pengjunjie/ragflow
1
0
Fork 1
mirror of https://github.com/infiniflow/ragflow.git synced 2025-12-08 20:42:30 +08:00
Code Issues Packages Projects Releases Wiki Activity

refine for English corpus (#135)

Browse Source
This commit is contained in:
KevinHuSh
2024-03-20 16:56:16 +08:00
committed by GitHub
parent 78727c8809
commit 6999598101
12 changed files with 216 additions and 125 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
rag/app/table.py
View File

@ -74,9 +74,9 @@ def trans_datatime(s):
def trans_bool(s):
if re.match(r"(true|yes|是|\*|✓|✔|☑|✅|√)$", str(s).strip(), flags=re.IGNORECASE):
return ["yes", "是"]
return "yes"
if re.match(r"(false|no|否|⍻|×)$", str(s).strip(), flags=re.IGNORECASE):
return ["no", "否"]
return "no"
def column_data_type(arr):
@ -92,7 +92,7 @@ def column_data_type(arr):
counts["int"] += 1
elif re.match(r"[+-]?[0-9.]+$", str(a).replace("%%", "")):
counts["float"] += 1
elif re.match(r"(true|false|yes|no|是|否)$", str(a), flags=re.IGNORECASE):
elif re.match(r"(true|yes|是|\*|✓|✔|☑|✅|√|false|no|否|⍻|×)$", str(a), flags=re.IGNORECASE):
counts["bool"] += 1
elif trans_datatime(str(a)):
counts["datetime"] += 1