mirrors/docling
1
0
Fork 0
mirror of https://github.com/DS4SD/docling.git synced 2025-07-25 19:44:34 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix/implementing the capture of pptx_image with the same method from docx backend by extracting the drawing blip

Browse Source 
Signed-off-by: Benichou <fbenichou@deloitte.ca>
This commit is contained in:
Benichou 2025-04-08 00:53:24 -04:00
parent 95e49705e8
commit b0553e8812
1 changed files with 23 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
docling/backend/mspowerpoint_backend.py
View File

@ -2,6 +2,8 @@ import logging
from io import BytesIO
from pathlib import Path
from typing import Set, Union
from lxml import etree
from lxml.etree import XPath
from docling_core.types.doc import (
BoundingBox,
@ -278,20 +280,31 @@ class MsPowerpointDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentB
)
return
def handle_pictures(self, shape, parent_slide, slide_ind, doc, slide_size):
def handle_pictures(self, shape, parent_slide, slide_ind, doc, slide_size, drawing_blip, slide):
def get_pptx_image(drawing_blip):
rId = drawing_blip[0].get(
"{http://schemas.openxmlformats.org/officeDocument/2006/relationships}embed"
)
rel = slide.part.rels.get(rId)
image_part = rel.target_part
image_data = image_part.blob
return image_data
# Open it with PIL
try:
# Get the image bytes
image = shape.image
image_bytes = image.blob
im_dpi, _ = image.dpi
pil_image = Image.open(BytesIO(image_bytes))
# Open it with PIL
image_data = get_pptx_image(drawing_blip)
image_bytes = BytesIO(image_data)
pil_image = Image.open(image_bytes)
im_dpi, _ = pil_image.info.get("dpi", (72, 72))
# shape has picture
prov = self.generate_prov(shape, slide_ind, "", slide_size)
doc.add_picture(
parent=parent_slide,
image=ImageRef.from_pil(image=pil_image, dpi=im_dpi),
image=ImageRef.from_pil(image=pil_image, dpi=int(im_dpi)),
caption=None,
prov=prov,
)
@ -392,10 +405,9 @@ class MsPowerpointDocumentBackend(DeclarativeDocumentBackend, PaginatedDocumentB
self.handle_tables(shape, parent_slide, slide_ind, doc, slide_size)
if shape.shape_type == MSO_SHAPE_TYPE.PICTURE:
# Handle Pictures
if hasattr(shape, "image"):
self.handle_pictures(
shape, parent_slide, slide_ind, doc, slide_size
)
drawing_blip = self.xpath_expr(shape.element)
if drawing_blip:
self.handle_pictures(shape, parent_slide, slide_ind, doc, drawing_blip, slide)
# If shape doesn't have any text, move on to the next shape
if not hasattr(shape, "text"):
return