()
| 102 | return cells |
| 103 | |
| 104 | def _process_page(): |
| 105 | page_ix = page_no - 1 |
| 106 | page = doc_result.pages[page_ix] |
| 107 | |
| 108 | page_cells = _process_page_cells(page=page) |
| 109 | page_segments = _process_page_segments(doc_items=doc_items, page=page) |
| 110 | content_md = doc.export_to_markdown( |
| 111 | main_text_start=start_ix, main_text_stop=end_ix |
| 112 | ) |
| 113 | # No page-tagging since we only do 1 page at the time |
| 114 | content_dt = doc.export_to_document_tokens( |
| 115 | main_text_start=start_ix, main_text_stop=end_ix, add_page_index=False |
| 116 | ) |
| 117 | |
| 118 | return content_text, content_md, content_dt, page_cells, page_segments, page |
| 119 | |
| 120 | if doc.main_text is None: |
| 121 | return |
no test coverage detected