add dockerfile and fix trival bugs (#78)

api/apps/document_app.py

@@ -146,6 +146,21 @@ def list():
         return server_error_response(e)
 
 
+@manager.route('/thumbnails', methods=['GET'])
+@login_required
+def thumbnails():
+    doc_ids = request.args.get("doc_ids").split(",")
+    if not doc_ids:
+        return get_json_result(
+            data=False, retmsg='Lack of "Document ID"', retcode=RetCode.ARGUMENT_ERROR)
+
+    try:
+        docs = DocumentService.get_thumbnails(doc_ids)
+        return get_json_result(data={d["id"]: d["thumbnail"] for d in docs})
+    except Exception as e:
+        return server_error_response(e)
+
+
 @manager.route('/change_status', methods=['POST'])
 @login_required
 @validate_request("doc_id", "status")
@@ -263,6 +278,7 @@ def rename():
 
 
 @manager.route('/get/<doc_id>', methods=['GET'])
+#@login_required
 def get(doc_id):
     try:
         e, doc = DocumentService.get_by_id(doc_id)

api/db/init_data.py

@@ -56,21 +56,21 @@ def init_superuser():
              "api_key": API_KEY})
 
     if not UserService.save(**user_info):
-        print("【ERROR】can't init admin.")
+        print("\033[93m【ERROR】\033[0mcan't init admin.")
         return
     TenantService.insert(**tenant)
     UserTenantService.insert(**usr_tenant)
     TenantLLMService.insert_many(tenant_llm)
-    print("【INFO】Super user initialized. user name: admin, password: admin. Changing the password after logining is strongly recomanded.")
+    print("【INFO】Super user initialized. \033[93muser name: admin, password: admin\033[0m. Changing the password after logining is strongly recomanded.")
 
     chat_mdl = LLMBundle(tenant["id"], LLMType.CHAT, tenant["llm_id"])
     msg = chat_mdl.chat(system="", history=[{"role": "user", "content": "Hello!"}], gen_conf={})
     if msg.find("ERROR: ") == 0:
-        print("【ERROR】: '{}' dosen't work. {}".format(tenant["llm_id"]), msg)
+        print("\33[91m【ERROR】\33[0m: ", "'{}' dosen't work. {}".format(tenant["llm_id"]), msg)
     embd_mdl = LLMBundle(tenant["id"], LLMType.EMBEDDING, tenant["embd_id"])
-    v,c = embd_mdl.encode(["Hello!"])
+    v, c = embd_mdl.encode(["Hello!"])
     if c == 0:
-        print("【ERROR】: '{}' dosen't work...".format(tenant["embd_id"]))
+        print("\33[91m【ERROR】\33[0m:", " '{}' dosen't work!".format(tenant["embd_id"]))
 
 
 def init_llm_factory():
@@ -89,12 +89,13 @@ def init_llm_factory():
             "logo": "",
             "tags": "LLM,TEXT EMBEDDING,SPEECH2TEXT,MODERATION",
             "status": "1",
-        },{
-            "name": "文心一言",
-            "logo": "",
-            "tags": "LLM,TEXT EMBEDDING,SPEECH2TEXT,MODERATION",
-            "status": "1",
         },
+        # {
+        #     "name": "文心一言",
+        #     "logo": "",
+        #     "tags": "LLM,TEXT EMBEDDING,SPEECH2TEXT,MODERATION",
+        #     "status": "1",
+        # },
     ]
     llm_infos = [
         # ---------------------- OpenAI ------------------------
@@ -198,7 +199,7 @@ def init_llm_factory():
             "llm_name": "embedding-2",
             "tags": "TEXT EMBEDDING",
             "max_tokens": 512,
-            "model_type": LLMType.SPEECH2TEXT.value
+            "model_type": LLMType.EMBEDDING.value
         },
     ]
     for info in factory_infos:

api/db/services/document_service.py

@@ -107,4 +107,11 @@ class DocumentService(CommonService):
         docs = cls.model.select(Knowledgebase.tenant_id).join(Knowledgebase, on=(Knowledgebase.id == cls.model.kb_id)).where(cls.model.id == doc_id, Knowledgebase.status==StatusEnum.VALID.value)
         docs = docs.dicts()
         if not docs:return
-        return docs[0]["tenant_id"]
+        return docs[0]["tenant_id"]
+
+
+    @classmethod
+    @DB.connection_context()
+    def get_thumbnails(cls, docids):
+        fields = [cls.model.id, cls.model.thumbnail]
+        return list(cls.model.select(*fields).where(cls.model.id.in_(docids)).dicts())

api/ragflow_server.py

@@ -33,6 +33,15 @@ from api.db.init_data import init_web_data
 from api.versions import get_versions
 
 if __name__ == '__main__':
+    print("""
+    ____                 ______ __               
+   / __ \ ____ _ ____ _ / ____// /____  _      __
+  / /_/ // __ `// __ `// /_   / // __ \| | /| / /
+ / _, _// /_/ // /_/ // __/  / // /_/ /| |/ |/ / 
+/_/ |_| \__,_/ \__, //_/    /_/ \____/ |__/|__/  
+              /____/                             
+
+    """)
     stat_logger.info(
         f'project base: {utils.file_utils.get_project_base_directory()}'
     )

api/settings.py

@@ -45,12 +45,36 @@ REQUEST_MAX_WAIT_SEC = 300
 
 USE_REGISTRY = get_base_config("use_registry")
 
+default_llm = {
+    "通义千问": {
+        "chat_model": "qwen-plus",
+        "embedding_model": "text-embedding-v2",
+        "image2text_model": "qwen-vl-max",
+        "asr_model": "paraformer-realtime-8k-v1",
+    },
+    "OpenAI": {
+        "chat_model": "gpt-3.5-turbo",
+        "embedding_model": "text-embedding-ada-002",
+        "image2text_model": "gpt-4-vision-preview",
+        "asr_model": "whisper-1",
+    },
+    "智普AI": {
+        "chat_model": "glm-3-turbo",
+        "embedding_model": "embedding-2",
+        "image2text_model": "glm-4v",
+        "asr_model": "",
+    },
+}
 LLM = get_base_config("user_default_llm", {})
-LLM_FACTORY=LLM.get("factory", "通义千问")
-CHAT_MDL = LLM.get("chat_model", "qwen-plus")
-EMBEDDING_MDL = LLM.get("embedding_model", "text-embedding-v2")
-ASR_MDL = LLM.get("asr_model", "paraformer-realtime-8k-v1")
-IMAGE2TEXT_MDL = LLM.get("image2text_model", "qwen-vl-max")
+LLM_FACTORY = LLM.get("factory", "通义千问")
+if LLM_FACTORY not in default_llm:
+    print("\33[91m【ERROR】\33[0m:", f"LLM factory {LLM_FACTORY} has not supported yet, switch to '通义千问/QWen' automatically, and please check the API_KEY in service_conf.yaml.")
+    LLM_FACTORY = "通义千问"
+CHAT_MDL = default_llm[LLM_FACTORY]["chat_model"]
+EMBEDDING_MDL = default_llm[LLM_FACTORY]["embedding_model"]
+ASR_MDL = default_llm[LLM_FACTORY]["asr_model"]
+IMAGE2TEXT_MDL = default_llm[LLM_FACTORY]["image2text_model"]
+
 API_KEY = LLM.get("api_key", "infiniflow API Key")
 PARSERS = LLM.get("parsers", "general:General,qa:Q&A,resume:Resume,naive:Naive,table:Table,laws:Laws,manual:Manual,book:Book,paper:Paper,presentation:Presentation,picture:Picture")
 
@@ -72,7 +96,7 @@ RANDOM_INSTANCE_ID = get_base_config(RAG_FLOW_SERVICE_NAME, {}).get("random_inst
 PROXY = get_base_config(RAG_FLOW_SERVICE_NAME, {}).get("proxy")
 PROXY_PROTOCOL = get_base_config(RAG_FLOW_SERVICE_NAME, {}).get("protocol")
 
-DATABASE = decrypt_database_config()
+DATABASE = decrypt_database_config(name="mysql")
 
 # Logger
 LoggerFactory.set_directory(os.path.join(get_project_base_directory(), "logs", "api"))

api/utils/__init__.py

@@ -264,7 +264,7 @@ def decrypt_database_password(password):
     return pwdecrypt_fun(private_key, password)
 
 
-def decrypt_database_config(database=None, passwd_key="passwd", name="database"):
+def decrypt_database_config(database=None, passwd_key="password", name="database"):
     if not database:
         database = get_base_config(name, {})