisper?...&word_confidence_threshold=0.3"
    LLMWhispererAPI-->>LLMWhispererClientV2: "{"status_code": 200, "extraction": {...}}"
    LLMWhispererClientV2-->>Caller: response dict

Loading

_{Reviews (3): Last reviewed commit: "test: delete pre-existing webhook before..." | Re-trigger Greptile}

[jaseemjaskp]

Automated PR Review Toolkit pass (Code Reviewer, Silent Failure Hunter, Type Design Analyzer, PR Test Analyzer, Comment Analyzer, Code Simplifier).

The change is small, additive, and follows the established whisper() convention exactly (new kwarg with default -> entry in the params dict). Blast radius is contained to this one public method; no callers need changes. Inline comments below, ordered by priority. Net recommendation: address the docstring "table" mode inaccuracy and tighten the test assertions; the rest are confirm/optional.

[jaseemjaskp]

Re-reviewed the latest revision with the PR Review Toolkit (code review, silent-failure, type-design, test, comment, and simplifier passes). All previously-raised threads have been addressed and resolved. Remaining observations are NIT/Suggestion-level only (no range validation by design — consistent with sibling numeric params; param forwarded for all modes and ignored server-side like include_line_confidence). No correctness, contract, or breaking-change issues found. LGTM 👍