Synchronize with enterprise version (#4325)

### Type of change

- [x] Refactoring

agent/templates/customer_service.json

@@ -336,7 +336,7 @@
                 "parameters": [],
                 "presencePenaltyEnabled": true,
                 "presence_penalty": 0.4,
-                "prompt": "Role: You are a customer support.  \n\nTask: Please answer the question based on content of knowledge base. \n\nReuirements & restrictions:\n  -  DO NOT make things up when all knowledge base content is irrelevant to the question. \n  - Answers need to consider chat history.\n  - Request about customer's contact information like, Wechat number, LINE number, twitter, discord, etc,. , when knowlegebase content can't answer his question. So,  product expert could contact him soon to solve his problem.\n\n      Knowledge base content is as following:\n      {input}\n      The above is the content of knowledge base.",
+                "prompt": "Role: You are a customer support.  \n\nTask: Please answer the question based on content of knowledge base. \n\nRequirements & restrictions:\n  -  DO NOT make things up when all knowledge base content is irrelevant to the question. \n  - Answers need to consider chat history.\n  - Request about customer's contact information like, Wechat number, LINE number, twitter, discord, etc,. , when knowledge base content can't answer his question. So,  product expert could contact him soon to solve his problem.\n\n      Knowledge base content is as following:\n      {input}\n      The above is the content of knowledge base.",
                 "temperature": 0.1,
                 "temperatureEnabled": true,
                 "topPEnabled": true,
@@ -603,7 +603,7 @@
           {
             "data": {
               "form": {
-                "text": "Static messages.\nDefine replys after recieve user's contact information."
+                "text": "Static messages.\nDefine response after receive user's contact information."
               },
               "label": "Note",
               "name": "N: What else?"
@@ -691,7 +691,7 @@
           {
             "data": {
               "form": {
-                "text": "Complete questions by conversation history.\nUser: What's RAGFlow?\nAssistant: RAGFlow is xxx.\nUser: How to deloy it?\n\nRefine it: How to deploy RAGFlow?"
+                "text": "Complete questions by conversation history.\nUser: What's RAGFlow?\nAssistant: RAGFlow is xxx.\nUser: How to deploy it?\n\nRefine it: How to deploy RAGFlow?"
               },
               "label": "Note",
               "name": "N: Refine Question"

rag/app/knowledge_graph.py

@@ -9,7 +9,7 @@ def chunk(filename, binary, tenant_id, from_page=0, to_page=100000,
           lang="Chinese", callback=None, **kwargs):
     parser_config = kwargs.get(
         "parser_config", {
-            "chunk_token_num": 512, "delimiter": "\n!?。；！？", "layout_recognize": True})
+            "chunk_token_num": 512, "delimiter": "\n!?;。；！？", "layout_recognize": True})
     eng = lang.lower() == "english"
 
     parser_config["layout_recognize"] = True
@@ -29,4 +29,4 @@ def chunk(filename, binary, tenant_id, from_page=0, to_page=100000,
     doc["title_sm_tks"] = rag_tokenizer.fine_grained_tokenize(doc["title_tks"])
     chunks.extend(tokenize_chunks(sections, doc, eng))
 
-    return chunks
+    return chunks

rag/app/manual.py

@@ -256,7 +256,7 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
         res.extend(tokenize_chunks(chunks, doc, eng, pdf_parser))
         return res
 
-    elif re.search(r"\.docx$", filename, re.IGNORECASE):
+    elif re.search(r"\.docx?$", filename, re.IGNORECASE):
         docx_parser = Docx()
         ti_list, tbls = docx_parser(filename, binary,
                                     from_page=0, to_page=10000, callback=callback)
@@ -279,4 +279,4 @@ if __name__ == "__main__":
         pass
 
 
-    chunk(sys.argv[1], callback=dummy)
+    chunk(sys.argv[1], callback=dummy)

rag/llm/chat_model.py

@@ -24,7 +24,6 @@ import openai
 from ollama import Client
 from rag.nlp import is_chinese, is_english
 from rag.utils import num_tokens_from_string
-from groq import Groq
 import os
 import json
 import requests
@@ -840,6 +839,7 @@ class GeminiChat(Base):
 
 class GroqChat:
     def __init__(self, key, model_name, base_url=''):
+        from groq import Groq
         self.client = Groq(api_key=key)
         self.model_name = model_name
 

rag/llm/tts_model.py

@@ -299,8 +299,6 @@ class SparkTTS:
             yield audio_chunk
 
 
-
-
 class XinferenceTTS:
     def __init__(self, key, model_name, **kwargs):
         self.base_url = kwargs.get("base_url", None)
@@ -330,3 +328,30 @@ class XinferenceTTS:
         for chunk in response.iter_content(chunk_size=1024):
             if chunk:
                 yield chunk
+
+
+class OllamaTTS(Base):
+    def __init__(self, key, model_name="ollama-tts", base_url="https://api.ollama.ai/v1"):
+        if not base_url: 
+            base_url = "https://api.ollama.ai/v1"
+        self.model_name = model_name
+        self.base_url = base_url
+        self.headers = {
+            "Content-Type": "application/json"
+        }
+
+    def tts(self, text, voice="standard-voice"):
+        payload = {
+            "model": self.model_name,
+            "voice": voice,
+            "input": text
+        }
+
+        response = requests.post(f"{self.base_url}/audio/tts", headers=self.headers, json=payload, stream=True)
+
+        if response.status_code != 200:
+            raise Exception(f"**Error**: {response.status_code}, {response.text}")
+
+        for chunk in response.iter_content():
+            if chunk:
+                yield chunk

rag/svr/cache_file_svr.py

@@ -1,60 +1,60 @@
-#
+#
-#  Copyright 2024 The InfiniFlow Authors. All Rights Reserved.
+#  Copyright 2024 The InfiniFlow Authors. All Rights Reserved.
-#
+#
-#  Licensed under the Apache License, Version 2.0 (the "License");
+#  Licensed under the Apache License, Version 2.0 (the "License");
-#  you may not use this file except in compliance with the License.
+#  you may not use this file except in compliance with the License.
-#  You may obtain a copy of the License at
+#  You may obtain a copy of the License at
-#
+#
-#      http://www.apache.org/licenses/LICENSE-2.0
+#      http://www.apache.org/licenses/LICENSE-2.0
-#
+#
-#  Unless required by applicable law or agreed to in writing, software
+#  Unless required by applicable law or agreed to in writing, software
-#  distributed under the License is distributed on an "AS IS" BASIS,
+#  distributed under the License is distributed on an "AS IS" BASIS,
-#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#  See the License for the specific language governing permissions and
+#  See the License for the specific language governing permissions and
-#  limitations under the License.
+#  limitations under the License.
-#
+#
-import logging
+import logging
-import time
+import time
-import traceback
+import traceback
-
+
-from api.db.db_models import close_connection
+from api.db.db_models import close_connection
-from api.db.services.task_service import TaskService
+from api.db.services.task_service import TaskService
-from rag.utils.storage_factory import STORAGE_IMPL
+from rag.utils.minio_conn import MINIOs
-from rag.utils.redis_conn import REDIS_CONN
+from rag.utils.redis_conn import REDIS_CONN
-
+
-
+
-def collect():
+def collect():
-    doc_locations = TaskService.get_ongoing_doc_name()
+    doc_locations = TaskService.get_ongoing_doc_name()
-    logging.debug(doc_locations)
+    logging.debug(doc_locations)
-    if len(doc_locations) == 0:
+    if len(doc_locations) == 0:
-        time.sleep(1)
+        time.sleep(1)
-        return
+        return
-    return doc_locations
+    return doc_locations
-
+
-def main():
+
-    locations = collect()
+def main():
-    if not locations:
+    locations = collect()
-        return
+    if not locations:
-    logging.info(f"TASKS: {len(locations)}")
+        return
-    for kb_id, loc in locations:
+    logging.info(f"TASKS: {len(locations)}")
-        try:
+    for kb_id, loc in locations:
-            if REDIS_CONN.is_alive():
+        try:
-                try:
+            if REDIS_CONN.is_alive():
-                    key = "{}/{}".format(kb_id, loc)
+                try:
-                    if REDIS_CONN.exist(key):
+                    key = "{}/{}".format(kb_id, loc)
-                        continue
+                    if REDIS_CONN.exist(key):
-                    file_bin = STORAGE_IMPL.get(kb_id, loc)
+                        continue
-                    REDIS_CONN.transaction(key, file_bin, 12 * 60)
+                    file_bin = MINIOs.get(kb_id, loc)
-                    logging.info("CACHE: {}".format(loc))
+                    REDIS_CONN.transaction(key, file_bin, 12 * 60)
-                except Exception as e:
+                    logging.info("CACHE: {}".format(loc))
-                    traceback.print_stack(e)
+                except Exception as e:
-        except Exception as e:
+                    traceback.print_stack(e)
-            traceback.print_stack(e)
+        except Exception as e:
-
+            traceback.print_stack(e)
-
+
-
+
-if __name__ == "__main__":
+if __name__ == "__main__":
-    while True:
+    while True:
-        main()
+        main()
-        close_connection()
+        close_connection()
         time.sleep(1)