mirror of
https://github.com/openai/codex.git
synced 2026-11-16 22:00:36 +00:00
Avoid JSON-schema canonicalization (preserve field order)
Motivation: - App-server JSON schema output is already stable; sorting object keys is unnecessary. - Alphabetical key canonicalization is potentially harmful because it obscures intentional ordering (e.g. struct field order) and makes diffs noisier. What changed: - Remove JSON key-sorting from schema emission. - Regenerate vendored JSON schema fixtures. Refs: openai/codex#8956 (fixture sync test pattern)
This commit is contained in:
@@ -0,0 +1,26 @@
|
||||
{
|
||||
"$schema": "http://json-schema.org/draft-07/schema#",
|
||||
"properties": {
|
||||
"cancellationToken": {
|
||||
"type": [
|
||||
"string",
|
||||
"null"
|
||||
]
|
||||
},
|
||||
"query": {
|
||||
"type": "string"
|
||||
},
|
||||
"roots": {
|
||||
"items": {
|
||||
"type": "string"
|
||||
},
|
||||
"type": "array"
|
||||
}
|
||||
},
|
||||
"required": [
|
||||
"query",
|
||||
"roots"
|
||||
],
|
||||
"title": "FuzzyFileSearchParams",
|
||||
"type": "object"
|
||||
}
|
||||
Reference in New Issue
Block a user