pengjunjie/ragflow
1
0
Fork 1
mirror of https://github.com/infiniflow/ragflow.git synced 2025-12-23 06:46:40 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix: model not authorized (#12001)

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

Fix model not authorized. #11973.


### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
This commit is contained in:
Yongteng Lei
2025-12-17 19:48:24 +08:00
committed by GitHub
parent 3820de916c
commit 672958a192
10 changed files with 100 additions and 33 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
rag/app/manual.py
View File

@ -27,6 +27,7 @@ from deepdoc.parser.figure_parser import vision_figure_parser_pdf_wrapper,vision
from docx import Document
from PIL import Image
from rag.app.naive import by_plaintext, PARSERS
from common.parser_config_utils import normalize_layout_recognizer
class Pdf(PdfParser):
def __init__(self):
@ -196,7 +197,9 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
# is it English
eng = lang.lower() == "english" # pdf_parser.is_english
if re.search(r"\.pdf$", filename, re.IGNORECASE):
layout_recognizer = parser_config.get("layout_recognize", "DeepDOC")
layout_recognizer, parser_model_name = normalize_layout_recognizer(
parser_config.get("layout_recognize", "DeepDOC")
)
if isinstance(layout_recognizer, bool):
layout_recognizer = "DeepDOC" if layout_recognizer else "Plain Text"
@ -205,6 +208,8 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
pdf_parser = PARSERS.get(name, by_plaintext)
callback(0.1, "Start to parse.")
kwargs.pop("parse_method", None)
kwargs.pop("mineru_llm_name", None)
sections, tbls, pdf_parser = pdf_parser(
filename = filename,
binary = binary,
@ -214,6 +219,7 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
callback = callback,
pdf_cls = Pdf,
layout_recognizer = layout_recognizer,
mineru_llm_name=parser_model_name,
parse_method = "manual",
**kwargs
)
@ -232,7 +238,7 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
poss = pdf_parser.extract_positions(poss)
if poss:
first = poss[0] # tuple: ([pn], x1, x2, y1, y2)
pn = first[0]
pn = first[0]
if isinstance(pn, list) and pn:
pn = pn[0] # [pn] -> pn
poss[0] = (pn, *first[1:])