fix bug about reload knowledgebase configuration reloading (#210)

### What problem does this PR solve?

_Briefly describe what this PR aims to solve. Include background context
that will help reviewers understand the purpose of the PR._

Issue link:#[[Link the issue
here](https://github.com/infiniflow/ragflow/issues/209)]

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

api/apps/chunk_app.py

@@ -79,7 +79,7 @@ def list():
         return get_json_result(data=res)
     except Exception as e:
         if str(e).find("not_found") > 0:
-            return get_json_result(data=False, retmsg=f'Index not found!',
+            return get_json_result(data=False, retmsg=f'No chunk found!',
                                    retcode=RetCode.DATA_ERROR)
         return server_error_response(e)
 
@@ -262,6 +262,6 @@ def retrieval_test():
         return get_json_result(data=ranks)
     except Exception as e:
         if str(e).find("not_found") > 0:
-            return get_json_result(data=False, retmsg=f'Index not found!',
+            return get_json_result(data=False, retmsg=f'No chunk found! Check the chunk status please!',
                                    retcode=RetCode.DATA_ERROR)
         return server_error_response(e)

api/db/services/knowledgebase_service.py

@@ -44,7 +44,8 @@ class KnowledgebaseService(CommonService):
     def get_detail(cls, kb_id):
         fields = [
             cls.model.id,
-            Tenant.embd_id,
+            #Tenant.embd_id,
+            cls.embd_id,
             cls.model.avatar,
             cls.model.name,
             cls.model.language,

rag/app/manual.py

@@ -125,7 +125,7 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
     for txt, sec_id, poss in sorted(sections, key=lambda x: (
             x[-1][0][0], x[-1][0][3], x[-1][0][1])):
         poss = "\t".join([tag(*pos) for pos in poss])
-        if tk_cnt < 2048 and (sec_id == last_sid or sec_id == -1):
+        if tk_cnt < 32 or (tk_cnt < 1024 and (sec_id == last_sid or sec_id == -1)):
             if chunks:
                 chunks[-1] += "\n" + txt + poss
                 tk_cnt += num_tokens_from_string(txt)
@@ -143,6 +143,9 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
 if __name__ == "__main__":
     import sys
 
+
     def dummy(prog=None, msg=""):
         pass
+
+
     chunk(sys.argv[1], callback=dummy)

rag/app/qa.py

@@ -133,7 +133,7 @@ def chunk(filename, binary=None, lang="Chinese", callback=None, **kwargs):
         return res
 
     raise NotImplementedError(
-        "file type not supported yet(pptx, pdf supported)")
+        "Excel and csv(txt) format files are supported.")
 
 
 if __name__ == "__main__":

rag/nlp/query.py

@@ -73,7 +73,7 @@ class EsQueryer:
             return True
 
         qs, keywords = [], []
-        for tt in self.tw.split(txt):  # .split(" "):
+        for tt in self.tw.split(txt)[:256]:  # .split(" "):
             if not tt:
                 continue
             twts = self.tw.weights([tt])