fix(python): Fix boolean path parameters serialized as True/False instead of true/false

[jsklan]

Description

Fix boolean path parameters in the Python SDK being serialized as Python-style True/False instead of JSON-style true/false in URL paths. This causes WireMock test failures and incorrect API calls when boolean values appear in path segments.

Root cause: The generator used jsonable_encoder() for path parameter interpolation. Since bool is a subclass of int in Python, jsonable_encoder returns booleans as-is, and Python's default str() produces True/False when interpolated into f-strings.

Changes Made

• Added encode_path_param() utility function in jsonable_encoder.py that explicitly handles booleans with lowercase true/false, and delegates to jsonable_encoder + str() for all other types
• Registered encode_path_param as an exported core utility in core_utilities.py
• Updated endpoint_function_generator.py to use encode_path_param instead of jsonable_encoder for path parameter interpolation
• Updated websocket_connect_method_generator.py to also use encode_path_param (same pattern as endpoint generator — flagged by Devin Review)
• Added getWithBooleanPath endpoint to the exhaustive test fixture to cover boolean path parameters
• Updated IR snapshot files for the exhaustive fixture (both ir and dynamic-snippets)
• Version bump to 5.3.4 with changelog entry

Important Review Notes

• The str(jsonable_encoder(obj)) fallback in encode_path_param mirrors the existing implicit stringification that happens when values are interpolated into f-strings, so non-boolean types should behave identically to before.
• ⚠️ Seed snapshot files: The seed-test-results (python-sdk) CI check passes (generated code compiles and tests pass), but the actual seed snapshot files under seed/python-sdk/exhaustive/ have not been committed in this PR. A reviewer may want to run pnpm seed:local test --generator python-sdk --fixture exhaustive locally and commit the updated snapshots, or confirm CI's passing result is sufficient.
• The test-ete CI check failure is unrelated — it's a timeout in generate-with-settings.test.ts > dependencies-based api and is not marked as required.

Testing

• Pre-commit hooks pass (poetry run pre-commit run -a)
• pnpm format passes
• IR snapshot tests updated and passing locally
• All required CI checks passing (test, compile, lint, seed-test-results, etc.)
• Reproduced the bug locally via pnpm seed:local run --generator python-sdk --path <customer-api-path> — confirmed WireMock error: Path parameter: ignoreEmpty = true | True <<<<< Path parameter does not match

For reviewer

• Are there other code paths beyond endpoint_function_generator.py and websocket_connect_method_generator.py that interpolate path parameters and may need the same fix?
• Verify encode_path_param handles edge cases (e.g., None, enums, nested types) — the str(jsonable_encoder(obj)) fallback should cover these, but worth confirming
• Decide whether seed snapshot files under seed/python-sdk/exhaustive/ need to be committed (CI seed tests pass without them)

Link to Devin session: https://app.devin.ai/sessions/25a25e3ba67447f59aa5d11486e44726
Requested by: @jsklan

---

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment and CI monitoring

[devin-ai-integration]

Devin Review found 1 potential issue.

🐛 1 issue in files not directly in the diff

🐛 Websocket path parameters not updated to use encode_path_param (generators/python/src/fern_python/generators/sdk/client_generator/websocket_connect_method_generator.py:542)

The PR's fix for boolean path parameter encoding was applied to endpoint_function_generator.py (line 1019) but the identical pattern in websocket_connect_method_generator.py:542 still uses jsonable_encoder. This means websocket endpoints with boolean path parameters will still serialize as Python-style True/False instead of lowercase true/false. The two methods (_get_path_for_endpoint and _get_path_for_websocket) are structurally identical, and the transformation was only applied to one of them.

View 2 additional findings in Devin Review.

[devin-ai-integration]

Addressed the Devin Review finding — updated websocket_connect_method_generator.py to also use encode_path_param instead of jsonable_encoder for path parameter interpolation (commit 9f6894f).

[claude]

Claude Code Review

This repository is configured for manual code reviews. Comment @claude review to trigger a review and subscribe this PR to future pushes, or @claude review once for a one-time review.

_{Tip: disable this comment in your organization's Code Review settings.}

[github-actions]

SDK Generation Benchmark Results

Comparing PR branch against latest nightly baseline on main (2026-04-08T04:49:09Z).

Full benchmark table (click to expand)

Generator	Spec	main (generator)	main (E2E)	PR (generator)	Delta
csharp-sdk	square	132s	171s	91s	-41s (-31.1%)
go-sdk	square	280s	345s	105s	-175s (-62.5%)
java-sdk	square	311s	371s	273s	-38s (-12.2%)
php-sdk	square	127s	155s	85s	-42s (-33.1%)
python-sdk	square	168s	204s	124s	-44s (-26.2%)
ruby-sdk-v2	square	152s	192s	112s	-40s (-26.3%)
rust-sdk	square	137s	127s	94s	-43s (-31.4%)
swift-sdk	square	141s	486s	102s	-39s (-27.7%)
ts-sdk	square	142s	173s	108s	-34s (-23.9%)

main (generator): generator-only time via --skip-scripts (includes Docker image build, container startup, IR parsing, and code generation — this is the same Docker-based flow customers use via fern generate). main (E2E): full customer-observable time including build/test scripts (nightly baseline, informational). Delta is computed against generator-only baseline.
⚠️ = generation exited with a non-zero exit code (timing may not reflect a successful run).
Baseline from nightly runs on main (latest: 2026-04-08T04:49:09Z). Trigger benchmark-baseline to refresh.