ocr_pdf_intermediate_dict_to_markdown_with_para输出nlp格式的markdown

opendatalab · Mar 24, 2024 · 07e4f11 · 07e4f11
1 parent bf8d8e2
commit 07e4f11
Showing 1 changed file with 3 additions and 2 deletions.
diff --git a/magic_pdf/pipeline.py b/magic_pdf/pipeline.py
@@ -7,7 +7,7 @@
     ocr_mk_nlp_markdown,
     ocr_mk_mm_markdown,
     ocr_mk_mm_standard_format,
-    ocr_mk_mm_markdown_with_para, ocr_mk_mm_markdown_with_para_and_pagination,
+    ocr_mk_mm_markdown_with_para, ocr_mk_mm_markdown_with_para_and_pagination, ocr_mk_nlp_markdown_with_para,
 )
 from magic_pdf.libs.commons import (
     read_file,
@@ -510,7 +510,8 @@ def ocr_pdf_intermediate_dict_to_markdown_with_para(jso: dict, debug_mode=False)
         pdf_intermediate_dict = jso["pdf_intermediate_dict"]
         # 将 pdf_intermediate_dict 解压
         pdf_intermediate_dict = JsonCompressor.decompress_json(pdf_intermediate_dict)
-        markdown_content = ocr_mk_mm_markdown_with_para(pdf_intermediate_dict)
+        # markdown_content = ocr_mk_mm_markdown_with_para(pdf_intermediate_dict)
+        markdown_content = ocr_mk_nlp_markdown_with_para(pdf_intermediate_dict)
         jso["content"] = markdown_content
         logger.info(
             f"book_name is:{get_data_source(jso)}/{jso['file_id']},markdown content length is {len(markdown_content)}",