AI/ragflow
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/infiniflow/ragflow.git synced 2025-08-14 01:25:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

let presentation do raptor (#2838)

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

#2837

### Type of change

- [x] New Feature (non-breaking change which adds functionality)
This commit is contained in:
Kevin Hu 2024-10-15 10:11:09 +08:00 committed by GitHub
parent 260d694bbc
commit b540d41cdc
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 13 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
api/apps/document_app.py
View File

@ -439,8 +439,9 @@ def change_parser():
else: else:
return get_json_result(data=True) return get_json_result(data=True)
if doc.type == FileType.VISUAL or re.search( if ((doc.type == FileType.VISUAL and req["parser_id"] != "picture")
r"\.(ppt|pptx|pages)$", doc.name): or (re.search(
r"\.(ppt|pptx|pages)$", doc.name) and req["parser_id"] != "presentation")):
return get_data_error_result(retmsg="Not supported yet!") return get_data_error_result(retmsg="Not supported yet!")
e = DocumentService.update_by_id(doc.id, e = DocumentService.update_by_id(doc.id,

11
rag/app/qa.py
View File

@ -68,6 +68,7 @@ class Excel(ExcelParser):
[rmPrefix(q) for q, _ in random_choices(res, k=30) if len(q) > 1]) [rmPrefix(q) for q, _ in random_choices(res, k=30) if len(q) > 1])
return res return res
class Pdf(PdfParser): class Pdf(PdfParser):
def __call__(self, filename, binary=None, from_page=0, def __call__(self, filename, binary=None, from_page=0,
to_page=100000, zoomin=3, callback=None): to_page=100000, zoomin=3, callback=None):
@ -155,6 +156,7 @@ class Pdf(PdfParser):
if last_q: if last_q:
qai_list.append((last_q, last_a, *self.crop(last_tag, need_position=True))) qai_list.append((last_q, last_a, *self.crop(last_tag, need_position=True)))
return qai_list, tbls return qai_list, tbls
def get_tbls_info(self, tbls, tbl_index): def get_tbls_info(self, tbls, tbl_index):
if tbl_index >= len(tbls): if tbl_index >= len(tbls):
return 1, 0, 0, 0, 0, '@@0\t0\t0\t0\t0##', '' return 1, 0, 0, 0, 0, '@@0\t0\t0\t0\t0##', ''
@ -166,10 +168,13 @@ class Pdf(PdfParser):
tbl_tag = "@@{}\t{:.1f}\t{:.1f}\t{:.1f}\t{:.1f}##" \ tbl_tag = "@@{}\t{:.1f}\t{:.1f}\t{:.1f}\t{:.1f}##" \
.format(tbl_pn, tbl_left, tbl_right, tbl_top, tbl_bottom) .format(tbl_pn, tbl_left, tbl_right, tbl_top, tbl_bottom)
tbl_text = ''.join(tbls[tbl_index][0][1]) tbl_text = ''.join(tbls[tbl_index][0][1])
return tbl_pn, tbl_left, tbl_right, tbl_top, tbl_bottom, tbl_tag, tbl_text return tbl_pn, tbl_left, tbl_right, tbl_top, tbl_bottom, tbl_tag,
class Docx(DocxParser): class Docx(DocxParser):
def __init__(self): def __init__(self):
pass pass
def get_picture(self, document, paragraph): def get_picture(self, document, paragraph):
img = paragraph._element.xpath('.//pic:pic') img = paragraph._element.xpath('.//pic:pic')
if not img: if not img:
@ -242,6 +247,7 @@ class Docx(DocxParser):
tbls.append(((None, html), "")) tbls.append(((None, html), ""))
return qai_list, tbls return qai_list, tbls
def rmPrefix(txt): def rmPrefix(txt):
return re.sub( return re.sub(
r"^(问题|答案|回答|user|assistant|Q|A|Question|Answer|问|答)[\t:： ]+", "", txt.strip(), flags=re.IGNORECASE) r"^(问题|答案|回答|user|assistant|Q|A|Question|Answer|问|答)[\t:： ]+", "", txt.strip(), flags=re.IGNORECASE)
@ -258,6 +264,7 @@ def beAdocPdf(d, q, a, eng, image, poss):
add_positions(d, poss) add_positions(d, poss)
return d return d
def beAdocDocx(d, q, a, eng, image): def beAdocDocx(d, q, a, eng, image):
qprefix = "Question: " if eng else "问题：" qprefix = "Question: " if eng else "问题："
aprefix = "Answer: " if eng else "回答：" aprefix = "Answer: " if eng else "回答："
@ -268,6 +275,7 @@ def beAdocDocx(d, q, a, eng, image):
d["image"] = image d["image"] = image
return d return d
def beAdoc(d, q, a, eng): def beAdoc(d, q, a, eng):
qprefix = "Question: " if eng else "问题：" qprefix = "Question: " if eng else "问题："
aprefix = "Answer: " if eng else "回答：" aprefix = "Answer: " if eng else "回答："
@ -282,6 +290,7 @@ def mdQuestionLevel(s):
match = re.match(r'#*', s) match = re.match(r'#*', s)
return (len(match.group(0)), s.lstrip('#').lstrip()) if match else (0, s) return (len(match.group(0)), s.lstrip('#').lstrip()) if match else (0, s)
def chunk(filename, binary=None, lang="Chinese", callback=None, **kwargs): def chunk(filename, binary=None, lang="Chinese", callback=None, **kwargs):
""" """
Excel and csv(txt) format files are supported. Excel and csv(txt) format files are supported.