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

Feat:Vision Model Image Enhancement in Manual/Paper/Book/One chunker (#10640)

Browse Source 
### What problem does this PR solve?
issue:
[#7472](https://github.com/infiniflow/ragflow/issues/7472)
change:
Vision Model Image Enhancement in Manual chunker
### Type of change

- [x] New Feature (non-breaking change which adds functionality)
This commit is contained in:
buua436
2025-10-21 09:36:27 +08:00
committed by GitHub
parent aaa4776657
commit 6ab96287c9
6 changed files with 71 additions and 46 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
rag/app/one.py
View File

@ -23,6 +23,7 @@ from deepdoc.parser.utils import get_text
from rag.app import naive
from rag.nlp import rag_tokenizer, tokenize
from deepdoc.parser import PdfParser, ExcelParser, PlainParser, HtmlParser
from deepdoc.parser.figure_parser import vision_figure_parser_pdf_wrapper,vision_figure_parser_docx_wrapper
class Pdf(PdfParser):
@ -57,13 +58,8 @@ class Pdf(PdfParser):
sections = [(b["text"], self.get_position(b, zoomin))
for i, b in enumerate(self.boxes)]
for (img, rows), poss in tbls:
if not rows:
continue
sections.append((rows if isinstance(rows, str) else rows[0],
[(p[0] + 1 - from_page, p[1], p[2], p[3], p[4]) for p in poss]))
return [(txt, "") for txt, _ in sorted(sections, key=lambda x: (
x[-1][0][0], x[-1][0][3], x[-1][0][1]))], None
x[-1][0][0], x[-1][0][3], x[-1][0][1]))], tbls
def chunk(filename, binary=None, from_page=0, to_page=100000,
@ -80,6 +76,7 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
if re.search(r"\.docx$", filename, re.IGNORECASE):
callback(0.1, "Start to parse.")
sections, tbls = naive.Docx()(filename, binary)
tbls=vision_figure_parser_docx_wrapper(sections=sections,tbls=tbls,callback=callback,**kwargs)
sections = [s for s, _ in sections if s]
for (_, html), _ in tbls:
sections.append(html)
@ -89,8 +86,14 @@ def chunk(filename, binary=None, from_page=0, to_page=100000,
pdf_parser = Pdf()
if parser_config.get("layout_recognize", "DeepDOC") == "Plain Text":
pdf_parser = PlainParser()
sections, _ = pdf_parser(
sections, tbls = pdf_parser(
filename if not binary else binary, to_page=to_page, callback=callback)
tbls=vision_figure_parser_pdf_wrapper(tbls=tbls,callback=callback,**kwargs)
for (img, rows), poss in tbls:
if not rows:
continue
sections.append((rows if isinstance(rows, str) else rows[0],
[(p[0] + 1 - from_page, p[1], p[2], p[3], p[4]) for p in poss]))
sections = [s for s, _ in sections if s]
elif re.search(r"\.xlsx?$", filename, re.IGNORECASE):