Fix compilation workflow (#16819)

Dataset_nav can not support large number of documents, introducing AHC clustering as well as retrieval engine as the clustering database
2026-07-11 14:15:40 +08:00 · 2026-07-10 21:11:19 +08:00
parent ea9b3789ce
commit 30739b7f8e
2 changed files with 767 additions and 255 deletions
--- a/conf/infinity_mapping.json
+++ b/conf/infinity_mapping.json
@@ -72,6 +72,7 @@
 	"superseded_by_chunk_id": {"type": "varchar", "default": ""},
 	"doc_count_int": {"type": "integer", "default": 0},
 	"depth_int": {"type": "integer", "default": 0},
+	"parent_kwd": {"type": "varchar", "default": "", "analyzer": "whitespace-#"},
 	"outlinks_int": {"type": "integer", "default": 0},
 	"token_num": {"type": "integer", "default": 0}
 }
--- a/rag/advanced_rag/knowlege_compile/dataset_nav.py
+++ b/rag/advanced_rag/knowlege_compile/dataset_nav.py