fix: reject empty/null input to match serverless platform behavior#302
Open
fix: reject empty/null input to match serverless platform behavior#302
Conversation
|
Promptless prepared a documentation update related to this change. Triggered by runpod/flash PR #302 Added documentation for the new empty input validation behavior — a warning block explaining that the input object must contain at least one field, HTTP 422 status code in the response codes table, and a troubleshooting section for empty/null input errors. Review at https://app.gopromptless.ai/suggestions/734d43e8-9534-4cab-80cb-76bca28b3d58 |
deanq
changed the title
deanq
force-pushed
the
deanq/ae-2744-payload-validation-parity
branch
from
d5ce86c to
0017c6c
Compare
Three-task TDD plan: (1) call_with_body validation in dev server, (2) deployed handler template validation, (3) create_deployed_handler rejection. Each task follows red-green-commit cycle.
call_with_body now checks model_fields_set and returns 422 when the
user sends {"input": {}}. This matches RunPod platform behavior,
which silently times out on empty input instead of dispatching the job.
Both function and class handler templates now return an error dict when job input is empty or null. Defense-in-depth: the RunPod platform typically blocks empty-input jobs, but this protects against platform behavior changes and direct handler testing.
Standardize on "Empty or null input." across all validation points for consistent developer experience when debugging across dev server and deployed endpoints.
Replace silent None-to-empty-dict fallback with explicit rejection.
Returns {success: false, error: ...} for empty/null input, consistent
with the dev server and handler template validation.
… exists When updating an endpoint that already has a templateId, the saveEndpoint payload included a fresh template object because _payload_exclude was called on new_config (templateId=None) instead of using the resolved templateId from the existing endpoint. This caused the API to create a new template with empty env, triggering a spurious release. Also removes superpowers docs that were incorrectly committed to the repo (they belong in flash-project/docs).
deanq
force-pushed
the
deanq/ae-2744-payload-validation-parity
branch
from
0017c6c to
d56737d
Compare
Contributor
There was a problem hiding this comment.
Pull request overview
This PR tightens input validation across the dev server (flash run) and deployed handler generation to reject empty/null input, aligning behavior with the RunPod serverless platform, and fixes an endpoint update path that could unintentionally trigger new template releases.
Changes:
- Reject empty/null job input for QB handlers (dev server + deployed handler factory) with a clear 422-style error message.
- Add empty-input guards to generated deployed handler templates (defense-in-depth) and update unit tests accordingly.
- Adjust
ServerlessResource.update()payload construction to avoid sending a freshtemplatewhen an existingtemplateIdshould be reused, preventing spurious releases.
Reviewed changes
Copilot reviewed 11 out of 11 changed files in this pull request and generated 3 comments.
Show a summary per file
| File | Description |
|---|---|
| tests/unit/test_p2_gaps.py | Updates template assertions to match new empty-input guard behavior. |
| tests/unit/runtime/test_generic_handler.py | Updates deployed handler tests to expect rejection for missing/None input. |
| tests/unit/runtime/test_generic_handler_extended.py | Adds/updates coverage around empty/null/missing input rejection. |
| tests/unit/runtime/test_deployed_handler.py | Updates deployed handler behavior expectations and test call shapes. |
| tests/unit/resources/test_serverless.py | Adds regression tests ensuring updates reuse existing templateId rather than sending a fresh template. |
| tests/unit/cli/commands/test_run_server_helpers.py | Adds tests for call_with_body rejecting empty Pydantic models while allowing explicit nulls. |
| tests/unit/cli/commands/build_utils/test_handler_generator.py | Adds tests asserting generated handler files include empty-input guards. |
| src/runpod_flash/runtime/generic_handler.py | Changes create_deployed_handler to reject empty/null input explicitly and return actionable errors. |
| src/runpod_flash/core/resources/serverless.py | Changes update payload exclusion logic to avoid sending template when an existing templateId is available. |
| src/runpod_flash/cli/commands/build_utils/handler_generator.py | Updates deployed handler templates to guard against empty/null input. |
| src/runpod_flash/cli/commands/_run_server_helpers.py | Adds dev-server-side empty-input validation in call_with_body via Pydantic model_fields_set. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
…emplateId override
Handler templates now use explicit None/empty-dict/non-dict checks
instead of `or {}` which could misclassify falsy non-dict inputs.
templateId injection in update() now only applies when new_config
does not explicitly set its own templateId, preserving intentional
template switches.
- Add "success": False to all handler template error dicts for consistency with create_deployed_handler - Document why create_handler intentionally accepts None input - Narrow call_with_body docstring to clarify Pydantic-only validation - Defensive copy of _payload_exclude() return value - Add debug log when template excluded from saveEndpoint payload - Add tests for empty plain dict body and falsy non-dict inputs - Fix stale test docstrings
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
flash run) QB routes now reject{"input": {}}with 422, matching RunPod platform behavior which silently times out on empty inputhandler_generator.py) return error dict for empty/null input (defense-in-depth)create_deployed_handlerreplaces silent None-to-empty-dict fallback with explicit rejectionsaveEndpointincluded a fresh template when updating an endpoint with an existingtemplateId, causing spurious releasesTest plan
flash runrejects{"input": {}}with 422 and actionable error messageflash runaccepts{"input": {"param_name": null}}for optional paramsflash buildgenerates handlers with empty-input validation guardmake quality-checkpasses (2512 tests, 85.59% coverage)Resolves AE-2744