zlei6/ragflow
1
0
Fork 0
mirror of https://github.com/infiniflow/ragflow.git synced 2026-07-29 12:09:31 +08:00
Code Issues Packages Projects Releases Wiki Activity

Refactor: reformat all code for lefthook using ruff and gofmt (#16585)

Browse Source
This commit is contained in:
Wang Qi
2026-07-03 12:53:39 +08:00
committed by GitHub
parent 19fcb4a981
commit 6a4b9be426
588 changed files with 11123 additions and 15412 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
deepdoc/parser/docx_parser.py
View File

@@ -29,6 +29,7 @@ from docx.image.exceptions import (
)
from rag.utils.lazy_image import LazyImage
class RAGFlowDocxParser:
def get_picture(self, document, paragraph):
imgs = paragraph._element.xpath(".//pic:pic")
@@ -69,7 +70,6 @@ class RAGFlowDocxParser:
return None
return LazyImage(image_blobs)
def __extract_table_content(self, tb):
df = []
for row in tb.rows:
@@ -91,7 +91,7 @@ class RAGFlowDocxParser:
(r"^[0-9A-Z/\._~-]+$", "Ca"),
(r"^[A-Z]*[a-z' -]+$", "En"),
(r"^[0-9.,+-]+[0-9A-Za-z/$￥%<>（）()' -]+$", "NE"),
(r"^.{1}$", "Sg")
(r"^.{1}$", "Sg"),
]
for p, n in pattern:
if re.search(p, b):
@@ -110,16 +110,14 @@ class RAGFlowDocxParser:
if len(df) < 2:
return []
max_type = Counter([blockType(str(df.iloc[i, j])) for i in range(
1, len(df)) for j in range(len(df.iloc[i, :]))])
max_type = Counter([blockType(str(df.iloc[i, j])) for i in range(1, len(df)) for j in range(len(df.iloc[i, :]))])
max_type = max(max_type.items(), key=lambda x: x[1])[0]
colnm = len(df.iloc[0, :])
hdrows = [0] # header is not necessarily appear in the first line
if max_type == "Nu":
for r in range(1, len(df)):
tys = Counter([blockType(str(df.iloc[r, j]))
for j in range(len(df.iloc[r, :]))])
tys = Counter([blockType(str(df.iloc[r, j])) for j in range(len(df.iloc[r, :]))])
tys = max(tys.items(), key=lambda x: x[1])[0]
if tys != max_type:
hdrows.append(r)
@@ -160,26 +158,25 @@ class RAGFlowDocxParser:
return ["\n".join(lines)]
def __call__(self, fnm, from_page=0, to_page=MAXIMUM_PAGE_NUMBER):
self.doc = Document(fnm) if isinstance(
fnm, str) else Document(BytesIO(fnm))
pn = 0 # parsed page
secs = [] # parsed contents
self.doc = Document(fnm) if isinstance(fnm, str) else Document(BytesIO(fnm))
pn = 0 # parsed page
secs = [] # parsed contents
for p in self.doc.paragraphs:
if pn > to_page:
break
runs_within_single_paragraph = [] # save runs within the range of pages
runs_within_single_paragraph = [] # save runs within the range of pages
for run in p.runs:
if pn > to_page:
break
if from_page <= pn < to_page and p.text.strip():
runs_within_single_paragraph.append(run.text) # append run.text first
runs_within_single_paragraph.append(run.text) # append run.text first
# wrap page break checker into a static method
if 'lastRenderedPageBreak' in run._element.xml:
if "lastRenderedPageBreak" in run._element.xml:
pn += 1
secs.append(("".join(runs_within_single_paragraph), p.style.name if hasattr(p.style, 'name') else '')) # then concat run.text as part of the paragraph
secs.append(("".join(runs_within_single_paragraph), p.style.name if hasattr(p.style, "name") else "")) # then concat run.text as part of the paragraph
tbls = [self.__extract_table_content(tb) for tb in self.doc.tables]
return secs, tbls