Create app.py

app.py

@@ -0,0 +1,96 @@
+from pptx import Presentation
+from pptx.util import Pt
+from pptx.shapes.group import GroupShape
+from pptx.dml.color import RGBColor
+from pptx.enum.text import PP_ALIGN
+import pptx_ea_font
+from pptx.shapes.picture import Picture
+from pptx.util import Inches
+
+import json
+import os
+from PIL import Image
+import io
+
+def print_json(item):
+    item_json = json.dumps(item, ensure_ascii=False, indent=4)
+    print(item_json)
+
+def transfer_textbox_content_in_group(group_shape):
+    """Edit the content of text boxes within a group shape."""
+    group_shape_item = {}
+    for l, shape in enumerate(group_shape.shapes):
+        shape_item = {}
+        if shape.has_text_frame:
+            shape_item['type'] = "text"
+            shape_item['location'] = (shape.left, shape.top)
+            text_frame = shape.text_frame
+            for r, paragraph in enumerate(text_frame.paragraphs):
+                original_run = paragraph.runs[0]
+                paragraph_item = {}
+                paragraph_item['text'] = paragraph.text
+#                 paragraph_item['location'] = paragraph.left
+                paragraph_item['align'] = paragraph.alignment
+                font_item = {}
+                font_item['name'] = original_run.font.name
+                font_item['bold'] = original_run.font.bold
+                font_item['italic'] = original_run.font.italic
+                font_item['underline'] = original_run.font.underline
+                font_item['color'] = original_run.font.color.rgb
+                font_item['language_id'] = original_run.font.language_id
+                paragraph_item['font'] = font_item
+                shape_item[f'paragraph_{r}'] = paragraph_item
+        group_shape_item[f"shape_{l}"] = shape_item
+    return group_shape_item
+
+def transfer_to_structure(prs, images_dir_path):
+    item = {}
+    
+    # Iterate through each slide in the presentation
+    for i, slide in enumerate(prs.slides):
+        # Iterate through each shape in the slide
+        slide_item = {}
+        for j, shape in enumerate(slide.shapes):
+            # If the shape is a group, process its shapes
+            shape_item = {}
+            if shape.has_text_frame:
+                text_frame = shape.text_frame
+                for paragraph in text_frame.paragraphs:
+                    # Clear the existing text but keep the paragraph
+                    paragraph.clear()
+                    # Add a new run with the new content and copy font style
+                    run = paragraph.add_run()
+                    run.text = new_content
+                    if paragraph.runs:
+                        original_run = paragraph.runs[0]
+                        copy_font_style(original_run, run)
+            elif isinstance(shape, GroupShape):
+                shape_item['type'] = "group"
+                group_shape_item = transfer_textbox_content_in_group(shape)
+                shape_item['group_content'] = group_shape_item
+                pass
+            elif isinstance(shape, Picture):
+                shape_item['type'] = "picture"
+                image_path = os.path.join(images_dir_path, f"picture_{j}.png")
+                shape_item['image_path'] = image_path
+                shape_item['size'] = shape.image.size # width, height
+                shape_item['dpi'] = shape.image.dpi # (horz_dpi, vert_dpi) 
+                shape_item['location'] = (shape.left, shape.top)
+                shape_item['location_inches'] = (Inches(shape.left).inches, Inches(shape.top).inches)
+                image_stream = io.BytesIO(shape.image.blob)
+                shape_image = Image.open(image_stream)
+                shape_image.save(image_path)
+#                 print(shape.image)
+#                 shape.image.save(image_path)
+                pass
+            slide_item[f"shape_{j}"] = shape_item
+        item[f"slide_{i}"] = slide_item
+                
+    print_json(item)
+    
+    
+if __name__ == "__main__":
+    pptx_file_path = "templates_from_gaoding/ppt/1.pptx"
+    images_dir_path = "images"
+    prs = Presentation(pptx_file_path)
+    transfer_to_structure(prs, images_dir_path)