AI/ragflow
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/infiniflow/ragflow.git synced 2025-08-13 10:09:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix: retrieval debug mode issue. (#6150)

Browse Source 
### What problem does this PR solve?

#6139

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
This commit is contained in:
Kevin Hu 2025-03-17 13:07:13 +08:00 committed by GitHub
parent 37f3486483
commit bfa8d342b3
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
4 changed files with 4 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
agent/component/base.py
View File

@ -521,7 +521,7 @@ class ComponentBase(ABC):
if u.lower().find("answer") >= 0: if u.lower().find("answer") >= 0:
for r, c in self._canvas.history[::-1]: for r, c in self._canvas.history[::-1]:
if r == "user": if r == "user":
upstream_outs.append(pd.DataFrame([{"content": f"USER:{c}", "component_id": u}])) upstream_outs.append(pd.DataFrame([{"content": c, "component_id": u}]))
break break
break break
if self.component_name.lower().find("answer") >= 0 and self.get_component_name(u) in ["relevant"]: if self.component_name.lower().find("answer") >= 0 and self.get_component_name(u) in ["relevant"]:

3
agent/component/retrieval.py
View File

@ -57,8 +57,7 @@ class Retrieval(ComponentBase, ABC):
query = self.get_input() query = self.get_input()
query = str(query["content"][0]) if "content" in query else "" query = str(query["content"][0]) if "content" in query else ""
lines = query.split('\n') lines = query.split('\n')
user_queries = [line.split("USER:", 1)[1] for line in lines if line.startswith("USER:")] query = lines[-1] if lines else ""
query = user_queries[-1] if user_queries else ""
kbs = KnowledgebaseService.get_by_ids(self._param.kb_ids) kbs = KnowledgebaseService.get_by_ids(self._param.kb_ids)
if not kbs: if not kbs:
return Retrieval.be_output("") return Retrieval.be_output("")

1
deepdoc/parser/pdf_parser.py
View File

@ -41,6 +41,7 @@ LOCK_KEY_pdfplumber = "global_shared_lock_pdfplumber"
if LOCK_KEY_pdfplumber not in sys.modules: if LOCK_KEY_pdfplumber not in sys.modules:
sys.modules[LOCK_KEY_pdfplumber] = threading.Lock() sys.modules[LOCK_KEY_pdfplumber] = threading.Lock()
class RAGFlowPdfParser: class RAGFlowPdfParser:
def __init__(self, parallel_devices: int | None = None): def __init__(self, parallel_devices: int | None = None):
""" """

2
rag/app/naive.py
View File

@ -197,7 +197,7 @@ class Markdown(MarkdownParser):
def chunk(filename, binary=None, from_page=0, to_page=100000, def chunk(filename, binary=None, from_page=0, to_page=100000,
lang="Chinese", parallel_devices=None, callback=None, **kwargs): lang="Chinese", callback=None, parallel_devices=None, **kwargs):
""" """
Supported file formats are docx, pdf, excel, txt. Supported file formats are docx, pdf, excel, txt.
This method apply the naive ways to chunk files. This method apply the naive ways to chunk files.