Spaces:
Sleeping
Sleeping
| import gradio as gr | |
| from docx import Document | |
| import os | |
| from docx import Document | |
| from docx.oxml import CT_P | |
| def split_by_headers(file_path, headers_per_chunk=1): | |
| doc = Document(file_path) | |
| chunks = [] | |
| current_chunk = Document() | |
| header_count = 0 | |
| for element in doc.element.body: | |
| if isinstance(element, CT_P): | |
| paragraph = element | |
| if any(style.val.startswith('Heading') for style in paragraph.xpath('.//w:pStyle')): | |
| header_count += 1 | |
| if header_count > headers_per_chunk: | |
| chunks.append(current_chunk) | |
| current_chunk = Document() | |
| header_count = 1 | |
| current_chunk.element.body.append(element) | |
| if len(current_chunk.element.body): | |
| chunks.append(current_chunk) | |
| return chunks | |
| def split_by_pages(file_path, pages_per_chunk=1): | |
| doc = Document(file_path) | |
| chunks = [] | |
| current_chunk = Document() | |
| page_count = 0 | |
| estimated_chars_per_page = 3000 # This is an estimation | |
| char_count = 0 | |
| for element in doc.element.body: | |
| if element.tag.endswith('p'): | |
| text = element.text | |
| char_count += len(text) | |
| if char_count >= estimated_chars_per_page: | |
| page_count += 1 | |
| char_count = 0 | |
| if page_count >= pages_per_chunk: | |
| chunks.append(current_chunk) | |
| current_chunk = Document() | |
| page_count = 0 | |
| current_chunk.element.body.append(element) | |
| if len(current_chunk.element.body): | |
| chunks.append(current_chunk) | |
| return chunks | |
| def save_chunks(chunks, original_filename): | |
| saved_files = [] | |
| base_name = os.path.splitext(original_filename)[0] | |
| for i, chunk in enumerate(chunks, 1): | |
| output_path = f"{base_name}_part{i}.docx" | |
| chunk.save(output_path) | |
| saved_files.append(output_path) | |
| return saved_files | |
| def process_document(file, split_type, headers_or_pages): | |
| if headers_or_pages < 1: | |
| return "Por favor, especifique un número positivo de encabezados o páginas por fragmento." | |
| if split_type == "Encabezados": | |
| chunks = split_by_headers(file.name, headers_or_pages) | |
| else: # Páginas | |
| chunks = split_by_pages(file.name, headers_or_pages) | |
| saved_files = save_chunks(chunks, os.path.basename(file.name)) | |
| return f"Documento dividido en {len(saved_files)} partes: {', '.join(saved_files)}" | |
| # Interfaz Gradio | |
| iface = gr.Interface( | |
| fn=process_document, | |
| inputs=[ | |
| gr.File(label="Seleccione el archivo DOCX"), | |
| gr.Radio(["Encabezados", "Páginas"], label="Método de división"), | |
| gr.Number(value=1, label="Número de encabezados/páginas por fragmento", minimum=1) | |
| ], | |
| outputs=gr.Text(label="Resultado"), | |
| title="Divisor de Documentos DOCX", | |
| description="Divida documentos DOCX por encabezados o páginas estimadas" | |
| ) | |
| if __name__ == "__main__": | |
| iface.launch() |