keep the python client generator within memory (#12546)

* fix: keep the python client generator within memory

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* chore: state the generator's real memory peak

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Ruben Fiszel
2026-11-19 17:51:06 +01:00
committed by GitHub
co-authored by Claude Opus 5.5
parent 6c2c5cf46c
commit 4472b5b94b
6 changed files with 10 additions and 6 deletions
+2 -1
View File
@@ -7,7 +7,8 @@ on:
jobs:
publish_pypi:
runs-on: ubicloud-standard-8
# openapi-python-client peaks near 32 GB on the dereferenced spec
runs-on: ubicloud-standard-16
if: startsWith(github.ref, 'refs/tags/v')
container:
image: ghcr.io/windmill-labs/python-client-builder
+1 -1
View File
File diff suppressed because one or more lines are too long
+4 -1
View File
@@ -1429,8 +1429,11 @@ components:
test_subflows:
type: object
description: In a flow test run, deployed subflow path -> the flow value with the test's step overrides applied, run in place of the deployed subflow.
# A FlowValue, left untyped: FlowStatus is inlined into every job schema, and the
# python client generator (which dereferences the whole spec) runs out of memory
# when each copy also carries a full FlowValue.
additionalProperties:
$ref: "#/components/schemas/FlowValue"
type: object
required:
- step
- modules
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long