mineru 2.6.6__py3-none-any.whl → 2.6.7__py3-none-any.whl

mineru/backend/vlm/vlm_magic_model.py

@@ -45,7 +45,7 @@ class MagicModel:
                 continue
 
             span_type = "unknown"
-            line_type = None
+            code_block_sub_type = None
             guess_lang = None
 
             if block_type in [
@@ -74,7 +74,7 @@ class MagicModel:
                 span_type = ContentType.TABLE
             elif block_type in ["code", "algorithm"]:
                 block_content = code_content_clean(block_content)
-                line_type = block_type
+                code_block_sub_type = block_type
                 block_type = BlockType.CODE_BODY
                 span_type = ContentType.TEXT
                 guess_lang = guess_language_by_text(block_content)
@@ -82,6 +82,9 @@ class MagicModel:
                 block_type = BlockType.INTERLINE_EQUATION
                 span_type = ContentType.INTERLINE_EQUATION
 
+            #  code 和 algorithm 类型的块，如果内容中包含行内公式，则需要将块类型切换为algorithm
+            switch_code_to_algorithm = False
+
             if span_type in ["image", "table"]:
                 span = {
                     "bbox": block_bbox,
@@ -102,6 +105,8 @@ class MagicModel:
 
                 if block_content and block_content.count("\\(") == block_content.count("\\)") and block_content.count("\\(") > 0:
 
+                    switch_code_to_algorithm = True
+
                     # 生成包含文本和公式的span列表
                     spans = []
                     last_end = 0
@@ -160,7 +165,9 @@ class MagicModel:
 
             # 构造line对象
             if block_type in [BlockType.CODE_BODY]:
-                line = {"bbox": block_bbox, "spans": spans, "extra": {"type": line_type, "guess_lang": guess_lang}}
+                if switch_code_to_algorithm and code_block_sub_type == "code":
+                    code_block_sub_type = "algorithm"
+                line = {"bbox": block_bbox, "spans": spans, "extra": {"type": code_block_sub_type, "guess_lang": guess_lang}}
             else:
                 line = {"bbox": block_bbox, "spans": spans}
 

mineru/backend/vlm/vlm_middle_json_mkcontent.py

@@ -1,7 +1,8 @@
 import os
 
+from loguru import logger
 from mineru.utils.config_reader import get_latex_delimiter_config, get_formula_enable, get_table_enable
-from mineru.utils.enum_class import MakeMode, BlockType, ContentType
+from mineru.utils.enum_class import MakeMode, BlockType, ContentType, ContentTypeV2
 
 latex_delimiters_config = get_latex_delimiter_config()
 
@@ -234,6 +235,277 @@ def make_blocks_to_content_list(para_block, img_buket_path, page_idx, page_size)
 
     return para_content
 
+
+def make_blocks_to_content_list_v2(para_block, img_buket_path, page_size):
+    para_type = para_block['type']
+    para_content = {}
+    if para_type in [
+        BlockType.HEADER,
+        BlockType.FOOTER,
+        BlockType.ASIDE_TEXT,
+        BlockType.PAGE_NUMBER,
+        BlockType.PAGE_FOOTNOTE,
+    ]:
+        if para_type == BlockType.HEADER:
+            content_type = ContentTypeV2.PAGE_HEADER
+        elif para_type == BlockType.FOOTER:
+            content_type = ContentTypeV2.PAGE_FOOTER
+        elif para_type == BlockType.ASIDE_TEXT:
+            content_type = ContentTypeV2.PAGE_ASIDE_TEXT
+        elif para_type == BlockType.PAGE_NUMBER:
+            content_type = ContentTypeV2.PAGE_NUMBER
+        elif para_type == BlockType.PAGE_FOOTNOTE:
+            content_type = ContentTypeV2.PAGE_FOOTNOTE
+        else:
+            raise ValueError(f"Unknown para_type: {para_type}")
+        para_content = {
+            'type': content_type,
+            'content': {
+                f"{content_type}_content": merge_para_with_text_v2(para_block),
+            }
+        }
+    elif para_type == BlockType.TITLE:
+        title_level = get_title_level(para_block)
+        if title_level != 0:
+            para_content = {
+                'type': ContentTypeV2.TITLE,
+                'content': {
+                    "title_content": merge_para_with_text_v2(para_block),
+                    "level": title_level
+                }
+            }
+        else:
+            para_content = {
+                'type': ContentTypeV2.PARAGRAPH,
+                'content': {
+                    "paragraph_content": merge_para_with_text_v2(para_block),
+                }
+            }
+    elif para_type in [
+        BlockType.TEXT,
+        BlockType.PHONETIC
+    ]:
+        para_content = {
+            'type': ContentTypeV2.PARAGRAPH,
+            'content': {
+                'paragraph_content': merge_para_with_text_v2(para_block),
+            }
+        }
+    elif para_type == BlockType.INTERLINE_EQUATION:
+        image_path, math_content = get_body_data(para_block)
+        para_content = {
+            'type': ContentTypeV2.EQUATION_INTERLINE,
+            'content': {
+                'math_content': math_content,
+                'math_type': 'latex',
+                'image_source': {'path': f"{img_buket_path}/{image_path}"},
+            }
+        }
+    elif para_type == BlockType.IMAGE:
+        image_caption = []
+        image_footnote = []
+        image_path, _ = get_body_data(para_block)
+        image_source = {
+            'path': f"{img_buket_path}/{image_path}",
+        }
+        for block in para_block['blocks']:
+            if block['type'] == BlockType.IMAGE_CAPTION:
+                image_caption.extend(merge_para_with_text_v2(block))
+            if block['type'] == BlockType.IMAGE_FOOTNOTE:
+                image_footnote.extend(merge_para_with_text_v2(block))
+        para_content = {
+            'type': ContentTypeV2.IMAGE,
+            'content': {
+                'image_source': image_source,
+                'image_caption': image_caption,
+                'image_footnote': image_footnote,
+            }
+        }
+    elif para_type == BlockType.TABLE:
+        table_caption = []
+        table_footnote = []
+        image_path, html = get_body_data(para_block)
+        image_source = {
+            'path': f"{img_buket_path}/{image_path}",
+        }
+        if html.count("<table") > 1:
+            table_nest_level = 2
+        else:
+            table_nest_level = 1
+        if (
+                "colspan" in html or
+                "rowspan" in html or
+                table_nest_level > 1
+        ):
+            table_type = ContentTypeV2.TABLE_COMPLEX
+        else:
+            table_type = ContentTypeV2.TABLE_SIMPLE
+
+        for block in para_block['blocks']:
+            if block['type'] == BlockType.TABLE_CAPTION:
+                table_caption.extend(merge_para_with_text_v2(block))
+            if block['type'] == BlockType.TABLE_FOOTNOTE:
+                table_footnote.extend(merge_para_with_text_v2(block))
+        para_content = {
+            'type': ContentTypeV2.TABLE,
+            'content': {
+                'image_source': image_source,
+                'table_caption': table_caption,
+                'table_footnote': table_footnote,
+                'html': html,
+                'table_type': table_type,
+                'table_nest_level': table_nest_level,
+            }
+        }
+    elif para_type == BlockType.CODE:
+        code_caption = []
+        code_content = []
+        for block in para_block['blocks']:
+            if block['type'] == BlockType.CODE_CAPTION:
+                code_caption.extend(merge_para_with_text_v2(block))
+            if block['type'] == BlockType.CODE_BODY:
+                code_content = merge_para_with_text_v2(block)
+        sub_type = para_block["sub_type"]
+        if sub_type == BlockType.CODE:
+            para_content = {
+                'type': ContentTypeV2.CODE,
+                'content': {
+                    'code_caption': code_caption,
+                    'code_content': code_content,
+                    'code_language': para_block.get('guess_lang', 'txt'),
+                }
+            }
+        elif sub_type == BlockType.ALGORITHM:
+            para_content = {
+                'type': ContentTypeV2.ALGORITHM,
+                'content': {
+                    'algorithm_caption': code_caption,
+                    'algorithm_content': code_content,
+                }
+            }
+        else:
+            raise ValueError(f"Unknown code sub_type: {sub_type}")
+    elif para_type == BlockType.REF_TEXT:
+        para_content = {
+            'type': ContentTypeV2.LIST,
+            'content': {
+                'list_type': ContentTypeV2.LIST_REF,
+                'list_items': [
+                    {
+                        'item_type': 'text',
+                        'item_content': merge_para_with_text_v2(para_block),
+                    }
+                ],
+            }
+        }
+    elif para_type == BlockType.LIST:
+        if 'sub_type' in para_block:
+            if para_block['sub_type'] == BlockType.REF_TEXT:
+                list_type = ContentTypeV2.LIST_REF
+            elif para_block['sub_type'] == BlockType.TEXT:
+                list_type = ContentTypeV2.LIST_TEXT
+            else:
+                raise ValueError(f"Unknown list sub_type: {para_block['sub_type']}")
+        else:
+            list_type = ContentTypeV2.LIST_TEXT
+        list_items = []
+        for block in para_block['blocks']:
+            item_content = merge_para_with_text_v2(block)
+            if item_content:
+                list_items.append({
+                    'item_type': 'text',
+                    'item_content': item_content,
+                })
+        para_content = {
+            'type': ContentTypeV2.LIST,
+            'content': {
+                'list_type': list_type,
+                'list_items': list_items,
+            }
+        }
+
+    page_width, page_height = page_size
+    para_bbox = para_block.get('bbox')
+    if para_bbox:
+        x0, y0, x1, y1 = para_bbox
+        para_content['bbox'] = [
+            int(x0 * 1000 / page_width),
+            int(y0 * 1000 / page_height),
+            int(x1 * 1000 / page_width),
+            int(y1 * 1000 / page_height),
+        ]
+
+    return para_content
+
+
+
+
+
+def get_body_data(para_block):
+    """
+    Extract image_path and html from para_block
+    Returns:
+        - For IMAGE/INTERLINE_EQUATION: (image_path, '')
+        - For TABLE: (image_path, html)
+        - Default: ('', '')
+    """
+
+    def get_data_from_spans(lines):
+        for line in lines:
+            for span in line.get('spans', []):
+                span_type = span.get('type')
+                if span_type == ContentType.TABLE:
+                    return span.get('image_path', ''), span.get('html', '')
+                elif span_type == ContentType.IMAGE:
+                    return span.get('image_path', ''), ''
+                elif span_type == ContentType.INTERLINE_EQUATION:
+                    return span.get('image_path', ''), span.get('content', '')
+                elif span_type == ContentType.TEXT:
+                    return '', span.get('content', '')
+        return '', ''
+
+    # 处理嵌套的 blocks 结构
+    if 'blocks' in para_block:
+        for block in para_block['blocks']:
+            block_type = block.get('type')
+            if block_type in [BlockType.IMAGE_BODY, BlockType.TABLE_BODY, BlockType.CODE_BODY]:
+                result = get_data_from_spans(block.get('lines', []))
+                if result != ('', ''):
+                    return result
+        return '', ''
+
+    # 处理直接包含 lines 的结构
+    return get_data_from_spans(para_block.get('lines', []))
+
+
+def merge_para_with_text_v2(para_block):
+    para_content = []
+    para_type = para_block['type']
+    for line in para_block['lines']:
+        for span in line['spans']:
+            span_type = span['type']
+            if span['content']:
+                if para_type == BlockType.PHONETIC and span_type == ContentTypeV2.SPAN_TEXT:
+                    span_type = ContentTypeV2.SPAN_PHONETIC
+                if span_type == ContentType.INLINE_EQUATION:
+                    span_type = ContentTypeV2.SPAN_EQUATION_INLINE
+                if span_type in [
+                    ContentTypeV2.SPAN_TEXT,
+                    ContentTypeV2.SPAN_PHONETIC,
+                    ContentTypeV2.SPAN_EQUATION_INLINE,
+                    ContentTypeV2.SPAN_MD,
+                    ContentTypeV2.SPAN_CODE_INLINE,
+                ]:
+                    span_content = {
+                        'type': span_type,
+                        'content': span['content'],
+                    }
+                    para_content.append(span_content)
+                else:
+                    logger.warning(f"Unknown span type in merge_para_with_text_v2: {span_type}")
+    return para_content
+
+
 def union_make(pdf_info_dict: list,
                make_mode: str,
                img_buket_path: str = '',
@@ -260,10 +532,20 @@ def union_make(pdf_info_dict: list,
             for para_block in para_blocks:
                 para_content = make_blocks_to_content_list(para_block, img_buket_path, page_idx, page_size)
                 output_content.append(para_content)
+        elif make_mode == MakeMode.CONTENT_LIST_V2:
+            # https://github.com/drunkpig/llm-webkit-mirror/blob/dev6/docs/specification/output_format/content_list_spec.md
+            page_contents = []
+            para_blocks = (paras_of_layout or []) + (paras_of_discarded or [])
+            if not para_blocks:
+                continue
+            for para_block in para_blocks:
+                para_content = make_blocks_to_content_list_v2(para_block, img_buket_path, page_size)
+                page_contents.append(para_content)
+            output_content.append(page_contents)
 
     if make_mode in [MakeMode.MM_MD, MakeMode.NLP_MD]:
         return '\n\n'.join(output_content)
-    elif make_mode == MakeMode.CONTENT_LIST:
+    elif make_mode in [MakeMode.CONTENT_LIST, MakeMode.CONTENT_LIST_V2]:
         return output_content
     return None
 

mineru/cli/common.py

@@ -144,6 +144,13 @@ def _process_output(
             f"{pdf_file_name}_content_list.json",
             json.dumps(content_list, ensure_ascii=False, indent=4),
         )
+        if not is_pipeline:
+            content_list_v2 = make_func(pdf_info, MakeMode.CONTENT_LIST_V2, image_dir)
+            md_writer.write_string(
+                f"{pdf_file_name}_content_list_v2.json",
+                json.dumps(content_list_v2, ensure_ascii=False, indent=4),
+            )
+
 
     if f_dump_middle_json:
         md_writer.write_string(

mineru/utils/enum_class.py

@@ -38,6 +38,31 @@ class ContentType:
     CODE = 'code'
 
 
+class ContentTypeV2:
+    CODE = 'code'
+    ALGORITHM = "algorithm"
+    EQUATION_INTERLINE = 'equation_interline'
+    IMAGE = 'image'
+    TABLE = 'table'
+    TABLE_SIMPLE = 'simple_table'
+    TABLE_COMPLEX = 'complex_table'
+    LIST = 'list'
+    LIST_TEXT = 'text_list'
+    LIST_REF = 'reference_list'
+    TITLE = 'title'
+    PARAGRAPH = 'paragraph'
+    SPAN_TEXT = 'text'
+    SPAN_EQUATION_INLINE = 'equation_inline'
+    SPAN_PHONETIC = 'phonetic'
+    SPAN_MD = 'md'
+    SPAN_CODE_INLINE = 'code_inline'
+    PAGE_HEADER = "page_header"
+    PAGE_FOOTER = "page_footer"
+    PAGE_NUMBER = "page_number"
+    PAGE_ASIDE_TEXT = "page_aside_text"
+    PAGE_FOOTNOTE = "page_footnote"
+
+
 class CategoryId:
     Title = 0
     Text = 1
@@ -60,6 +85,7 @@ class MakeMode:
     MM_MD = 'mm_markdown'
     NLP_MD = 'nlp_markdown'
     CONTENT_LIST = 'content_list'
+    CONTENT_LIST_V2 = 'content_list_v2'
 
 
 class ModelPath:

mineru/utils/guess_suffix_or_lang.py

@@ -1,9 +1,11 @@
 from pathlib import Path
 
+from loguru import logger
 from magika import Magika
 
 
 DEFAULT_LANG = "txt"
+PDF_SIG_BYTES = b'%PDF'
 magika = Magika()
 
 def guess_language_by_text(code):
@@ -14,7 +16,7 @@ def guess_language_by_text(code):
 
 def guess_suffix_by_bytes(file_bytes, file_path=None) -> str:
     suffix = magika.identify_bytes(file_bytes).prediction.output.label
-    if file_path and suffix in ["ai"] and Path(file_path).suffix.lower() in [".pdf"]:
+    if file_path and suffix in ["ai", "html"] and Path(file_path).suffix.lower() in [".pdf"] and file_bytes[:4] == PDF_SIG_BYTES:
         suffix = "pdf"
     return suffix
 
@@ -23,6 +25,11 @@ def guess_suffix_by_path(file_path) -> str:
     if not isinstance(file_path, Path):
         file_path = Path(file_path)
     suffix = magika.identify_path(file_path).prediction.output.label
-    if suffix in ["ai"] and file_path.suffix.lower() in [".pdf"]:
-        suffix = "pdf"
+    if suffix in ["ai", "html"] and file_path.suffix.lower() in [".pdf"]:
+        try:
+            with open(file_path, 'rb') as f:
+                if f.read(4) == PDF_SIG_BYTES:
+                    suffix = "pdf"
+        except Exception as e:
+            logger.warning(f"Failed to read file {file_path} for PDF signature check: {e}")
     return suffix

mineru/utils/table_merge.py

@@ -7,6 +7,9 @@ from mineru.backend.vlm.vlm_middle_json_mkcontent import merge_para_with_text
 from mineru.utils.enum_class import BlockType, SplitFlag
 
 
+CONTINUATION_MARKERS = ["(续)", "(续表)", "(continued)", "(cont.)"]
+
+
 def full_to_half(text: str) -> str:
     """Convert full-width characters to half-width characters using code point manipulation.
 
@@ -174,8 +177,13 @@ def can_merge_tables(current_table_block, previous_table_block):
     # 如果有TABLE_CAPTION类型的块,检查是否至少有一个以"(续)"结尾
     caption_blocks = [block for block in current_table_block["blocks"] if block["type"] == BlockType.TABLE_CAPTION]
     if caption_blocks:
-        # 如果所有caption都不以"(续)"结尾,则不合并
-        if not any(full_to_half(merge_para_with_text(block).strip()).endswith("(续)") for block in caption_blocks):
+        # 如果所有caption都不以"(续)"、"(续表)"、"(continued)"或"(cont.)"结尾,则不合并
+
+        if not any(
+                any(full_to_half(merge_para_with_text(block).strip()).lower().endswith(marker.lower())
+                    for marker in CONTINUATION_MARKERS)
+                for block in caption_blocks
+        ):
             return False, None, None, None, None
 
     if any(block["type"] == BlockType.TABLE_FOOTNOTE for block in previous_table_block["blocks"]):

mineru/version.py

@@ -1 +1 @@
-__version__ = "2.6.6"
+__version__ = "2.6.7"

{mineru-2.6.6.dist-info → mineru-2.6.7.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mineru
-Version: 2.6.6
+Version: 2.6.7
 Summary: A practical tool for converting PDF to Markdown
 License: AGPL-3.0
 Project-URL: homepage, https://mineru.net/
@@ -135,6 +135,9 @@ Dynamic: license-file
 
 # Changelog
 
+- 2025/12/12 2.6.7 Release
+  - Bug fix: #4168
+
 - 2025/12/02 2.6.6 Release
   - `mineru-api` tool optimizations
     - Added descriptive text to `mineru-api` interface parameters to improve API documentation readability.

{mineru-2.6.6.dist-info → mineru-2.6.7.dist-info}/RECORD

@@ -1,5 +1,5 @@
 mineru/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
-mineru/version.py,sha256=MJHGx-Qo0nycI7WHSavnK8Mok6HS_De_qLfGWXih6Og,22
+mineru/version.py,sha256=Kc2cyTvfIdfaMxraNhj61Hw9Mr3C-eTCdQPLK5KH1AQ,22
 mineru/backend/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
 mineru/backend/utils.py,sha256=GLJU3IznDmhE1_qNmkU1UOtsuskIHBezgsEVO6Uar-Y,698
 mineru/backend/pipeline/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
@@ -15,11 +15,11 @@ mineru/backend/vlm/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5
 mineru/backend/vlm/model_output_to_middle_json.py,sha256=AqYX44gS9crUO_t7SuUatD71EVjow6pI6yA2Ik3gQ0s,5139
 mineru/backend/vlm/utils.py,sha256=JMgS3SMFcHJYH2jIx-Xhs-P2a1bmT8U6Kn60IL0OmQA,3570
 mineru/backend/vlm/vlm_analyze.py,sha256=wP3vuYGVec0hRsDAuzfSm2HD4Muu7wSWL767qxd_yqw,11690
-mineru/backend/vlm/vlm_magic_model.py,sha256=Pd0sOr7G1crAJIVeq6h_03gNSuxmV5U8dvGTGT_rrjs,23452
-mineru/backend/vlm/vlm_middle_json_mkcontent.py,sha256=5V-AU9KkxxMn0DDSQBrb15I4GVpEyiQy8uNI_tQhS6M,13498
+mineru/backend/vlm/vlm_magic_model.py,sha256=e8BWkfeRkZNJjFdm9oPmvIs9ATjdzCy_5OZw2qTziZA,23839
+mineru/backend/vlm/vlm_middle_json_mkcontent.py,sha256=9ig7kfbKDrG5QOwvYoXfcsK5EY9_VyAFmqz733Il-P0,24071
 mineru/cli/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
 mineru/cli/client.py,sha256=XSEIr4klUuufMAWn5IioZdXpg1xAxqRZF0HkaVIhxh0,6815
-mineru/cli/common.py,sha256=zhNOJCOnTSMbWdUWSZG-nf0odv5vBRtdZYZ1UbUPH3g,14369
+mineru/cli/common.py,sha256=Rld3P4FzbTYngy3BP-Irto9aMS7EPq2lIWgeIF1xPMc,14667
 mineru/cli/fast_api.py,sha256=lLxQKKHmD8ruoZGcE6LrXzr3pQIxvw8OdJrQq_FNLSM,14447
 mineru/cli/gradio_app.py,sha256=EUPuRHHCOECrE3E3VNEeuMDYeC3nicurOYfk8YJSOMw,15646
 mineru/cli/models_download.py,sha256=LNfoIpUlJM7m7qb2SiCxtjMDw4jILBQtZwNP2JoY81U,4815
@@ -159,9 +159,9 @@ mineru/utils/cli_parser.py,sha256=4seFAu1kulsYnw6WM2q_cxgEOt2tErZVkI-LNEF_kGw,14
 mineru/utils/config_reader.py,sha256=IRVWTpBnbnRpck6eXZUKw-fcLt7hon5S4uqWW-RBb1w,4075
 mineru/utils/cut_image.py,sha256=g3m4nfcJNWlxi-P0kpXTtlmspXkMcLCfGwmYuQ-Z2hE,751
 mineru/utils/draw_bbox.py,sha256=FkgppjUzRhN-uxvChdkhHXcDavJEaApMD6qC6qoRwfQ,20292
-mineru/utils/enum_class.py,sha256=-_Ey03vGNEQHkl6x7pZ43GgrakwhSCOa1RXdr1m-I3A,2503
+mineru/utils/enum_class.py,sha256=NO9FiM2Ni6-SI1rzMm5X98ssNDxgJTyykZddJSscgJs,3221
 mineru/utils/format_utils.py,sha256=2s89vHcSISjuolk8Hvg3K-5-rRbiT3Us7eFLzUKrNKs,10233
-mineru/utils/guess_suffix_or_lang.py,sha256=nznyQpUn1BSA8JNw9HuG3pVV-xtVAtrtcGuHZ-VXt9M,856
+mineru/utils/guess_suffix_or_lang.py,sha256=aUC2wAJwa5LH0SHxwTbOEJqVVgvpdUCWFF6of8eDVkc,1198
 mineru/utils/hash_utils.py,sha256=UPS_8NRBmVumdyOv16Lmv6Ly2xK8OVDJEe5gG6gKIFk,857
 mineru/utils/language.py,sha256=7RT3mxSa7jdpoC5ySd7ZddHA7TO7UsnmDOWiYZAxuyg,1433
 mineru/utils/llm_aided.py,sha256=9WUytvxenSAuaWR4sTQhVPQ5h8pY0wVOH1O2sj_6dLs,5149
@@ -178,10 +178,10 @@ mineru/utils/pdf_text_tool.py,sha256=KEztjfdqsIHHuiTEAMAL7Lr1OS3R7Ur-uTqGiCRjReQ
 mineru/utils/run_async.py,sha256=rPeP4BCZerR8VByRDhiYzfZiahLVqoZEBVAS54dAjNg,1286
 mineru/utils/span_block_fix.py,sha256=0eVQjJCrT03woRt9hoh6Uu42Tp1dacfGTv2x3B9qq94,8797
 mineru/utils/span_pre_proc.py,sha256=h41q2uQajI0xQbc_30hqaju1dv3oVYxBAlKgURl8HIc,13692
-mineru/utils/table_merge.py,sha256=d98zNbM1ZQ8V1kUt6RugParNUNPv7DGL-XKIzR3iJVQ,15360
-mineru-2.6.6.dist-info/licenses/LICENSE.md,sha256=jVa0BUaKrRH4erV2P5AeJ24I2WRv9chIGxditreJ6e0,34524
-mineru-2.6.6.dist-info/METADATA,sha256=9f-9lcSQXdLCxbYmHItJbLgDc-TZG7u7dVUWMS0SzXA,73095
-mineru-2.6.6.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-mineru-2.6.6.dist-info/entry_points.txt,sha256=JbtrCPhx1T32s7TONUsteKg-24ZwRT1HSiFtW5jypVw,376
-mineru-2.6.6.dist-info/top_level.txt,sha256=zuGQfZcbsHv4I4oKI9gaKPqEWBFm6xJroKuug2LnKP8,7
-mineru-2.6.6.dist-info/RECORD,,
+mineru/utils/table_merge.py,sha256=pZHP0mIUcWPcdu0HecZDEFZ_ms_rT7C43m0Eg2iEEF4,15592
+mineru-2.6.7.dist-info/licenses/LICENSE.md,sha256=jVa0BUaKrRH4erV2P5AeJ24I2WRv9chIGxditreJ6e0,34524
+mineru-2.6.7.dist-info/METADATA,sha256=2SEMOR8-0ZzJL85pLQR4Gkm61YroeJ6BSDjB44YCUhs,73142
+mineru-2.6.7.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+mineru-2.6.7.dist-info/entry_points.txt,sha256=JbtrCPhx1T32s7TONUsteKg-24ZwRT1HSiFtW5jypVw,376
+mineru-2.6.7.dist-info/top_level.txt,sha256=zuGQfZcbsHv4I4oKI9gaKPqEWBFm6xJroKuug2LnKP8,7
+mineru-2.6.7.dist-info/RECORD,,