Compaction/Safeguard: add summary quality audit retries (#25556)

Merged via squash.

Prepared head SHA: be473efd16
Co-authored-by: rodrigouroz <384037+rodrigouroz@users.noreply.github.com>
Co-authored-by: jalehman <550978+jalehman@users.noreply.github.com>
Reviewed-by: @jalehman
This commit is contained in:
Rodrigo Uroz
2026-03-05 18:39:25 -03:00
committed by GitHub
parent 029c473727
commit 036c329716
15 changed files with 967 additions and 65 deletions

View File

@@ -630,6 +630,18 @@ const STOP_WORDS_ZH = new Set([
"告诉",
]);
export function isQueryStopWordToken(token: string): boolean {
return (
STOP_WORDS_EN.has(token) ||
STOP_WORDS_ES.has(token) ||
STOP_WORDS_PT.has(token) ||
STOP_WORDS_AR.has(token) ||
STOP_WORDS_ZH.has(token) ||
STOP_WORDS_KO.has(token) ||
STOP_WORDS_JA.has(token)
);
}
/**
* Check if a token looks like a meaningful keyword.
* Returns false for short tokens, numbers-only, etc.
@@ -727,15 +739,7 @@ export function extractKeywords(query: string): string[] {
for (const token of tokens) {
// Skip stop words
if (
STOP_WORDS_EN.has(token) ||
STOP_WORDS_ES.has(token) ||
STOP_WORDS_PT.has(token) ||
STOP_WORDS_AR.has(token) ||
STOP_WORDS_ZH.has(token) ||
STOP_WORDS_KO.has(token) ||
STOP_WORDS_JA.has(token)
) {
if (isQueryStopWordToken(token)) {
continue;
}
// Skip invalid keywords