shell knowledge commands

src/knowledge/core_object/document_object.py

@@ -49,7 +49,7 @@ class DocumentObjectBuilder:
         self.text = text
         return self
 
-    def build(self, relation_store: ObjectRelationStore) -> DocumentObject:
+    def build(self) -> DocumentObject:
         chunk_list = KnowledgeStore().get_chunk_list_writer().create_chunk_list_from_text(
             self.text,
             1024 * 4,
@@ -60,6 +60,6 @@ class DocumentObjectBuilder:
         
         # Add relation to store
         for chunk_id in chunk_list.chunk_list:
-            relation_store.add_relation(chunk_id, doc_id)
+            KnowledgeStore().get_relation_store().add_relation(chunk_id, doc_id)
             
         return doc

src/knowledge/core_object/email_object.py

@@ -94,7 +94,7 @@ class EmailObjectBuilder:
                 with open(content_file, "r", encoding="utf-8") as f:
                     text = f.read()
 
-                document = DocumentObjectBuilder({}, {}, text).build(relation_store=relation)
+                document = DocumentObjectBuilder({}, {}, text).build()
                 document_id = document.calculate_id()
                 store.put_object(document_id, document.encode())
                 documents = {"email.txt": document_id}

src/knowledge/core_object/image_object.py

@@ -52,7 +52,7 @@ def get_exif_data(image_path: str):
         return {
             TAGS.get(key): exif_data[key]
             for key in exif_data.keys()
-            if key in TAGS and isinstance(exif_data[key], (bytes, str))
+            if key in TAGS and isinstance(exif_data[key], str)
         }
     else:
         return {}

src/knowledge/data/writer.py

@@ -46,7 +46,7 @@ class ChunkListWriter:
                     )
 
         file_hash = HashValue(hash_obj.digest())
-        print(f"calc file hash: {file_path}, {file_hash}")
+        # print(f"calc file hash: {file_path}, {file_hash}")
         
         return ChunkList(chunk_list, file_hash)
 

src/knowledge/store.py

@@ -4,7 +4,7 @@ from .object import ObjectStore, ObjectRelationStore
 from .data import ChunkStore, ChunkTracker, ChunkListWriter, ChunkReader
 from .vector import ChromaVectorStore, VectorBase
 import logging
-
+import aios_kernel
 
 # KnowledgeStore class, which aggregates ChunkStore, ChunkTracker, and ObjectStore, and is a global singleton that makes it easy to use these three built-in store examples
 class KnowledgeStore:
@@ -13,16 +13,12 @@ class KnowledgeStore:
     def __new__(cls):
         if cls._instance is None:
             cls._instance = super().__new__(cls)
-            directory = os.path.join(
-                os.path.dirname(__file__), "../../rootfs/data/"
-            )
-            directory = os.path.normpath(directory)
-            print(directory)
+            knowledge_dir = aios_kernel.storage.AIStorage().get_myai_dir() / "knowledge"
 
-            if not os.path.exists(directory):
-                os.makedirs(directory)
+            if not os.path.exists(knowledge_dir):
+                os.makedirs(knowledge_dir)
 
-            cls._instance.__singleton_init__(directory)
+            cls._instance.__singleton_init__(knowledge_dir)
 
         return cls._instance
 
@@ -64,5 +60,5 @@ class KnowledgeStore:
     
     def get_vector_store(self, model_name: str) -> VectorBase:
         if model_name not in self.vector_store:
-            self.vector_store[model_name] = ChromaVectorStore(model_name)
+            self.vector_store[model_name] = ChromaVectorStore(self.root, model_name)
         return self.vector_store[model_name]

src/knowledge/vector/chroma_store.py

@@ -6,16 +6,14 @@ import os
 
 
 class ChromaVectorStore(VectorBase):
-    def __init__(self, model_name: str) -> None:
+    def __init__(self, root_dir, model_name: str) -> None:
         super().__init__(model_name)
 
         logging.info(
             "will init chroma vector store, model={}".format(model_name)
         )
 
-        directory = os.path.join(
-            os.path.dirname(__file__), "../../../rootfs/data/vector"
-        )
+        directory = os.path.join(root_dir, "vector")
         logging.info("will use vector store: {}".format(directory))
 
         client = chromadb.PersistentClient(