Spaces:
Sleeping
Sleeping
Update ppt_parser.py
Browse files- ppt_parser.py +41 -33
ppt_parser.py
CHANGED
|
@@ -10,6 +10,35 @@ import io
|
|
| 10 |
def print_json(item):
|
| 11 |
return json.dumps(item, ensure_ascii=False, indent=4)
|
| 12 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
def transfer_textbox_content_in_group(group_shape):
|
| 14 |
group_shape_item = {}
|
| 15 |
for l, shape in enumerate(group_shape.shapes):
|
|
@@ -19,21 +48,9 @@ def transfer_textbox_content_in_group(group_shape):
|
|
| 19 |
shape_item['location'] = (shape.left, shape.top)
|
| 20 |
text_frame = shape.text_frame
|
| 21 |
for r, paragraph in enumerate(text_frame.paragraphs):
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
'text': paragraph.text,
|
| 26 |
-
'align': paragraph.alignment,
|
| 27 |
-
'font': {
|
| 28 |
-
'name': original_run.font.name,
|
| 29 |
-
'bold': original_run.font.bold,
|
| 30 |
-
'italic': original_run.font.italic,
|
| 31 |
-
'underline': original_run.font.underline,
|
| 32 |
-
'color': str(original_run.font.color.rgb),
|
| 33 |
-
'language_id': original_run.font.language_id,
|
| 34 |
-
}
|
| 35 |
-
}
|
| 36 |
-
shape_item[f'paragraph_{r}'] = paragraph_item
|
| 37 |
group_shape_item[f"shape_{l}"] = shape_item
|
| 38 |
return group_shape_item
|
| 39 |
|
|
@@ -53,21 +70,9 @@ def transfer_to_structure(pptx_file, images_dir_path):
|
|
| 53 |
shape_item['type'] = "text"
|
| 54 |
text_frame = shape.text_frame
|
| 55 |
for r, paragraph in enumerate(text_frame.paragraphs):
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
'text': paragraph.text,
|
| 60 |
-
'align': paragraph.alignment,
|
| 61 |
-
'font': {
|
| 62 |
-
'name': original_run.font.name,
|
| 63 |
-
'bold': original_run.font.bold,
|
| 64 |
-
'italic': original_run.font.italic,
|
| 65 |
-
'underline': original_run.font.underline,
|
| 66 |
-
'color': str(original_run.font.color.rgb),
|
| 67 |
-
'language_id': original_run.font.language_id,
|
| 68 |
-
}
|
| 69 |
-
}
|
| 70 |
-
shape_item[f'paragraph_{r}'] = paragraph_item
|
| 71 |
|
| 72 |
# Case 2: Grouped shapes
|
| 73 |
elif isinstance(shape, GroupShape):
|
|
@@ -84,9 +89,12 @@ def transfer_to_structure(pptx_file, images_dir_path):
|
|
| 84 |
shape_item['dpi'] = shape.image.dpi
|
| 85 |
shape_item['location'] = (shape.left, shape.top)
|
| 86 |
shape_item['location_inches'] = (Inches(shape.left).inches, Inches(shape.top).inches)
|
| 87 |
-
|
| 88 |
-
|
| 89 |
-
|
|
|
|
|
|
|
|
|
|
| 90 |
|
| 91 |
slide_item[f"shape_{j}"] = shape_item
|
| 92 |
|
|
|
|
| 10 |
def print_json(item):
|
| 11 |
return json.dumps(item, ensure_ascii=False, indent=4)
|
| 12 |
|
| 13 |
+
def safe_font_attribute(run, attr):
|
| 14 |
+
try:
|
| 15 |
+
return getattr(run.font, attr)
|
| 16 |
+
except Exception:
|
| 17 |
+
return None
|
| 18 |
+
|
| 19 |
+
def safe_color(run):
|
| 20 |
+
try:
|
| 21 |
+
return str(run.font.color.rgb) if run.font.color and run.font.color.rgb else None
|
| 22 |
+
except Exception:
|
| 23 |
+
return None
|
| 24 |
+
|
| 25 |
+
def extract_paragraph_data(paragraph):
|
| 26 |
+
if not paragraph.runs:
|
| 27 |
+
return None
|
| 28 |
+
run = paragraph.runs[0]
|
| 29 |
+
return {
|
| 30 |
+
'text': paragraph.text,
|
| 31 |
+
'align': paragraph.alignment,
|
| 32 |
+
'font': {
|
| 33 |
+
'name': safe_font_attribute(run, 'name'),
|
| 34 |
+
'bold': safe_font_attribute(run, 'bold'),
|
| 35 |
+
'italic': safe_font_attribute(run, 'italic'),
|
| 36 |
+
'underline': safe_font_attribute(run, 'underline'),
|
| 37 |
+
'color': safe_color(run),
|
| 38 |
+
'language_id': safe_font_attribute(run, 'language_id'),
|
| 39 |
+
}
|
| 40 |
+
}
|
| 41 |
+
|
| 42 |
def transfer_textbox_content_in_group(group_shape):
|
| 43 |
group_shape_item = {}
|
| 44 |
for l, shape in enumerate(group_shape.shapes):
|
|
|
|
| 48 |
shape_item['location'] = (shape.left, shape.top)
|
| 49 |
text_frame = shape.text_frame
|
| 50 |
for r, paragraph in enumerate(text_frame.paragraphs):
|
| 51 |
+
data = extract_paragraph_data(paragraph)
|
| 52 |
+
if data:
|
| 53 |
+
shape_item[f'paragraph_{r}'] = data
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 54 |
group_shape_item[f"shape_{l}"] = shape_item
|
| 55 |
return group_shape_item
|
| 56 |
|
|
|
|
| 70 |
shape_item['type'] = "text"
|
| 71 |
text_frame = shape.text_frame
|
| 72 |
for r, paragraph in enumerate(text_frame.paragraphs):
|
| 73 |
+
data = extract_paragraph_data(paragraph)
|
| 74 |
+
if data:
|
| 75 |
+
shape_item[f'paragraph_{r}'] = data
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 76 |
|
| 77 |
# Case 2: Grouped shapes
|
| 78 |
elif isinstance(shape, GroupShape):
|
|
|
|
| 89 |
shape_item['dpi'] = shape.image.dpi
|
| 90 |
shape_item['location'] = (shape.left, shape.top)
|
| 91 |
shape_item['location_inches'] = (Inches(shape.left).inches, Inches(shape.top).inches)
|
| 92 |
+
try:
|
| 93 |
+
image_stream = io.BytesIO(shape.image.blob)
|
| 94 |
+
shape_image = Image.open(image_stream)
|
| 95 |
+
shape_image.save(image_path)
|
| 96 |
+
except Exception:
|
| 97 |
+
pass # Corrupt or unsupported image
|
| 98 |
|
| 99 |
slide_item[f"shape_{j}"] = shape_item
|
| 100 |
|