MindWorkAI/AI-Studio
MindWorkAI/AI-Studio
1
0
Fork 0
mirror of https://github.com/MindWorkAI/AI-Studio.git synced 2026-08-24 20:52:11 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

disabled images for RAG for now

Browse Source
This commit is contained in:
Paul Koudelka 2026-08-12 11:43:47 +02:00
parent c92e83ee59
commit d95487aa6e
1 changed files with 6 additions and 38 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

44
app/MindWork AI Studio/Tools/Services/DataSourceEmbeddingService.Files.cs
View File

@ -15,6 +15,7 @@ public sealed partial class DataSourceEmbeddingService
{ {
private const string OFFICE_LOCK_FILE_PREFIX = "~$"; private const string OFFICE_LOCK_FILE_PREFIX = "~$";
private const int DEFAULT_CHUNK_OVERLAP_TOKEN_LENGTH = 300; private const int DEFAULT_CHUNK_OVERLAP_TOKEN_LENGTH = 300;
private const bool IMAGE_EMBEDDING_ENABLED = false;
private static readonly string[] RAG_DELIMITED_TABLE_FILE_EXTENSIONS = ["csv", "tsv"]; private static readonly string[] RAG_DELIMITED_TABLE_FILE_EXTENSIONS = ["csv", "tsv"];
private static readonly string[] RAG_SPREADSHEET_FILE_EXTENSIONS = ["ods", "xlsm", "xlsb"]; private static readonly string[] RAG_SPREADSHEET_FILE_EXTENSIONS = ["ods", "xlsm", "xlsb"];
@ -46,17 +47,7 @@ public sealed partial class DataSourceEmbeddingService
{ {
var options = this.GetChunkingOptions(dataSource, embeddingProvider); var options = this.GetChunkingOptions(dataSource, embeddingProvider);
var strategy = this.GetChunkingStrategy(filePath); var strategy = this.GetChunkingStrategy(filePath);
ExtractedFileContent content; var content = await this.ReadExtractedFileContentAsync(filePath, embeddingProvider, token);
if (this.IsImageFilePath(filePath))
{
var imageIndexText = this.BuildImageIndexText(filePath);
content = new(imageIndexText, [new(imageIndexText, null)]);
}
else
{
content = await this.ReadExtractedFileContentAsync(filePath, embeddingProvider, token);
}
await foreach (var chunk in this.SplitByChunkingStrategyAsync(content, strategy, options, embeddingProvider, token)) await foreach (var chunk in this.SplitByChunkingStrategyAsync(content, strategy, options, embeddingProvider, token))
yield return chunk; yield return chunk;
@ -567,12 +558,6 @@ public sealed partial class DataSourceEmbeddingService
private ChunkingStrategy GetChunkingStrategy(string filePath) private ChunkingStrategy GetChunkingStrategy(string filePath)
{ {
if (this.IsImageFilePath(filePath))
return new("image", [
new("Whitespace", SplitByWhitespace),
new("Hard cut", null),
]);
if (this.IsPresentationFilePath(filePath)) if (this.IsPresentationFilePath(filePath))
return new("presentation", [ return new("presentation", [
new("Slide", SplitBySourceSegments, true), new("Slide", SplitBySourceSegments, true),
@ -917,7 +902,7 @@ public sealed partial class DataSourceEmbeddingService
private bool IsSupportedRagFilePath(string filePath) private bool IsSupportedRagFilePath(string filePath)
{ {
var extension = Path.GetExtension(filePath).TrimStart('.'); var extension = Path.GetExtension(filePath).TrimStart('.');
return FileTypes.IsAllowedPath(filePath, FileTypes.DOCUMENT, FileTypes.IMAGE) return FileTypes.IsAllowedPath(filePath, FileTypes.DOCUMENT)
|| RAG_DELIMITED_TABLE_FILE_EXTENSIONS.Contains(extension, StringComparer.OrdinalIgnoreCase) || RAG_DELIMITED_TABLE_FILE_EXTENSIONS.Contains(extension, StringComparer.OrdinalIgnoreCase)
|| RAG_SPREADSHEET_FILE_EXTENSIONS.Contains(extension, StringComparer.OrdinalIgnoreCase) || RAG_SPREADSHEET_FILE_EXTENSIONS.Contains(extension, StringComparer.OrdinalIgnoreCase)
|| RAG_SPREADSHEET_ADD_IN_FILE_EXTENSIONS.Contains(extension, StringComparer.OrdinalIgnoreCase); || RAG_SPREADSHEET_ADD_IN_FILE_EXTENSIONS.Contains(extension, StringComparer.OrdinalIgnoreCase);
@ -928,6 +913,9 @@ public sealed partial class DataSourceEmbeddingService
if (this.IsSkippedRagFile(file)) if (this.IsSkippedRagFile(file))
return RagFileIndexingDecision.EXCLUDED; return RagFileIndexingDecision.EXCLUDED;
if (!IMAGE_EMBEDDING_ENABLED && this.IsImageFilePath(file.FullName))
return RagFileIndexingDecision.EXCLUDED;
return this.IsSupportedRagFilePath(file.FullName) return this.IsSupportedRagFilePath(file.FullName)
? RagFileIndexingDecision.INDEXABLE ? RagFileIndexingDecision.INDEXABLE
: RagFileIndexingDecision.UNSUPPORTED; : RagFileIndexingDecision.UNSUPPORTED;
@ -975,26 +963,6 @@ public sealed partial class DataSourceEmbeddingService
} }
} }
private string BuildImageIndexText(string filePath)
{
var fileName = Path.GetFileName(filePath);
var fileNameWithoutExtension = Path.GetFileNameWithoutExtension(filePath);
var extension = Path.GetExtension(filePath).TrimStart('.');
var normalizedName = fileNameWithoutExtension
.Replace('_', ' ')
.Replace('-', ' ')
.Trim();
return $$"""
Image asset
File name: {{fileName}}
Type: {{extension}}
Search terms: {{normalizedName}}
Path: {{filePath}}
Note: The current RAG embedding pipeline stores image files by metadata only. Visual content is not OCRed or captioned yet.
""";
}
private string BuildEmbeddingSignature(IDataSource dataSource, EmbeddingProvider embeddingProvider, ChunkingOptions chunkingOptions) private string BuildEmbeddingSignature(IDataSource dataSource, EmbeddingProvider embeddingProvider, ChunkingOptions chunkingOptions)
{ {
return string.Join('|', return string.Join('|',