Bound song evidence vector chunks

This commit is contained in:
Meta Vens 2026-05-31 03:55:51 +01:00
commit 7d5ae1368e
3 changed files with 28 additions and 2 deletions

View file

@ -1229,7 +1229,7 @@ static IEnumerable<EvidenceChunk> BuildEvidenceChunks(IReadOnlyList<IpaTrialResu
content_type {document.ContentType}
content_hash {document.ContentHash}
source {document.SourcePath}
{document.Content}
{EvidenceIndexText(document)}
""",
store);
}
@ -1237,6 +1237,27 @@ static IEnumerable<EvidenceChunk> BuildEvidenceChunks(IReadOnlyList<IpaTrialResu
}
}
static string EvidenceIndexText(SongChallengeEvidenceDocument document)
{
const int maxIndexedChars = 32768;
if (document.Content.Length <= maxIndexedChars)
{
return document.Content;
}
var builder = new StringBuilder();
builder.AppendLine($"indexed_excerpt_only true");
builder.AppendLine($"full_content_hash {document.ContentHash}");
builder.AppendLine($"full_content_chars {document.Content.Length.ToString(CultureInfo.InvariantCulture)}");
if (document.Kind.Contains("spectrogram", StringComparison.OrdinalIgnoreCase))
{
builder.AppendLine("spectrogram evidence is stored in full in CultCache; vector index carries a bounded CSV excerpt.");
}
builder.Append(document.Content.AsSpan(0, maxIndexedChars));
return builder.ToString();
}
static string[] EvidenceTags(IpaTrialResult result, SearchDocument document)
{
var tags = new List<string>();

View file

@ -371,7 +371,9 @@ Return a short final summary naming the patch and report.
"--store",
$storePath,
"--output",
(Join-Path $passDir "vector-index-after.md")
(Join-Path $passDir "vector-index-after.md"),
"--timeout-seconds",
"300"
) `
-LogPath (Join-Path $logRoot "$passId-index-after.log")