magic-pdf 0.10.1__py3-none-any.whl → 0.10.3__py3-none-any.whl

magic_pdf/pre_proc/ocr_span_list_modify.py

@@ -1,10 +1,7 @@
 
 from magic_pdf.config.drop_tag import DropTag
-from magic_pdf.config.ocr_content_type import BlockType, ContentType
-from magic_pdf.libs.boxbase import (__is_overlaps_y_exceeds_threshold,
-                                    calculate_iou,
-                                    calculate_overlap_area_in_bbox1_area_ratio,
-                                    get_minbox_if_overlap_by_ratio)
+from magic_pdf.config.ocr_content_type import BlockType
+from magic_pdf.libs.boxbase import calculate_iou, get_minbox_if_overlap_by_ratio
 
 
 def remove_overlaps_low_confidence_spans(spans):
@@ -59,253 +56,6 @@ def remove_overlaps_min_spans(spans):
     return spans, dropped_spans
 
 
-def remove_spans_by_bboxes(spans, need_remove_spans_bboxes):
-    # 遍历spans, 判断是否在removed_span_block_bboxes中
-    # 如果是, 则删除该span 否则, 保留该span
-    need_remove_spans = []
-    for span in spans:
-        for removed_bbox in need_remove_spans_bboxes:
-            if (
-                calculate_overlap_area_in_bbox1_area_ratio(span['bbox'], removed_bbox)
-                > 0.5
-            ):
-                if span not in need_remove_spans:
-                    need_remove_spans.append(span)
-                    break
-
-    if len(need_remove_spans) > 0:
-        for span in need_remove_spans:
-            spans.remove(span)
-
-    return spans
-
-
-def remove_spans_by_bboxes_dict(spans, need_remove_spans_bboxes_dict):
-    dropped_spans = []
-    for drop_tag, removed_bboxes in need_remove_spans_bboxes_dict.items():
-        # logger.info(f"remove spans by bbox dict, drop_tag: {drop_tag}, removed_bboxes: {removed_bboxes}")
-        need_remove_spans = []
-        for span in spans:
-            # 通过判断span的bbox是否在removed_bboxes中, 判断是否需要删除该span
-            for removed_bbox in removed_bboxes:
-                if (
-                    calculate_overlap_area_in_bbox1_area_ratio(
-                        span['bbox'], removed_bbox
-                    )
-                    > 0.5
-                ):
-                    need_remove_spans.append(span)
-                    break
-                # 当drop_tag为DropTag.FOOTNOTE时, 判断span是否在removed_bboxes中任意一个的下方，如果是,则删除该span
-                elif (
-                    drop_tag == DropTag.FOOTNOTE
-                    and (span['bbox'][1] + span['bbox'][3]) / 2 > removed_bbox[3]
-                    and removed_bbox[0]
-                    < (span['bbox'][0] + span['bbox'][2]) / 2
-                    < removed_bbox[2]
-                ):
-                    need_remove_spans.append(span)
-                    break
-
-        for span in need_remove_spans:
-            spans.remove(span)
-            span['tag'] = drop_tag
-            dropped_spans.append(span)
-
-    return spans, dropped_spans
-
-
-def adjust_bbox_for_standalone_block(spans):
-    # 对tpye=["interline_equation", "image", "table"]进行额外处理,如果左边有字的话,将该span的bbox中y0调整至不高于文字的y0
-    for sb_span in spans:
-        if sb_span['type'] in [
-            ContentType.InterlineEquation,
-            ContentType.Image,
-            ContentType.Table,
-        ]:
-            for text_span in spans:
-                if text_span['type'] in [ContentType.Text, ContentType.InlineEquation]:
-                    # 判断span2的纵向高度是否被span所覆盖
-                    if (
-                        sb_span['bbox'][1] < text_span['bbox'][1]
-                        and sb_span['bbox'][3] > text_span['bbox'][3]
-                    ):
-                        # 判断span2是否在span左边
-                        if text_span['bbox'][0] < sb_span['bbox'][0]:
-                            # 调整span的y0和span2的y0一致
-                            sb_span['bbox'][1] = text_span['bbox'][1]
-    return spans
-
-
-def modify_y_axis(spans: list, displayed_list: list, text_inline_lines: list):
-    # displayed_list = []
-    # 如果spans为空,则不处理
-    if len(spans) == 0:
-        pass
-    else:
-        spans.sort(key=lambda span: span['bbox'][1])
-
-        lines = []
-        current_line = [spans[0]]
-        if spans[0]['type'] in [
-            ContentType.InterlineEquation,
-            ContentType.Image,
-            ContentType.Table,
-        ]:
-            displayed_list.append(spans[0])
-
-        line_first_y0 = spans[0]['bbox'][1]
-        line_first_y = spans[0]['bbox'][3]
-        # 用于给行间公式搜索
-        # text_inline_lines = []
-        for span in spans[1:]:
-            # if span.get("content","") == "78.":
-            #     print("debug")
-            # 如果当前的span类型为"interline_equation" 或者 当前行中已经有"interline_equation"
-            # image和table类型，同上
-            if span['type'] in [
-                ContentType.InterlineEquation,
-                ContentType.Image,
-                ContentType.Table,
-            ] or any(
-                s['type']
-                in [ContentType.InterlineEquation, ContentType.Image, ContentType.Table]
-                for s in current_line
-            ):
-                # 传入
-                if span['type'] in [
-                    ContentType.InterlineEquation,
-                    ContentType.Image,
-                    ContentType.Table,
-                ]:
-                    displayed_list.append(span)
-                # 则开始新行
-                lines.append(current_line)
-                if len(current_line) > 1 or current_line[0]['type'] in [
-                    ContentType.Text,
-                    ContentType.InlineEquation,
-                ]:
-                    text_inline_lines.append(
-                        (current_line, (line_first_y0, line_first_y))
-                    )
-                current_line = [span]
-                line_first_y0 = span['bbox'][1]
-                line_first_y = span['bbox'][3]
-                continue
-
-            # 如果当前的span与当前行的最后一个span在y轴上重叠，则添加到当前行
-            if __is_overlaps_y_exceeds_threshold(
-                span['bbox'], current_line[-1]['bbox']
-            ):
-                if span['type'] == 'text':
-                    line_first_y0 = span['bbox'][1]
-                    line_first_y = span['bbox'][3]
-                current_line.append(span)
-
-            else:
-                # 否则，开始新行
-                lines.append(current_line)
-                text_inline_lines.append((current_line, (line_first_y0, line_first_y)))
-                current_line = [span]
-                line_first_y0 = span['bbox'][1]
-                line_first_y = span['bbox'][3]
-
-            # 添加最后一行
-        if current_line:
-            lines.append(current_line)
-            if len(current_line) > 1 or current_line[0]['type'] in [
-                ContentType.Text,
-                ContentType.InlineEquation,
-            ]:
-                text_inline_lines.append((current_line, (line_first_y0, line_first_y)))
-        for line in text_inline_lines:
-            # 按照x0坐标排序
-            current_line = line[0]
-            current_line.sort(key=lambda span: span['bbox'][0])
-
-        # 调整每一个文字行内bbox统一
-        for line in text_inline_lines:
-            current_line, (line_first_y0, line_first_y) = line
-            for span in current_line:
-                span['bbox'][1] = line_first_y0
-                span['bbox'][3] = line_first_y
-
-        # return spans, displayed_list, text_inline_lines
-
-
-def modify_inline_equation(spans: list, displayed_list: list, text_inline_lines: list):
-    # 错误行间公式转行内公式
-    j = 0
-    for i in range(len(displayed_list)):
-        # if i == 8:
-        #     print("debug")
-        span = displayed_list[i]
-        span_y0, span_y = span['bbox'][1], span['bbox'][3]
-
-        while j < len(text_inline_lines):
-            text_line = text_inline_lines[j]
-            y0, y1 = text_line[1]
-            if (
-                span_y0 < y0 < span_y
-                or span_y0 < y1 < span_y
-                or span_y0 < y0
-                and span_y > y1
-            ) and __is_overlaps_y_exceeds_threshold(span['bbox'], (0, y0, 0, y1)):
-                # 调整公式类型
-                if span['type'] == ContentType.InterlineEquation:
-                    # 最后一行是行间公式
-                    if j + 1 >= len(text_inline_lines):
-                        span['type'] = ContentType.InlineEquation
-                        span['bbox'][1] = y0
-                        span['bbox'][3] = y1
-                    else:
-                        # 行间公式旁边有多行文字或者行间公式比文字高3倍则不转换
-                        y0_next, y1_next = text_inline_lines[j + 1][1]
-                        if (
-                            not __is_overlaps_y_exceeds_threshold(
-                                span['bbox'], (0, y0_next, 0, y1_next)
-                            )
-                            and 3 * (y1 - y0) > span_y - span_y0
-                        ):
-                            span['type'] = ContentType.InlineEquation
-                            span['bbox'][1] = y0
-                            span['bbox'][3] = y1
-                break
-            elif (
-                span_y < y0
-                or span_y0 < y0 < span_y
-                and not __is_overlaps_y_exceeds_threshold(span['bbox'], (0, y0, 0, y1))
-            ):
-                break
-            else:
-                j += 1
-
-    return spans
-
-
-def get_qa_need_list(blocks):
-    # 创建 images, tables, interline_equations, inline_equations 的副本
-    images = []
-    tables = []
-    interline_equations = []
-    inline_equations = []
-
-    for block in blocks:
-        for line in block['lines']:
-            for span in line['spans']:
-                if span['type'] == ContentType.Image:
-                    images.append(span)
-                elif span['type'] == ContentType.Table:
-                    tables.append(span)
-                elif span['type'] == ContentType.InlineEquation:
-                    inline_equations.append(span)
-                elif span['type'] == ContentType.InterlineEquation:
-                    interline_equations.append(span)
-                else:
-                    continue
-    return images, tables, interline_equations, inline_equations
-
-
 def get_qa_need_list_v2(blocks):
     # 创建 images, tables, interline_equations, inline_equations 的副本
     images = []

magic_pdf/rw/S3ReaderWriter.py

@@ -1,5 +1,5 @@
 from magic_pdf.rw.AbsReaderWriter import AbsReaderWriter
-from magic_pdf.libs.commons import parse_aws_param, parse_bucket_key, join_path
+from magic_pdf.libs.commons import parse_bucket_key, join_path
 import boto3
 from loguru import logger
 from botocore.config import Config

{magic_pdf-0.10.1.dist-info → magic_pdf-0.10.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: magic-pdf
-Version: 0.10.1
+Version: 0.10.3
 Summary: A practical tool for converting PDF to Markdown
 Home-page: https://github.com/opendatalab/MinerU
 Requires-Python: >=3.9
@@ -12,7 +12,6 @@ Requires-Dist: click>=8.1.7
 Requires-Dist: fast-langdetect==0.2.0
 Requires-Dist: loguru>=0.6.0
 Requires-Dist: numpy<2.0.0,>=1.21.6
-Requires-Dist: pdfminer.six==20231228
 Requires-Dist: pydantic<2.8.0,>=2.7.2
 Requires-Dist: PyMuPDF>=1.24.9
 Requires-Dist: scikit-learn>=1.0.2
@@ -320,88 +319,14 @@ If your device supports CUDA and meets the GPU requirements of the mainline envi
 
 ### Command Line
 
-```bash
-magic-pdf --help
-Usage: magic-pdf [OPTIONS]
-
-Options:
-  -v, --version                display the version and exit
-  -p, --path PATH              local pdf filepath or directory  [required]
-  -o, --output-dir PATH        output local directory  [required]
-  -m, --method [ocr|txt|auto]  the method for parsing pdf. ocr: using ocr
-                               technique to extract information from pdf. txt:
-                               suitable for the text-based pdf only and
-                               outperform ocr. auto: automatically choose the
-                               best method for parsing pdf from ocr and txt.
-                               without method specified, auto will be used by
-                               default.
-  -l, --lang TEXT              Input the languages in the pdf (if known) to
-                               improve OCR accuracy.  Optional. You should
-                               input "Abbreviation" with language form url: ht
-                               tps://paddlepaddle.github.io/PaddleOCR/latest/en
-                               /ppocr/blog/multi_languages.html#5-support-languages-
-                               and-abbreviations
-  -d, --debug BOOLEAN          Enables detailed debugging information during
-                               the execution of the CLI commands.
-  -s, --start INTEGER          The starting page for PDF parsing, beginning
-                               from 0.
-  -e, --end INTEGER            The ending page for PDF parsing, beginning from
-                               0.
-  --help                       Show this message and exit.
-
-
-## show version
-magic-pdf -v
-
-## command line example
-magic-pdf -p {some_pdf} -o {some_output_dir} -m auto
-```
+[Using MinerU via Command Line](https://mineru.readthedocs.io/en/latest/user_guide/quick_start/command_line.html)
 
-`{some_pdf}` can be a single PDF file or a directory containing multiple PDFs.
-The results will be saved in the `{some_output_dir}` directory. The output file list is as follows:
-
-```text
-├── some_pdf.md                          # markdown file
-├── images                               # directory for storing images
-├── some_pdf_layout.pdf                  # layout diagram (Include layout reading order)
-├── some_pdf_middle.json                 # MinerU intermediate processing result
-├── some_pdf_model.json                  # model inference result
-├── some_pdf_origin.pdf                  # original PDF file
-├── some_pdf_spans.pdf                   # smallest granularity bbox position information diagram
-└── some_pdf_content_list.json           # Rich text JSON arranged in reading order
-```
 > [!TIP]
 > For more information about the output files, please refer to the [Output File Description](docs/output_file_en_us.md).
 
 ### API
 
-Processing files from local disk
-
-```python
-image_writer = DiskReaderWriter(local_image_dir)
-image_dir = str(os.path.basename(local_image_dir))
-jso_useful_key = {"_pdf_type": "", "model_list": []}
-pipe = UNIPipe(pdf_bytes, jso_useful_key, image_writer)
-pipe.pipe_classify()
-pipe.pipe_analyze()
-pipe.pipe_parse()
-md_content = pipe.pipe_mk_markdown(image_dir, drop_mode="none")
-```
-
-Processing files from object storage
-
-```python
-s3pdf_cli = S3ReaderWriter(pdf_ak, pdf_sk, pdf_endpoint)
-image_dir = "s3://img_bucket/"
-s3image_cli = S3ReaderWriter(img_ak, img_sk, img_endpoint, parent_path=image_dir)
-pdf_bytes = s3pdf_cli.read(s3_pdf_path, mode=s3pdf_cli.MODE_BIN)
-jso_useful_key = {"_pdf_type": "", "model_list": []}
-pipe = UNIPipe(pdf_bytes, jso_useful_key, s3image_cli)
-pipe.pipe_classify()
-pipe.pipe_analyze()
-pipe.pipe_parse()
-md_content = pipe.pipe_mk_markdown(image_dir, drop_mode="none")
-```
+[Using MinerU via Python API](https://mineru.readthedocs.io/en/latest/user_guide/quick_start/to_markdown.html)
 
 For detailed implementation, refer to:
 

{magic_pdf-0.10.1.dist-info → magic_pdf-0.10.3.dist-info}/RECORD

@@ -1,8 +1,7 @@
 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.py,sha256=w90lFIMOYUMAq4iv8bpsbBtLXFphPV4HyYeqbOTYQUI,12420
-magic_pdf/pdf_parse_union_core_v2.py,sha256=EqEi9AahBBh2JbXoY8uOCmClvi9W_H_26U4jK8RwPwU,31308
+magic_pdf/pdf_parse_union_core_v2.py,sha256=6Apku7-pW450HbHNTtbVLDyroRSKlQ57w9f0ScOaZv4,30879
 magic_pdf/user_api.py,sha256=Sh6U7iD5VsH7Qkav_0o5GTx-Rlj7vhmhHQHZSBKR5T8,4006
 magic_pdf/config/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/config/constants.py,sha256=gqhUEtso7rCop-k-VvEPAMW_6pA6Tv2Y9smrr_0Iajo,1173
@@ -28,51 +27,39 @@ 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/mkcontent.py,sha256=bMQK7uiay76YaWA92VIK57YajINV20SnOs65wOEXyKE,18667
-magic_pdf/dict2md/ocr_mkcontent.py,sha256=ohjhEFS9YFrzTCC9c9yrvi4QuZe9iZm1qlkQWB6xxIw,13038
+magic_pdf/dict2md/ocr_mkcontent.py,sha256=hwcHTEx1tbIlM9ukmPBOAyH0G6rmbOTu87nVtZ1gE6k,12354
 magic_pdf/filter/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/filter/pdf_classify_by_type.py,sha256=spmDO-f2ihAl1d6-EP-j271Yi50oyu6mw4X2kRd_m0s,42320
-magic_pdf/filter/pdf_meta_scan.py,sha256=h4D4O0OeAlEy2A8mJ6E0aQ8wIizIfsIxEagbjaomnAo,17823
+magic_pdf/filter/pdf_meta_scan.py,sha256=3ba7SxXu1z2r5N97Dxmp_L10Lo7llsrBlvtEAJeIJBQ,17403
 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
 magic_pdf/integrations/rag/type.py,sha256=Z_1g_ZIOCsb7-FmZBudReIXj8nzGrgj_BygCalhJdmk,3193
 magic_pdf/integrations/rag/utils.py,sha256=DCb-UhC8TElb6Eq7_6NmmETreKEk5DVE18hNL8sTEBk,11762
-magic_pdf/layout/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/layout/bbox_sort.py,sha256=PzzaBf6MC_AZ-ZWGU0Kg-KIsw874l_gML73mM3hE4Ps,30807
-magic_pdf/layout/layout_det_utils.py,sha256=NCYBTvsrULE3Cue53aMD1MfXTmOL9Xy0nivl6ku2cls,9137
-magic_pdf/layout/layout_sort.py,sha256=jtacQVcxnuYAksvEqtS0DH-v6U8qyjX-jmyZgDJ-egA,37005
-magic_pdf/layout/layout_spiler_recog.py,sha256=QjBSgB-a7J2yjUR1eaCs9ZD7URtiRnV6W934hpAeuC4,3067
-magic_pdf/layout/mcol_sort.py,sha256=ADnLisBJBHXDKYChcf2lzTb_TC_vZ4q89_CSN8mwEJc,11331
 magic_pdf/libs/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/libs/boxbase.py,sha256=ELMHWolgWROxOAQDgwmL7VS5kveZp4ifvEzRmPul2Ws,16925
-magic_pdf/libs/calc_span_stats.py,sha256=5vnU27DcbkFDRSAoLqAmX0KQ3I9ehWkEgh_t9hxg_zI,10147
 magic_pdf/libs/clean_memory.py,sha256=BIOmEWuwR7c_p4OwTSW2muE3PRaGhmOplS-wTXt_EXk,211
-magic_pdf/libs/commons.py,sha256=6Zu9-OyamyCNDY7qj0SxR-rux-ggj9im3CVPtC4ubB8,7108
+magic_pdf/libs/commons.py,sha256=xD0fGA16KNB5rhbl4zRrOqdrNHYwaRablT_s9W2ZTbw,1174
 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/detect_language_from_model.py,sha256=Uln8F9qs8EJOw4EgI7KRlaU3lD_mK8KMTlADLFtz8fk,816
-magic_pdf/libs/draw_bbox.py,sha256=NhAfqib5HYuGjjrAG_SvJR-yOHZTy6tzDxLXdxKlULQ,17676
+magic_pdf/libs/draw_bbox.py,sha256=2IXr4TUxm0-pXYIPkNaELWo9pOysZC6etpqzTE5eg-w,17588
 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=cLxLXjRhrNp_wCHvtglrGA_FVdrvfd1KULeTtj1p18w,944
-magic_pdf/libs/nlp_utils.py,sha256=-X9W3-Ns5ZdDYFvyyEq6i6P2b5hCATaFEZeOjwNOH9M,6901
+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=MAe8wzwT0qvPf_I72wEZG7k1g4haNHS7oUtLqkB5rlE,2145
-magic_pdf/libs/pdf_image_tools.py,sha256=sh8hgBQu_83R71qBLodOFdByBUuQujsOMfgpSD9mrhE,1981
+magic_pdf/libs/pdf_check.py,sha256=wCVOcwEPeMRcHW5OGN-GSQnPT5qNXUYHWWowoUknxF4,3178
+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/textbase.py,sha256=SC1Frhz3Fb7V7n2SFRBsl7Bmg0JZdlvZskq0lfW1vIk,732
-magic_pdf/libs/version.py,sha256=v7Gyp89umFzDtY45tTjCdXqZnQ2RN01AibdYNxEvxYo,23
-magic_pdf/libs/vis_utils.py,sha256=hTOTEakKV0pGMbk0tbRkVI_tku7A3dGc96ynObZ4kwI,10207
+magic_pdf/libs/version.py,sha256=0C8KcY1dzs3hdkAre06v0NCQ0Uxcqv6g9a93bRcVLW0,23
 magic_pdf/model/__init__.py,sha256=1QcfMKET0xQhSaZMjNQHi_TjzSSDR6PI5mjkmaXHPe8,52
-magic_pdf/model/doc_analyze_by_custom_model.py,sha256=HOT6chGx2VPyH6O9WB0c6xGPeDs9m_6oZn3iOa745yw,7125
-magic_pdf/model/magic_model.py,sha256=8nJLzNCa0Ag4JhMAQbjj5qrkj617qKPCXVJAiT9DnaA,43472
+magic_pdf/model/doc_analyze_by_custom_model.py,sha256=YZwlhIgidy1_MUyTM_MRSLfKR_rpi508Bra6Vpj8PJ4,7125
+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/pp_structure_v2.py,sha256=BKPN7W4BjG0eWPAPjPEac1RMnb5eIzmAz4E4Rq-9b1U,3019
+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
@@ -107,8 +94,8 @@ magic_pdf/model/sub_modules/mfr/unimernet/Unimernet.py,sha256=jeJkqID6L1ZivPMdK1
 magic_pdf/model/sub_modules/mfr/unimernet/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/model/sub_modules/ocr/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/model/sub_modules/ocr/paddleocr/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/model/sub_modules/ocr/paddleocr/ocr_utils.py,sha256=UP7fADPGoxAMj2SUKmeW-fe_AcAQxlT9Mfy4WF6vHmU,9796
-magic_pdf/model/sub_modules/ocr/paddleocr/ppocr_273_mod.py,sha256=a6xkQHqLMUL4NCaORp8oo4Tfa8GB8PN9MVvG7Pj6jIE,7316
+magic_pdf/model/sub_modules/ocr/paddleocr/ocr_utils.py,sha256=rwKphio9SZgiNgqASWOBWZIf6PPi3kvgQO_qJLc_diE,10726
+magic_pdf/model/sub_modules/ocr/paddleocr/ppocr_273_mod.py,sha256=d__xICejA_Q-Cz4cfajwroDjfA0dT4TL18XAFYYc4OQ,7265
 magic_pdf/model/sub_modules/ocr/paddleocr/ppocr_291_mod.py,sha256=VouMTvi6M5TV6pQdlpusgfyZapxiZ_Wi7Ff53eMC3rE,8996
 magic_pdf/model/sub_modules/reading_oreder/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/model/sub_modules/reading_oreder/layoutreader/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -121,66 +108,28 @@ magic_pdf/model/sub_modules/table/rapidtable/rapid_table.py,sha256=_FKKOSKeceusx
 magic_pdf/model/sub_modules/table/structeqtable/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/model/sub_modules/table/structeqtable/struct_eqtable.py,sha256=SrNPm-uOFEvN5muFGbXTAuwzXm-rCiaihVdqbydIBIA,1131
 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=AdH3UGu4BEoII0uFjPKUf61W7HmG4fDlWgR1xxMeFlE,2775
+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/block_continuation_processor.py,sha256=IkReB5hirjm0OAirNzQQpnWe2f2bdP3Hug3Ef8qTRDA,22749
-magic_pdf/para/block_termination_processor.py,sha256=YU3ZYqJy9e3OQmOuQYZrR6AUpmAlQ0mhj0PgZZPZ_fM,17957
-magic_pdf/para/commons.py,sha256=VdJ8SY9qJTtcRyx8HH-PFeZSJwL4Tsf50197RD_-dwc,5414
-magic_pdf/para/denoise.py,sha256=J7dM2KNnbdzAd2A3agB04U6L1GL9RrhAs-MLrq-_Ftg,10443
-magic_pdf/para/draw.py,sha256=KyWc03do_WuBKQ028HYzepYwbIkel9ID0uqRhuPVOHc,5643
-magic_pdf/para/exceptions.py,sha256=kpjGxrSZ-drNmoKlmuQ0asTjI8cKKKWsdDDBoDHQP9M,4978
-magic_pdf/para/layout_match_processor.py,sha256=yr4FEO7GJ502udShqGRqIJQ_FQxoa0aG_mhmWd8nLwI,1554
-magic_pdf/para/para_split.py,sha256=z7nYeg86BjZOAdJNMwYKSu51W9evurtl3cy1ZUcQLlw,33222
-magic_pdf/para/para_split_v2.py,sha256=vJJqqMMKbv8D702nODThL-5hjkgZ7Vl2BTmEIdwmmDw,39051
-magic_pdf/para/para_split_v3.py,sha256=atfELVRx-90paAS3nZptgP0qG8UpTTaj3LG_2x3NAlQ,15977
-magic_pdf/para/raw_processor.py,sha256=mHxD9FrdOSXH7NqM41s55URyCyuyACvm9kKtowkIb3k,6317
-magic_pdf/para/stats.py,sha256=-6Pf9Y8jkP1uJOYWiHUjw9Lb-Fb9GY7MHr_ok7x2GX0,9731
-magic_pdf/para/title_processor.py,sha256=pYZv9vEkIjAtCz8jIUtl9AVUy_ib5SdAZmMVoZtsMRI,38593
+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/pipe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/post_proc/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/post_proc/detect_para.py,sha256=5LX86ueHQGOV9CNimAxqZH4R3KTi78leum1de_Na0pw,126181
-magic_pdf/post_proc/pdf_post_filter.py,sha256=3EJDovQPckPKJaBY1wvAty-LGKyRG63WICY_bA_Kfbs,2501
-magic_pdf/post_proc/remove_footnote.py,sha256=701P7xRu6gzLaEHfb2xkYpLZI4CwK2FAo7Ggho4bOTI,7596
 magic_pdf/pre_proc/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/pre_proc/citationmarker_remove.py,sha256=IitOERaK9fGaktsYMyiaaL_71uMIrlG5ZdmpZaR6dsA,6640
-magic_pdf/pre_proc/construct_page_dict.py,sha256=lp3zBmInlWYYIcGC1-NSqT9s44AjDvlnWxDPeZoBVSY,3043
-magic_pdf/pre_proc/cut_image.py,sha256=TghshkDTgdUbyLSbKZoFI9-n-xaFub02IYPyu0IAnRY,2761
-magic_pdf/pre_proc/detect_equation.py,sha256=9omDHKTI8QO9Qd46eVFHWhZeMmTNx7XDuWRgjXI-KFA,6627
-magic_pdf/pre_proc/detect_footer_by_model.py,sha256=_EghAM_zWBcqVY8XBkbSoprKqKUa0mlN1U8YNWxNNLI,2848
-magic_pdf/pre_proc/detect_footer_header_by_statistics.py,sha256=924soXZ51QVpitPgVgnwbC7BqOZI30j5hGW5zP86y-w,11250
-magic_pdf/pre_proc/detect_footnote.py,sha256=UxFuTCRwXdAv3wKCgRQJJVt12hM9O9oPTwzPAChQXoM,8309
-magic_pdf/pre_proc/detect_header.py,sha256=KOmRehgKMuMqNa_2weXkdNSiRVWMFgLMQE4e1itbY7g,2848
-magic_pdf/pre_proc/detect_images.py,sha256=8DwGGTb5IjxqADZDTc_ngwJrTYXxK2qpRqI2FBoPr00,30432
-magic_pdf/pre_proc/detect_page_number.py,sha256=qvYrBbCtBbREvw-MySL_p7byCRvcm1fkLJ5ZB4TP8OM,2848
-magic_pdf/pre_proc/detect_tables.py,sha256=srJzgLVeVuOsqnESqfdJfVukTF84K8qmI5mgFX_BZGs,2800
-magic_pdf/pre_proc/equations_replace.py,sha256=7mexRPwD9C_UJ-SbvO_-XnpcnN7YtGUUznmPjHbjhnw,20526
-magic_pdf/pre_proc/fix_image.py,sha256=5MOfkXc8abfIp49g-68vll40wwTUZ5tcQ2gtsJuFmvs,11486
-magic_pdf/pre_proc/fix_table.py,sha256=20sqJe27fAXcL7_C0qQ9mpsggmH37WuX-wPYWyRgACA,13227
-magic_pdf/pre_proc/main_text_font.py,sha256=1gkjvPuBdKC4oVFkLvnRm2zghsLtVlfAEMKXouyVonM,1048
-magic_pdf/pre_proc/ocr_detect_all_bboxes.py,sha256=DMc2H2xGqVePBReZu5AQbPdvDw3sxOssmujCLlNW3Vs,14143
-magic_pdf/pre_proc/ocr_detect_layout.py,sha256=DW0_HXzmcbW22cXKIYFsyZNFh8mEjSHXIFVjXndJsvQ,5878
-magic_pdf/pre_proc/ocr_dict_merge.py,sha256=Au8y1NBhbWpq_VuPLg3b9dAMUhyPS71xtTghtd21K5M,14273
-magic_pdf/pre_proc/ocr_span_list_modify.py,sha256=9DxEyy1pH87g4T_JEgI3cTVCL2TVrEBl38wsmqhQM4k,12758
-magic_pdf/pre_proc/pdf_pre_filter.py,sha256=qvNlNyj4Mc3qa73mgfkp0PMR-ucABbx3mMcyVipaEpQ,2776
-magic_pdf/pre_proc/post_layout_split.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/ocr_detect_all_bboxes.py,sha256=xQ1L6pwQjN4xBSKEXslheip1aMFaiB0grqlX3BF-kh0,9282
+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
 magic_pdf/pre_proc/remove_bbox_overlap.py,sha256=8eXNdsz9s06LX0kS0AxYSkaY1tWQQMkJfVtVSdjTQNE,3090
-magic_pdf/pre_proc/remove_colored_strip_bbox.py,sha256=WVKhgeWifRdO-u2ETYffkcMOFVYIbiaZu5pMr1RpEdA,4090
-magic_pdf/pre_proc/remove_footer_header.py,sha256=Igdr4jH7BUGuTcapWPiKEGKxhWH12c3VVmX5xwUVn7w,5680
-magic_pdf/pre_proc/remove_rotate_bbox.py,sha256=di7geS7AFhSaAvkWZHT6J3dlXEq8uu9Z4oBYtolQjl0,8803
-magic_pdf/pre_proc/resolve_bbox_conflict.py,sha256=ABl0vo8kkcCPSTI8dpXQTOH1b9R-lbzsJDDFONU6ELk,7313
-magic_pdf/pre_proc/solve_line_alien.py,sha256=aNoQptPcC38Sm1I2ABhgw8jeH_5kjsRHx3VYlFFtm1g,853
-magic_pdf/pre_proc/statistics.py,sha256=_9jGlXq0iXd03UMxB92ZqCiu7cjNkG5vHvFlTF_9ytA,220
 magic_pdf/resources/fasttext-langdetect/lid.176.ftz,sha256=jzRyz-hzintgmejpmcPL-uDc0VaWqsfXc4qAOdtgPoM,938013
 magic_pdf/resources/model_config/model_configs.yaml,sha256=S2BnVQxPd0xsZswn9WqJKTfnqd7ayY5lRwDVifTEAfw,290
 magic_pdf/resources/model_config/UniMERNet/demo.yaml,sha256=Jdaim2D2lAYrV9rhc1X5Sy2_IacGOrfysJhxEUgSElo,827
 magic_pdf/resources/model_config/layoutlmv3/layoutlmv3_base_inference.yaml,sha256=9aNAEYgpHTAWpcUrDvuPG2y4V-Qw8QdcJefi96y8yDU,6109
 magic_pdf/rw/AbsReaderWriter.py,sha256=2H5SDJfAAOX9kPfel06a8VRCHxD1Y8aPbWEkQDdn9JM,452
 magic_pdf/rw/DiskReaderWriter.py,sha256=7ZAekH8V6xlBo_1WeSZ6sNwAj2WGPtjNl50zq1CoMDY,2614
-magic_pdf/rw/S3ReaderWriter.py,sha256=_DmL45Ubio-_VsKD84KrqOQ-VNDUTzcXSrXfNMb5vww,5310
+magic_pdf/rw/S3ReaderWriter.py,sha256=LmbtA-pZlC745nnSUs67C1iqSrBDS7IzE6QC8YMB644,5293
 magic_pdf/rw/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/spark/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/spark/spark_api.py,sha256=BYO6zlRW0cEnIUB3ZzNQTu_LsPHEVitqiUN7gy3x_wo,1124
@@ -190,9 +139,9 @@ magic_pdf/tools/cli_dev.py,sha256=3RbubfTIagWoFYdu8wSDanr-BJDjFGeDet55jTy7He0,39
 magic_pdf/tools/common.py,sha256=ILTv8YjnK-XTVV5nzak3Sm-EJJXjG1hJJghlYKgYVBQ,6809
 magic_pdf/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/utils/annotations.py,sha256=82ou3uELNbQWa9hOFFkVt0gsIskAKf5msCv5J2IJ5V0,211
-magic_pdf-0.10.1.dist-info/LICENSE.md,sha256=jVa0BUaKrRH4erV2P5AeJ24I2WRv9chIGxditreJ6e0,34524
-magic_pdf-0.10.1.dist-info/METADATA,sha256=QdRsUeX9lmB2tTEFLT92qEWnPcgxIu7L0GeqTOHBGms,40300
-magic_pdf-0.10.1.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
-magic_pdf-0.10.1.dist-info/entry_points.txt,sha256=wXwYke3j8fqDQTocUspL-CqDUEv3Tfcwp09fM8dZAhA,98
-magic_pdf-0.10.1.dist-info/top_level.txt,sha256=J9I0AzmHWGkp9c6DL8Oe4mEx3yYphLzkRn4H25Lg1rE,10
-magic_pdf-0.10.1.dist-info/RECORD,,
+magic_pdf-0.10.3.dist-info/LICENSE.md,sha256=jVa0BUaKrRH4erV2P5AeJ24I2WRv9chIGxditreJ6e0,34524
+magic_pdf-0.10.3.dist-info/METADATA,sha256=R86XDaSfj1tcu3etkvhQfg3FSoARv8mKW2KpwjsdqWs,36992
+magic_pdf-0.10.3.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
+magic_pdf-0.10.3.dist-info/entry_points.txt,sha256=wXwYke3j8fqDQTocUspL-CqDUEv3Tfcwp09fM8dZAhA,98
+magic_pdf-0.10.3.dist-info/top_level.txt,sha256=J9I0AzmHWGkp9c6DL8Oe4mEx3yYphLzkRn4H25Lg1rE,10
+magic_pdf-0.10.3.dist-info/RECORD,,