Why a grep result does not cover everything the caller asked for. Absent means the scan was complete
over the requested scope, and only then may an empty matches be read as the pattern being absent.
At most one applies; the server picks the most significant.
timedOut — the scan hit its server-side deadline. Routine on large knowledge bases, since the
match is unindexed. Says nothing at all about whether the pattern occurs.
noIndexedText — nothing in scope has stored text, so nothing was searched. Contexts ingested
before literal search existed sit here until they are re-ingested.
filterMatchedNoContexts — metadataFilter, composed with any visibility filter the application
applies, admitted no contexts, so the scan had nothing to read.
scopeTruncated — metadataFilter admitted more contexts than one scan can cover, so only a prefix
of them was searched. Narrow the filter.
partsCapReached — the scan filled its per-call budget of matching pages/sheets, so more matching
places exist than it read.
truncatedContent — some in-scope text was cut at ingest, either because one page or sheet exceeded
the per-part storage cap or because a document had more pages than the per-document cap, so that
content was never searchable.
coverageUnknown — the scan itself succeeded, but how much of the scope holds stored text could not
be determined, so completeness cannot be asserted either way.
Why a grep result does not cover everything the caller asked for. Absent means the scan was complete over the requested scope, and only then may an empty
matchesbe read as the pattern being absent. At most one applies; the server picks the most significant.timedOut— the scan hit its server-side deadline. Routine on large knowledge bases, since the match is unindexed. Says nothing at all about whether the pattern occurs.noIndexedText— nothing in scope has stored text, so nothing was searched. Contexts ingested before literal search existed sit here until they are re-ingested.filterMatchedNoContexts—metadataFilter, composed with any visibility filter the application applies, admitted no contexts, so the scan had nothing to read.scopeTruncated—metadataFilteradmitted more contexts than one scan can cover, so only a prefix of them was searched. Narrow the filter.partsCapReached— the scan filled its per-call budget of matching pages/sheets, so more matching places exist than it read.partialCoverage— some contexts in scope have stored text and some do not; see GrepKnowledgeBaseResponse.unscannedContextCount.truncatedContent— some in-scope text was cut at ingest, either because one page or sheet exceeded the per-part storage cap or because a document had more pages than the per-document cap, so that content was never searchable.coverageUnknown— the scan itself succeeded, but how much of the scope holds stored text could not be determined, so completeness cannot be asserted either way.