magic-pdf 0.10.4__py3-none-any.whl → 0.10.6__py3-none-any.whl

magic_pdf/tools/common.py

@@ -1,5 +1,3 @@
-import copy
-import json as json_parse
 import os
 
 import click
@@ -7,13 +5,12 @@ import fitz
 from loguru import logger
 
 import magic_pdf.model as model_config
+from magic_pdf.config.enums import SupportedPdfParseMethod
 from magic_pdf.config.make_content_config import DropMode, MakeMode
 from magic_pdf.data.data_reader_writer import FileBasedDataWriter
-from magic_pdf.libs.draw_bbox import (draw_layout_bbox, draw_line_sort_bbox,
-                                      draw_model_bbox, draw_span_bbox)
-from magic_pdf.pipe.OCRPipe import OCRPipe
-from magic_pdf.pipe.TXTPipe import TXTPipe
-from magic_pdf.pipe.UNIPipe import UNIPipe
+from magic_pdf.data.dataset import PymuDocDataset
+from magic_pdf.model.doc_analyze_by_custom_model import doc_analyze
+from magic_pdf.model.operators import InferenceResult
 
 # from io import BytesIO
 # from pypdf import PdfReader, PdfWriter
@@ -56,7 +53,11 @@ def prepare_env(output_dir, pdf_file_name, method):
 def convert_pdf_bytes_to_bytes_by_pymupdf(pdf_bytes, start_page_id=0, end_page_id=None):
     document = fitz.open('pdf', pdf_bytes)
     output_document = fitz.open()
-    end_page_id = end_page_id if end_page_id is not None and end_page_id >= 0 else len(document) - 1
+    end_page_id = (
+        end_page_id
+        if end_page_id is not None and end_page_id >= 0
+        else len(document) - 1
+    )
     if end_page_id > len(document) - 1:
         logger.warning('end_page_id is out of range, use pdf_docs length')
         end_page_id = len(document) - 1
@@ -94,78 +95,126 @@ def do_parse(
         f_draw_model_bbox = True
         f_draw_line_sort_bbox = True
 
-    if lang == "":
+    if lang == '':
         lang = None
 
-    pdf_bytes = convert_pdf_bytes_to_bytes_by_pymupdf(pdf_bytes, start_page_id, end_page_id)
+    pdf_bytes = convert_pdf_bytes_to_bytes_by_pymupdf(
+        pdf_bytes, start_page_id, end_page_id
+    )
 
-    orig_model_list = copy.deepcopy(model_list)
-    local_image_dir, local_md_dir = prepare_env(output_dir, pdf_file_name,
-                                                parse_method)
+    local_image_dir, local_md_dir = prepare_env(output_dir, pdf_file_name, parse_method)
 
-    image_writer, md_writer = FileBasedDataWriter(
-        local_image_dir), FileBasedDataWriter(local_md_dir)
+    image_writer, md_writer = FileBasedDataWriter(local_image_dir), FileBasedDataWriter(
+        local_md_dir
+    )
     image_dir = str(os.path.basename(local_image_dir))
 
-    if parse_method == 'auto':
-        jso_useful_key = {'_pdf_type': '', 'model_list': model_list}
-        pipe = UNIPipe(pdf_bytes, jso_useful_key, image_writer, is_debug=True,
-                       # start_page_id=start_page_id, end_page_id=end_page_id,
-                       lang=lang,
-                       layout_model=layout_model, formula_enable=formula_enable, table_enable=table_enable)
-    elif parse_method == 'txt':
-        pipe = TXTPipe(pdf_bytes, model_list, image_writer, is_debug=True,
-                       # start_page_id=start_page_id, end_page_id=end_page_id,
-                       lang=lang,
-                       layout_model=layout_model, formula_enable=formula_enable, table_enable=table_enable)
-    elif parse_method == 'ocr':
-        pipe = OCRPipe(pdf_bytes, model_list, image_writer, is_debug=True,
-                       # start_page_id=start_page_id, end_page_id=end_page_id,
-                       lang=lang,
-                       layout_model=layout_model, formula_enable=formula_enable, table_enable=table_enable)
-    else:
-        logger.error('unknown parse method')
-        exit(1)
-
-    pipe.pipe_classify()
+    ds = PymuDocDataset(pdf_bytes)
 
     if len(model_list) == 0:
         if model_config.__use_inside_model__:
-            pipe.pipe_analyze()
-            orig_model_list = copy.deepcopy(pipe.model_list)
+            if parse_method == 'auto':
+                if ds.classify() == SupportedPdfParseMethod.TXT:
+                    infer_result = ds.apply(
+                        doc_analyze,
+                        ocr=False,
+                        lang=lang,
+                        layout_model=layout_model,
+                        formula_enable=formula_enable,
+                        table_enable=table_enable,
+                    )
+                    pipe_result = infer_result.pipe_txt_mode(
+                        image_writer, debug_mode=True, lang=lang
+                    )
+                else:
+                    infer_result = ds.apply(
+                        doc_analyze,
+                        ocr=True,
+                        lang=lang,
+                        layout_model=layout_model,
+                        formula_enable=formula_enable,
+                        table_enable=table_enable,
+                    )
+                    pipe_result = infer_result.pipe_ocr_mode(
+                        image_writer, debug_mode=True, lang=lang
+                    )
+
+            elif parse_method == 'txt':
+                infer_result = ds.apply(
+                    doc_analyze,
+                    ocr=False,
+                    lang=lang,
+                    layout_model=layout_model,
+                    formula_enable=formula_enable,
+                    table_enable=table_enable,
+                )
+                pipe_result = infer_result.pipe_txt_mode(
+                    image_writer, debug_mode=True, lang=lang
+                )
+            elif parse_method == 'ocr':
+                infer_result = ds.apply(
+                    doc_analyze,
+                    ocr=True,
+                    lang=lang,
+                    layout_model=layout_model,
+                    formula_enable=formula_enable,
+                    table_enable=table_enable,
+                )
+                pipe_result = infer_result.pipe_ocr_mode(
+                    image_writer, debug_mode=True, lang=lang
+                )
+            else:
+                logger.error('unknown parse method')
+                exit(1)
         else:
             logger.error('need model list input')
             exit(2)
+    else:
+        infer_result = InferenceResult(model_list, ds)
+        if parse_method == 'ocr':
+            pipe_result = infer_result.pipe_ocr_mode(
+                image_writer, debug_mode=True, lang=lang
+            )
+        elif parse_method == 'txt':
+            pipe_result = infer_result.pipe_txt_mode(
+                image_writer, debug_mode=True, lang=lang
+            )
+        else:
+            pipe_result = infer_result.pipe_auto_mode(
+                image_writer, debug_mode=True, lang=lang
+            )
+
+    if f_draw_model_bbox:
+        infer_result.draw_model(
+            os.path.join(local_md_dir, f'{pdf_file_name}_model.pdf')
+        )
 
-    pipe.pipe_parse()
-    pdf_info = pipe.pdf_mid_data['pdf_info']
     if f_draw_layout_bbox:
-        draw_layout_bbox(pdf_info, pdf_bytes, local_md_dir, pdf_file_name)
+        pipe_result.draw_layout(
+            os.path.join(local_md_dir, f'{pdf_file_name}_layout.pdf')
+        )
     if f_draw_span_bbox:
-        draw_span_bbox(pdf_info, pdf_bytes, local_md_dir, pdf_file_name)
-    if f_draw_model_bbox:
-        draw_model_bbox(copy.deepcopy(orig_model_list), pdf_bytes, local_md_dir, pdf_file_name)
+        pipe_result.draw_span(os.path.join(local_md_dir, f'{pdf_file_name}_spans.pdf'))
+
     if f_draw_line_sort_bbox:
-        draw_line_sort_bbox(pdf_info, pdf_bytes, local_md_dir, pdf_file_name)
+        pipe_result.draw_line_sort(
+            os.path.join(local_md_dir, f'{pdf_file_name}_line_sort.pdf')
+        )
 
-    md_content = pipe.pipe_mk_markdown(image_dir, drop_mode=DropMode.NONE, md_make_mode=f_make_md_mode)
     if f_dump_md:
-        md_writer.write_string(
+        pipe_result.dump_md(
+            md_writer,
             f'{pdf_file_name}.md',
-            md_content
+            image_dir,
+            drop_mode=DropMode.NONE,
+            md_make_mode=f_make_md_mode,
         )
 
     if f_dump_middle_json:
-        md_writer.write_string(
-            f'{pdf_file_name}_middle.json',
-            json_parse.dumps(pipe.pdf_mid_data, ensure_ascii=False, indent=4)
-        )
+        pipe_result.dump_middle_json(md_writer, f'{pdf_file_name}_middle.json')
 
     if f_dump_model_json:
-        md_writer.write_string(
-            f'{pdf_file_name}_model.json',
-            json_parse.dumps(orig_model_list, ensure_ascii=False, indent=4)
-        )
+        infer_result.dump_model(md_writer, f'{pdf_file_name}_model.json')
 
     if f_dump_orig_pdf:
         md_writer.write(
@@ -173,11 +222,11 @@ def do_parse(
             pdf_bytes,
         )
 
-    content_list = pipe.pipe_mk_uni_format(image_dir, drop_mode=DropMode.NONE)
     if f_dump_content_list:
-        md_writer.write_string(
+        pipe_result.dump_content_list(
+            md_writer,
             f'{pdf_file_name}_content_list.json',
-            json_parse.dumps(content_list, ensure_ascii=False, indent=4)
+            image_dir
         )
 
     logger.info(f'local output dir is {local_md_dir}')

magic_pdf/user_api.py

@@ -10,22 +10,29 @@
 from loguru import logger
 
 from magic_pdf.data.data_reader_writer import DataWriter
+from magic_pdf.data.dataset import Dataset
 from magic_pdf.libs.version import __version__
 from magic_pdf.model.doc_analyze_by_custom_model import doc_analyze
 from magic_pdf.pdf_parse_by_ocr import parse_pdf_by_ocr
 from magic_pdf.pdf_parse_by_txt import parse_pdf_by_txt
-
-PARSE_TYPE_TXT = 'txt'
-PARSE_TYPE_OCR = 'ocr'
-
-
-def parse_txt_pdf(pdf_bytes: bytes, pdf_models: list, imageWriter: DataWriter, is_debug=False,
-                  start_page_id=0, end_page_id=None, lang=None,
-                  *args, **kwargs):
+from magic_pdf.config.constants import PARSE_TYPE_TXT, PARSE_TYPE_OCR
+
+
+def parse_txt_pdf(
+    dataset: Dataset,
+    model_list: list,
+    imageWriter: DataWriter,
+    is_debug=False,
+    start_page_id=0,
+    end_page_id=None,
+    lang=None,
+    *args,
+    **kwargs
+):
     """解析文本类pdf."""
     pdf_info_dict = parse_pdf_by_txt(
-        pdf_bytes,
-        pdf_models,
+        dataset,
+        model_list,
         imageWriter,
         start_page_id=start_page_id,
         end_page_id=end_page_id,
@@ -43,13 +50,21 @@ def parse_txt_pdf(pdf_bytes: bytes, pdf_models: list, imageWriter: DataWriter, i
     return pdf_info_dict
 
 
-def parse_ocr_pdf(pdf_bytes: bytes, pdf_models: list, imageWriter: DataWriter, is_debug=False,
-                  start_page_id=0, end_page_id=None, lang=None,
-                  *args, **kwargs):
+def parse_ocr_pdf(
+    dataset: Dataset,
+    model_list: list,
+    imageWriter: DataWriter,
+    is_debug=False,
+    start_page_id=0,
+    end_page_id=None,
+    lang=None,
+    *args,
+    **kwargs
+):
     """解析ocr类pdf."""
     pdf_info_dict = parse_pdf_by_ocr(
-        pdf_bytes,
-        pdf_models,
+        dataset,
+        model_list,
         imageWriter,
         start_page_id=start_page_id,
         end_page_id=end_page_id,
@@ -67,17 +82,24 @@ def parse_ocr_pdf(pdf_bytes: bytes, pdf_models: list, imageWriter: DataWriter, i
     return pdf_info_dict
 
 
-def parse_union_pdf(pdf_bytes: bytes, pdf_models: list, imageWriter: DataWriter, is_debug=False,
-                    input_model_is_empty: bool = False,
-                    start_page_id=0, end_page_id=None, lang=None,
-                    *args, **kwargs):
+def parse_union_pdf(
+    dataset: Dataset,
+    model_list: list,
+    imageWriter: DataWriter,
+    is_debug=False,
+    start_page_id=0,
+    end_page_id=None,
+    lang=None,
+    *args,
+    **kwargs
+):
     """ocr和文本混合的pdf，全部解析出来."""
 
     def parse_pdf(method):
         try:
             return method(
-                pdf_bytes,
-                pdf_models,
+                dataset,
+                model_list,
                 imageWriter,
                 start_page_id=start_page_id,
                 end_page_id=end_page_id,
@@ -91,12 +113,12 @@ def parse_union_pdf(pdf_bytes: bytes, pdf_models: list, imageWriter: DataWriter,
     pdf_info_dict = parse_pdf(parse_pdf_by_txt)
     if pdf_info_dict is None or pdf_info_dict.get('_need_drop', False):
         logger.warning('parse_pdf_by_txt drop or error, switch to parse_pdf_by_ocr')
-        if input_model_is_empty:
+        if len(model_list) == 0:
             layout_model = kwargs.get('layout_model', None)
             formula_enable = kwargs.get('formula_enable', None)
             table_enable = kwargs.get('table_enable', None)
-            pdf_models = doc_analyze(
-                pdf_bytes,
+            infer_res = doc_analyze(
+                dataset,
                 ocr=True,
                 start_page_id=start_page_id,
                 end_page_id=end_page_id,
@@ -105,6 +127,7 @@ def parse_union_pdf(pdf_bytes: bytes, pdf_models: list, imageWriter: DataWriter,
                 formula_enable=formula_enable,
                 table_enable=table_enable,
             )
+            model_list = infer_res.get_infer_res()
         pdf_info_dict = parse_pdf(parse_pdf_by_ocr)
         if pdf_info_dict is None:
             raise Exception('Both parse_pdf_by_txt and parse_pdf_by_ocr failed.')

{magic_pdf-0.10.4.dist-info → magic_pdf-0.10.6.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: magic-pdf
-Version: 0.10.4
+Version: 0.10.6
 Summary: A practical tool for converting PDF to Markdown
 Home-page: https://github.com/opendatalab/MinerU
 Requires-Python: >=3.9
@@ -15,11 +15,14 @@ Requires-Dist: numpy<2.0.0,>=1.21.6
 Requires-Dist: pydantic<2.8.0,>=2.7.2
 Requires-Dist: PyMuPDF>=1.24.9
 Requires-Dist: scikit-learn>=1.0.2
-Requires-Dist: torch<=2.3.1,>=2.2.2
+Requires-Dist: torch>=2.2.2
 Requires-Dist: transformers
+Requires-Dist: pdfminer.six==20231228
 Provides-Extra: full
-Requires-Dist: unimernet==0.2.1; extra == "full"
-Requires-Dist: ultralytics; extra == "full"
+Requires-Dist: unimernet==0.2.2; extra == "full"
+Requires-Dist: torch<=2.3.1,>=2.2.2; extra == "full"
+Requires-Dist: torchvision<=0.18.1,>=0.17.2; extra == "full"
+Requires-Dist: ultralytics>=8.3.48; extra == "full"
 Requires-Dist: paddleocr==2.7.3; extra == "full"
 Requires-Dist: struct-eqtable==0.3.2; extra == "full"
 Requires-Dist: einops; extra == "full"

{magic_pdf-0.10.4.dist-info → magic_pdf-0.10.6.dist-info}/RECORD

@@ -1,10 +1,10 @@
 magic_pdf/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/pdf_parse_by_ocr.py,sha256=WTaLVSU2wRpgtldasnqbrw1B0OvVi8VvcB_t-dAIfmw,880
-magic_pdf/pdf_parse_by_txt.py,sha256=dh3ZM6BVrFzwbH4137BPUdKhgacGlpS2N4mn74_-UaA,762
-magic_pdf/pdf_parse_union_core_v2.py,sha256=6Apku7-pW450HbHNTtbVLDyroRSKlQ57w9f0ScOaZv4,30879
-magic_pdf/user_api.py,sha256=Sh6U7iD5VsH7Qkav_0o5GTx-Rlj7vhmhHQHZSBKR5T8,4006
+magic_pdf/pdf_parse_by_ocr.py,sha256=WFk6jhHSGvy8-hU2Qlpo5q-VORdSK_5Erh9IA_H7ZbQ,840
+magic_pdf/pdf_parse_by_txt.py,sha256=1-xieVOP8qmAC957ftzSzaeviv0-QC4yL6Lv6Pcg_6Y,722
+magic_pdf/pdf_parse_union_core_v2.py,sha256=Hl8PSJOJFHAGCdTwX4YY2_MMgjAuat47yALLb_E-DYg,30879
+magic_pdf/user_api.py,sha256=EAalk3WfQTfBq4qKMcISuHSjQg2Ku61ox_WiOPeFfuY,4060
 magic_pdf/config/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/config/constants.py,sha256=gqhUEtso7rCop-k-VvEPAMW_6pA6Tv2Y9smrr_0Iajo,1173
+magic_pdf/config/constants.py,sha256=CEhNtP8o_2zcK6DesO6cNDlpS9fUdRv-QUyHw0_vsso,1222
 magic_pdf/config/drop_reason.py,sha256=CqjMzBE96Qo8OeFvhhhItY8WhyqsKhE3DmyJLoQZNCc,2248
 magic_pdf/config/drop_tag.py,sha256=CjveyzhAsHm_bfXB7ZZNKruw1NR-WdKD8Hz6OhQdG0A,680
 magic_pdf/config/enums.py,sha256=CImYuw4sbKpq9zrj6zrrEvtdoGkjxDt8S8ByNVDpypU,89
@@ -13,12 +13,12 @@ magic_pdf/config/make_content_config.py,sha256=J2eJIhVHBPGwX18zVQomQUOxs8LcfeGLx
 magic_pdf/config/model_block_type.py,sha256=y5ie2ZLvo-h8OdVk8HOEha6qK0OJFtLmtOhYjrV680g,166
 magic_pdf/config/ocr_content_type.py,sha256=e_7RBTdShaWvWhMO2SFou7GM521elMH_Jtn5usbHWdY,890
 magic_pdf/data/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/data/dataset.py,sha256=n8rGw1-wizABR8giSk_XWPCXzx3478u5DK2Z0wOCOeI,5089
+magic_pdf/data/dataset.py,sha256=NpljxcttgRk4_Rl8Rf191t_vNIdbqIpK5x1xHAGE2iI,10686
 magic_pdf/data/read_api.py,sha256=hGpSVg9EcyM2mIlOsDIwsl7Y_ybWf9kkoxRumIXSzQQ,3566
 magic_pdf/data/schemas.py,sha256=oIUTBzK8Wq8Wuy8A_uilWAbVhucRvOs9_f3lSKYgcmQ,664
-magic_pdf/data/utils.py,sha256=dJZiqygwNier0UG5tbt5jAPjgwcnfsAN6-m-G1kVPLQ,917
+magic_pdf/data/utils.py,sha256=uaSHprh80D_puPUmd1slQDoE4uecNn4zZMzYWY0-a-8,917
 magic_pdf/data/data_reader_writer/__init__.py,sha256=QtevUaeSivv9dQKi3Tomfn4Z0E4To0cB8qXTnglxaHc,705
-magic_pdf/data/data_reader_writer/base.py,sha256=gUrHCMTHYBrWpqgHdIc-hN7HHwUC2ApK_VXrDUrnfdg,1320
+magic_pdf/data/data_reader_writer/base.py,sha256=nqmAcdHOXMOJO6RAT3ILligDFaw8Op0STyCw5yOzAbI,1706
 magic_pdf/data/data_reader_writer/filebase.py,sha256=iVjBT1M_89F5HB2uMsBzLfCNhsTVnmurJk4VUuAA2tw,2111
 magic_pdf/data/data_reader_writer/multi_bucket_s3.py,sha256=_HA8NJO1Be7KwozlwOJ90o8Ik2vfjlvlDPXppESeIfk,5885
 magic_pdf/data/data_reader_writer/s3.py,sha256=9Oy1cNuXMwG1e8PgZ7AR-pn_MqHAhkgAGnyEZCYoYAA,2408
@@ -27,10 +27,10 @@ magic_pdf/data/io/base.py,sha256=SqNQqe30ZvoVvg7GVv-hLMCjN6yBgDyQQWeLgGsTfhQ,111
 magic_pdf/data/io/http.py,sha256=XlKB0DNf4a_uUnfgcclvaaOtmE7lmddx0DnK8A-emAM,958
 magic_pdf/data/io/s3.py,sha256=hyA7sbNriQy64xd_uyJ7acN_oneQ1Pdmoc7_xcvkue8,3606
 magic_pdf/dict2md/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/dict2md/ocr_mkcontent.py,sha256=pE-lEUsYAhZC3nSmbgYO42Kvk_bW8Ds-AL-QMPHFu8c,12941
-magic_pdf/filter/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+magic_pdf/dict2md/ocr_mkcontent.py,sha256=zmEbxuIdFPfy3W72Zx_EEgyYtIOKcTa-0JoXHgXkEJ8,13046
+magic_pdf/filter/__init__.py,sha256=rV4dvUxfKyVErDx9ZbUp8DVq_fRIlv0lfSXp1ND4STc,1503
 magic_pdf/filter/pdf_classify_by_type.py,sha256=spmDO-f2ihAl1d6-EP-j271Yi50oyu6mw4X2kRd_m0s,42320
-magic_pdf/filter/pdf_meta_scan.py,sha256=3ba7SxXu1z2r5N97Dxmp_L10Lo7llsrBlvtEAJeIJBQ,17403
+magic_pdf/filter/pdf_meta_scan.py,sha256=rqTuStW2_ICr3HmV_9IQ5jnsl4JnSh7-bL11vbtH3i0,17470
 magic_pdf/integrations/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/integrations/rag/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/integrations/rag/api.py,sha256=t38wvIBzLje4_JzTP3dewMLqV-tQJ-A3B92Sj2oyrfs,2507
@@ -43,26 +43,27 @@ magic_pdf/libs/commons.py,sha256=xD0fGA16KNB5rhbl4zRrOqdrNHYwaRablT_s9W2ZTbw,117
 magic_pdf/libs/config_reader.py,sha256=vDsxw2xbW7Gb1mKqERTSlttbXFNtVU0BDdae2dG7wEI,4068
 magic_pdf/libs/convert_utils.py,sha256=Ov-lsfCLBPz_15iSJXIslBNmrSf_E_1g_XDWJy8NgO8,143
 magic_pdf/libs/coordinate_transform.py,sha256=Bbop2cP2uz2ZG0U0gwd7J6EKkgABq5Rv03qf2LMPw80,429
-magic_pdf/libs/draw_bbox.py,sha256=2IXr4TUxm0-pXYIPkNaELWo9pOysZC6etpqzTE5eg-w,17588
+magic_pdf/libs/draw_bbox.py,sha256=Z7-OOETUo90yj3tCV8MwbiJwckThcC0bjs4MXI9ocac,17561
 magic_pdf/libs/hash_utils.py,sha256=VEKK9WfFoZgrPfi8kfITjLpr8Ahufs8tXh9R1Y5lAL8,404
 magic_pdf/libs/json_compressor.py,sha256=6-KCu0lb5ksmyqWtQGb4QqmP-FjRb5dP7P-Hevcn68g,875
 magic_pdf/libs/language.py,sha256=Hj5-lrGoNExxdHLbkcNG-c27U4AjJ9AZPdZblaNSehU,1099
 magic_pdf/libs/local_math.py,sha256=tqljQOgqh3fZc146HYhO88JXJaiXMVwArBkk_CSGICc,177
 magic_pdf/libs/markdown_utils.py,sha256=86v2BmsSV4NkoRZrH4uQD1youJhYFF3vIKr_vDeg3z0,270
 magic_pdf/libs/path_utils.py,sha256=Hykw_l5CU736b2egHV9P7B-qh3QNKO4nZSGCbsi0Z8E,1043
-magic_pdf/libs/pdf_check.py,sha256=wCVOcwEPeMRcHW5OGN-GSQnPT5qNXUYHWWowoUknxF4,3178
+magic_pdf/libs/pdf_check.py,sha256=zBwUThKKBtnrNPmgE10lYsTy1Kq7j_6IejO7JR0J4pA,3118
 magic_pdf/libs/pdf_image_tools.py,sha256=kjzSEbm7K0yiHv8kJ4VbZ9HHktM8qvAv3LhxRyDZEQk,1987
 magic_pdf/libs/safe_filename.py,sha256=ckwcM_eqoysTb5id8czp-tXq2G9da0-l3pshZDCHQtE,236
-magic_pdf/libs/version.py,sha256=fGZMaoPHZfTX9I4TDkr07gp-kj_1U_SD-gjQC_2flQs,23
-magic_pdf/model/__init__.py,sha256=1QcfMKET0xQhSaZMjNQHi_TjzSSDR6PI5mjkmaXHPe8,52
-magic_pdf/model/doc_analyze_by_custom_model.py,sha256=YZwlhIgidy1_MUyTM_MRSLfKR_rpi508Bra6Vpj8PJ4,7125
+magic_pdf/libs/version.py,sha256=7qmFu9Qmzy5OxKJPN-LQOkzV_2T4cJYrUSLTfq7F3kE,23
+magic_pdf/model/__init__.py,sha256=R6uhAQucHJa87V81ahYHWEffG0-3F1792J4kaSxZpi8,3698
+magic_pdf/model/doc_analyze_by_custom_model.py,sha256=KAPRDgWUAzsXbofZ6i0ll9eaanPdPnfjM1nn4Pl8-Zo,7588
 magic_pdf/model/magic_model.py,sha256=ppMkMqtP7sKncHTZ2SbXuPOoR988iRPexBEMA6QeiIc,42208
 magic_pdf/model/model_list.py,sha256=tJ9jtMB93HGx8Rmt8wmQSDFXZBUIPQrwaaYsep4luTM,183
-magic_pdf/model/pdf_extract_kit.py,sha256=ceYWlSU1BhakfsHPVM9SrUx35EvCBa20uJmgDO5PAtE,10933
+magic_pdf/model/operators.py,sha256=qcacETf6j-gDUj9g0zYJgBrkq0YWe6ZlfoPjJhCMUYU,6628
+magic_pdf/model/pdf_extract_kit.py,sha256=6JdWkdKOgL9UyAlI5znPMexs0AMZzn1SgrIpJUxWiGs,11839
 magic_pdf/model/pp_structure_v2.py,sha256=NcqFWL4nUtjl82MFak8HX_8V3i4Aw_fK4dATrIp5uGs,3840
 magic_pdf/model/sub_modules/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/model/sub_modules/model_init.py,sha256=CnlZLsiSOmGJXQRASH-hMmuPiF6hYKCNfmzDTjQqy5g,5073
-magic_pdf/model/sub_modules/model_utils.py,sha256=ToiuwXbrvH_CPIwW2AXzz9miadUN5FA7lthwBljtIco,2118
+magic_pdf/model/sub_modules/model_init.py,sha256=Sp4I2tQ2oFsTIBRHXv8-44WU1PvPSx4L3VfwnQUaRFo,5438
+magic_pdf/model/sub_modules/model_utils.py,sha256=svV5bn_Xw3QqSa22h7OrmlQQQySSqe3DdE6KMEURr2c,2219
 magic_pdf/model/sub_modules/layout/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/model/sub_modules/layout/doclayout_yolo/DocLayoutYOLO.py,sha256=roe6Rth6cvBrCw0MWXcj1CBjvK3S_Ni7GC4DxY4-yBQ,886
 magic_pdf/model/sub_modules/layout/doclayout_yolo/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -110,15 +111,16 @@ magic_pdf/model/sub_modules/table/structeqtable/struct_eqtable.py,sha256=SrNPm-u
 magic_pdf/model/sub_modules/table/tablemaster/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/model/sub_modules/table/tablemaster/tablemaster_paddle.py,sha256=QEQ-56AzoIAU7UWsEidWW_KDOY5r16qm2kSpox8cxq4,2755
 magic_pdf/para/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/para/para_split_v3.py,sha256=UOQe0HUVX7FAlMbJp1OkGfdM7JECWeqscv3s8Hge7ps,16922
-magic_pdf/pipe/AbsPipe.py,sha256=jPtAa0pz_vPddya3ZpUk6UrGqp8PcBdLONO1spzavQo,4371
-magic_pdf/pipe/OCRPipe.py,sha256=nuN-zpUzu--gyrC0_vsvvilAyK7Mp3Tom_UOnsur1ps,2158
-magic_pdf/pipe/TXTPipe.py,sha256=5OFo2e8U5Y24wJrFDEJghBDpklnKFEnzKTYVnnhQssE,2159
-magic_pdf/pipe/UNIPipe.py,sha256=ik0xXPdsHo7Un0gFpLC5ul04BP3Omd2mp5gqem40deE,4807
+magic_pdf/para/para_split_v3.py,sha256=v4SdQn4OZdHRXpWQMfQ-FGJz_tglQ88uFUqpwY542Fo,16922
+magic_pdf/pipe/AbsPipe.py,sha256=_Lx4Ags5suEvmJEvgHEvg6n0RP4Yqjc1VBWaCP0la2o,4410
+magic_pdf/pipe/OCRPipe.py,sha256=nH21Rq7mQEw7pS7AVD2MRFdSE0DxGc1wk9VXB6T0m3A,2396
+magic_pdf/pipe/TXTPipe.py,sha256=JXJ7hzD7TNq5VnCt33dck2FM15GpozJoHibaRlYD14s,2196
+magic_pdf/pipe/UNIPipe.py,sha256=i0kWflZ5BFHrx8p8vDntRcN6jecaxOfGq11ANtYvrZY,5011
 magic_pdf/pipe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+magic_pdf/pipe/operators.py,sha256=5z7kF95IWyBGxs4tIhqJml2YMlfDkU9B5xy__NiUxz0,4962
 magic_pdf/pre_proc/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/pre_proc/construct_page_dict.py,sha256=OFmq5XRKi6fYIo-lmGlL-NB16Sf0egzsfEx-fT2uYrc,660
-magic_pdf/pre_proc/cut_image.py,sha256=U-ttnl3lAhhmgtkR1GGyPAVm0i0-6VscXf3E2EDy3lE,1187
+magic_pdf/pre_proc/cut_image.py,sha256=NDzbxwD7z7Tb4uAxL4KR6LzURFdN1Tzr4nPvj-VmEqc,1225
 magic_pdf/pre_proc/ocr_detect_all_bboxes.py,sha256=nt88ttXCEI_1ihAF7HU15SQjwM69V-iJmk-L_nyzA6o,9328
 magic_pdf/pre_proc/ocr_dict_merge.py,sha256=Ycgz2whzotL7kwl0-mHNV48QOQ2j4tRXqLSQrJRojYg,4847
 magic_pdf/pre_proc/ocr_span_list_modify.py,sha256=a5OmIwtkXkz6fTQg6p8R-f1nA_w0rgMwKFQjfs_HwrE,2864
@@ -136,12 +138,12 @@ magic_pdf/spark/spark_api.py,sha256=BYO6zlRW0cEnIUB3ZzNQTu_LsPHEVitqiUN7gy3x_wo,
 magic_pdf/tools/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/tools/cli.py,sha256=83a8p4_DvVdDOTuviE6WqexSXsDE_MUY-af3QDxXeoU,3067
 magic_pdf/tools/cli_dev.py,sha256=3RbubfTIagWoFYdu8wSDanr-BJDjFGeDet55jTy7He0,3948
-magic_pdf/tools/common.py,sha256=ILTv8YjnK-XTVV5nzak3Sm-EJJXjG1hJJghlYKgYVBQ,6809
+magic_pdf/tools/common.py,sha256=x3dNHT9wEpdmkkEb4Y70DmUMMPavre5C82T0v9OmA2g,7894
 magic_pdf/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/utils/annotations.py,sha256=82ou3uELNbQWa9hOFFkVt0gsIskAKf5msCv5J2IJ5V0,211
-magic_pdf-0.10.4.dist-info/LICENSE.md,sha256=jVa0BUaKrRH4erV2P5AeJ24I2WRv9chIGxditreJ6e0,34524
-magic_pdf-0.10.4.dist-info/METADATA,sha256=pujqC_qUWiPT-L6R065MoL0QO9q4IEra0iW4BCRkxr4,36992
-magic_pdf-0.10.4.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
-magic_pdf-0.10.4.dist-info/entry_points.txt,sha256=wXwYke3j8fqDQTocUspL-CqDUEv3Tfcwp09fM8dZAhA,98
-magic_pdf-0.10.4.dist-info/top_level.txt,sha256=J9I0AzmHWGkp9c6DL8Oe4mEx3yYphLzkRn4H25Lg1rE,10
-magic_pdf-0.10.4.dist-info/RECORD,,
+magic_pdf-0.10.6.dist-info/LICENSE.md,sha256=jVa0BUaKrRH4erV2P5AeJ24I2WRv9chIGxditreJ6e0,34524
+magic_pdf-0.10.6.dist-info/METADATA,sha256=CbT8tghajhhMHEawiHakbU-ndjeJ_J9J1011PFoYDbA,37144
+magic_pdf-0.10.6.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
+magic_pdf-0.10.6.dist-info/entry_points.txt,sha256=wXwYke3j8fqDQTocUspL-CqDUEv3Tfcwp09fM8dZAhA,98
+magic_pdf-0.10.6.dist-info/top_level.txt,sha256=J9I0AzmHWGkp9c6DL8Oe4mEx3yYphLzkRn4H25Lg1rE,10
+magic_pdf-0.10.6.dist-info/RECORD,,