PictureItem.export_to_markdown() defaults to ImageRefMode.EMBEDDED, which embeds base64 image data. When _extract_item_text() called this method during context expansion, base64 data would leak into results. Now explicitly handles PictureItem with ImageRefMode.PLACEHOLDER to prevent base64 while still including VLM descriptions and captions. |
||
|---|---|---|
| .. | ||
| test_chunk | ||
| test_chunker | ||
| test_client | ||
| test_context_enhancement | ||
| test_converters | ||
| test_database_autocreate | ||
| test_embedder | ||
| test_filter | ||
| test_qa | ||
| test_rebuild | ||
| test_reranker | ||
| test_search | ||
| test_versioning | ||