magic-pdf 1.2.0__py3-none-any.whl → 1.2.2__py3-none-any.whl

magic_pdf/dict2md/ocr_mkcontent.py

@@ -138,12 +138,9 @@ def full_to_half(text: str) -> str:
     result = []
     for char in text:
         code = ord(char)
-        # Full-width ASCII variants (FF01-FF5E)
-        if 0xFF01 <= code <= 0xFF5E:
+        # Full-width letters and numbers (FF21-FF3A for A-Z, FF41-FF5A for a-z, FF10-FF19 for 0-9)
+        if (0xFF21 <= code <= 0xFF3A) or (0xFF41 <= code <= 0xFF5A) or (0xFF10 <= code <= 0xFF19):
             result.append(chr(code - 0xFEE0))  # Shift to ASCII range
-        # Full-width space
-        elif code == 0x3000:
-            result.append(' ')
         else:
             result.append(char)
     return ''.join(result)

magic_pdf/libs/performance_stats.py

@@ -0,0 +1,54 @@
+import time
+import functools
+from collections import defaultdict
+from typing import Dict, List
+
+
+class PerformanceStats:
+    """性能统计类，用于收集和展示方法执行时间"""
+
+    _stats: Dict[str, List[float]] = defaultdict(list)
+
+    @classmethod
+    def add_execution_time(cls, func_name: str, execution_time: float):
+        """添加执行时间记录"""
+        cls._stats[func_name].append(execution_time)
+
+    @classmethod
+    def get_stats(cls) -> Dict[str, dict]:
+        """获取统计结果"""
+        results = {}
+        for func_name, times in cls._stats.items():
+            results[func_name] = {
+                'count': len(times),
+                'total_time': sum(times),
+                'avg_time': sum(times) / len(times),
+                'min_time': min(times),
+                'max_time': max(times)
+            }
+        return results
+
+    @classmethod
+    def print_stats(cls):
+        """打印统计结果"""
+        stats = cls.get_stats()
+        print("\n性能统计结果:")
+        print("-" * 80)
+        print(f"{'方法名':<40} {'调用次数':>8} {'总时间(s)':>12} {'平均时间(s)':>12}")
+        print("-" * 80)
+        for func_name, data in stats.items():
+            print(f"{func_name:<40} {data['count']:8d} {data['total_time']:12.6f} {data['avg_time']:12.6f}")
+
+
+def measure_time(func):
+    """测量方法执行时间的装饰器"""
+
+    @functools.wraps(func)
+    def wrapper(*args, **kwargs):
+        start_time = time.time()
+        result = func(*args, **kwargs)
+        execution_time = time.time() - start_time
+        PerformanceStats.add_execution_time(func.__name__, execution_time)
+        return result
+
+    return wrapper

magic_pdf/libs/version.py

@@ -1 +1 @@
-__version__ = "1.2.0"
+__version__ = "1.2.2"

magic_pdf/model/doc_analyze_by_custom_model.py

@@ -170,11 +170,7 @@ def doc_analyze(
         gpu_memory = int(os.getenv("VIRTUAL_VRAM_SIZE", round(get_vram(device))))
         if gpu_memory is not None and gpu_memory >= 8:
 
-            if gpu_memory >= 40:
-                batch_ratio = 32
-            elif gpu_memory >=20:
-                batch_ratio = 16
-            elif gpu_memory >= 16:
+            if gpu_memory >= 16:
                 batch_ratio = 8
             elif gpu_memory >= 10:
                 batch_ratio = 4

magic_pdf/model/magic_model.py

@@ -528,14 +528,13 @@ class MagicModel:
             pair_dis = bbox_distance(subjects[sub_idx]['bbox'], objects[obj_idx]['bbox'])
             nearest_dis = float('inf')
             for i in range(N):
-                if i in seen_idx:continue
+                if i in seen_idx or i == sub_idx:continue
                 nearest_dis = min(nearest_dis, bbox_distance(subjects[i]['bbox'], objects[obj_idx]['bbox']))
 
             if pair_dis >= 3*nearest_dis:
                 seen_idx.add(sub_idx)
                 continue
 
-
             seen_idx.add(sub_idx)
             seen_idx.add(obj_idx + OBJ_IDX_OFFSET)
             seen_sub_idx.add(sub_idx)

magic_pdf/model/sub_modules/mfr/unimernet/Unimernet.py

@@ -100,20 +100,61 @@ class UnimernetModel(object):
             res["latex"] = latex_rm_whitespace(latex)
         return formula_list
 
-    def batch_predict(
-        self, images_mfd_res: list, images: list, batch_size: int = 64
-    ) -> list:
+    # def batch_predict(
+    #     self, images_mfd_res: list, images: list, batch_size: int = 64
+    # ) -> list:
+    #     images_formula_list = []
+    #     mf_image_list = []
+    #     backfill_list = []
+    #     for image_index in range(len(images_mfd_res)):
+    #         mfd_res = images_mfd_res[image_index]
+    #         pil_img = Image.fromarray(images[image_index])
+    #         formula_list = []
+    #
+    #         for xyxy, conf, cla in zip(
+    #             mfd_res.boxes.xyxy, mfd_res.boxes.conf, mfd_res.boxes.cls
+    #         ):
+    #             xmin, ymin, xmax, ymax = [int(p.item()) for p in xyxy]
+    #             new_item = {
+    #                 "category_id": 13 + int(cla.item()),
+    #                 "poly": [xmin, ymin, xmax, ymin, xmax, ymax, xmin, ymax],
+    #                 "score": round(float(conf.item()), 2),
+    #                 "latex": "",
+    #             }
+    #             formula_list.append(new_item)
+    #             bbox_img = pil_img.crop((xmin, ymin, xmax, ymax))
+    #             mf_image_list.append(bbox_img)
+    #
+    #         images_formula_list.append(formula_list)
+    #         backfill_list += formula_list
+    #
+    #     dataset = MathDataset(mf_image_list, transform=self.mfr_transform)
+    #     dataloader = DataLoader(dataset, batch_size=batch_size, num_workers=0)
+    #     mfr_res = []
+    #     for mf_img in dataloader:
+    #         mf_img = mf_img.to(self.device)
+    #         with torch.no_grad():
+    #             output = self.model.generate({"image": mf_img})
+    #         mfr_res.extend(output["pred_str"])
+    #     for res, latex in zip(backfill_list, mfr_res):
+    #         res["latex"] = latex_rm_whitespace(latex)
+    #     return images_formula_list
+
+    def batch_predict(self, images_mfd_res: list, images: list, batch_size: int = 64) -> list:
         images_formula_list = []
         mf_image_list = []
         backfill_list = []
+        image_info = []  # Store (area, original_index, image) tuples
+
+        # Collect images with their original indices
         for image_index in range(len(images_mfd_res)):
             mfd_res = images_mfd_res[image_index]
             pil_img = Image.fromarray(images[image_index])
             formula_list = []
 
-            for xyxy, conf, cla in zip(
-                mfd_res.boxes.xyxy, mfd_res.boxes.conf, mfd_res.boxes.cls
-            ):
+            for idx, (xyxy, conf, cla) in enumerate(zip(
+                    mfd_res.boxes.xyxy, mfd_res.boxes.conf, mfd_res.boxes.cls
+            )):
                 xmin, ymin, xmax, ymax = [int(p.item()) for p in xyxy]
                 new_item = {
                     "category_id": 13 + int(cla.item()),
@@ -123,19 +164,43 @@ class UnimernetModel(object):
                 }
                 formula_list.append(new_item)
                 bbox_img = pil_img.crop((xmin, ymin, xmax, ymax))
+                area = (xmax - xmin) * (ymax - ymin)
+
+                curr_idx = len(mf_image_list)
+                image_info.append((area, curr_idx, bbox_img))
                 mf_image_list.append(bbox_img)
 
             images_formula_list.append(formula_list)
             backfill_list += formula_list
 
-        dataset = MathDataset(mf_image_list, transform=self.mfr_transform)
+        # Stable sort by area
+        image_info.sort(key=lambda x: x[0])  # sort by area
+        sorted_indices = [x[1] for x in image_info]
+        sorted_images = [x[2] for x in image_info]
+
+        # Create mapping for results
+        index_mapping = {new_idx: old_idx for new_idx, old_idx in enumerate(sorted_indices)}
+
+        # Create dataset with sorted images
+        dataset = MathDataset(sorted_images, transform=self.mfr_transform)
         dataloader = DataLoader(dataset, batch_size=batch_size, num_workers=0)
+
+        # Process batches and store results
         mfr_res = []
         for mf_img in dataloader:
             mf_img = mf_img.to(self.device)
             with torch.no_grad():
                 output = self.model.generate({"image": mf_img})
             mfr_res.extend(output["pred_str"])
-        for res, latex in zip(backfill_list, mfr_res):
-            res["latex"] = latex_rm_whitespace(latex)
+
+        # Restore original order
+        unsorted_results = [""] * len(mfr_res)
+        for new_idx, latex in enumerate(mfr_res):
+            original_idx = index_mapping[new_idx]
+            unsorted_results[original_idx] = latex_rm_whitespace(latex)
+
+        # Fill results back
+        for res, latex in zip(backfill_list, unsorted_results):
+            res["latex"] = latex
+
         return images_formula_list

magic_pdf/pdf_parse_union_core_v2.py

@@ -21,9 +21,12 @@ from magic_pdf.libs.config_reader import get_local_layoutreader_model_dir, get_l
 from magic_pdf.libs.convert_utils import dict_to_list
 from magic_pdf.libs.hash_utils import compute_md5
 from magic_pdf.libs.pdf_image_tools import cut_image_to_pil_image
+from magic_pdf.libs.performance_stats import measure_time, PerformanceStats
 from magic_pdf.model.magic_model import MagicModel
 from magic_pdf.post_proc.llm_aided import llm_aided_formula, llm_aided_text, llm_aided_title
 
+from concurrent.futures import ThreadPoolExecutor
+
 try:
     import torchtext
 
@@ -215,7 +218,7 @@ def calculate_contrast(img, img_mode) -> float:
     # logger.info(f"contrast: {contrast}")
     return round(contrast, 2)
 
-
+# @measure_time
 def txt_spans_extract_v2(pdf_page, spans, all_bboxes, all_discarded_blocks, lang):
     # cid用0xfffd表示，连字符拆开
     # text_blocks_raw = pdf_page.get_text('rawdict', flags=fitz.TEXT_PRESERVE_WHITESPACE | fitz.TEXT_MEDIABOX_CLIP)['blocks']
@@ -489,7 +492,7 @@ def insert_lines_into_block(block_bbox, line_height, page_w, page_h):
     else:
         return [[x0, y0, x1, y1]]
 
-
+# @measure_time
 def sort_lines_by_model(fix_blocks, page_w, page_h, line_height):
     page_line_list = []
 
@@ -923,7 +926,6 @@ def pdf_parse_union(
     magic_model = MagicModel(model_list, dataset)
 
     """根据输入的起始范围解析pdf"""
-    # end_page_id = end_page_id if end_page_id else len(pdf_docs) - 1
     end_page_id = (
         end_page_id
         if end_page_id is not None and end_page_id >= 0
@@ -960,6 +962,8 @@ def pdf_parse_union(
             )
         pdf_info_dict[f'page_{page_id}'] = page_info
 
+    # PerformanceStats.print_stats()
+
     """分段"""
     para_split(pdf_info_dict)
 

magic_pdf/post_proc/para_split_v3.py

@@ -108,29 +108,32 @@ def __is_list_or_index_block(block):
         ):
             multiple_para_flag = True
 
-        for line in block['lines']:
-            line_mid_x = (line['bbox'][0] + line['bbox'][2]) / 2
-            block_mid_x = (block['bbox_fs'][0] + block['bbox_fs'][2]) / 2
-            if (
-                line['bbox'][0] - block['bbox_fs'][0] > 0.7 * line_height
-                and block['bbox_fs'][2] - line['bbox'][2] > 0.7 * line_height
-            ):
-                external_sides_not_close_num += 1
-            if abs(line_mid_x - block_mid_x) < line_height / 2:
-                center_close_num += 1
+        block_text = ''
 
+        for line in block['lines']:
             line_text = ''
 
             for span in line['spans']:
                 span_type = span['type']
                 if span_type == ContentType.Text:
                     line_text += span['content'].strip()
-
             # 添加所有文本，包括空行，保持与block['lines']长度一致
             lines_text_list.append(line_text)
             block_text = ''.join(lines_text_list)
-            block_lang = detect_lang(block_text)
-            # logger.info(f"block_lang: {block_lang}")
+
+        block_lang = detect_lang(block_text)
+        # logger.info(f"block_lang: {block_lang}")
+
+        for line in block['lines']:
+            line_mid_x = (line['bbox'][0] + line['bbox'][2]) / 2
+            block_mid_x = (block['bbox_fs'][0] + block['bbox_fs'][2]) / 2
+            if (
+                line['bbox'][0] - block['bbox_fs'][0] > 0.7 * line_height
+                and block['bbox_fs'][2] - line['bbox'][2] > 0.7 * line_height
+            ):
+                external_sides_not_close_num += 1
+            if abs(line_mid_x - block_mid_x) < line_height / 2:
+                center_close_num += 1
 
             # 计算line左侧顶格数量是否大于2，是否顶格用abs(block['bbox_fs'][0] - line['bbox'][0]) < line_height/2 来判断
             if abs(block['bbox_fs'][0] - line['bbox'][0]) < line_height / 2:

magic_pdf/pre_proc/ocr_dict_merge.py

@@ -64,7 +64,7 @@ def span_block_type_compatible(span_type, block_type):
     if span_type in [ContentType.Text, ContentType.InlineEquation]:
         return block_type in [BlockType.Text, BlockType.Title, BlockType.ImageCaption, BlockType.ImageFootnote, BlockType.TableCaption, BlockType.TableFootnote]
     elif span_type == ContentType.InterlineEquation:
-        return block_type in [BlockType.InterlineEquation]
+        return block_type in [BlockType.InterlineEquation, BlockType.Text]
     elif span_type == ContentType.Image:
         return block_type in [BlockType.ImageBody]
     elif span_type == ContentType.Table:

{magic_pdf-1.2.0.dist-info → magic_pdf-1.2.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: magic-pdf
-Version: 1.2.0
+Version: 1.2.2
 Summary: A practical tool for converting PDF to Markdown
 Home-page: https://github.com/opendatalab/MinerU
 Requires-Python: >=3.9
@@ -9,7 +9,7 @@ License-File: LICENSE.md
 Requires-Dist: boto3>=1.28.43
 Requires-Dist: Brotli>=1.1.0
 Requires-Dist: click>=8.1.7
-Requires-Dist: fast-langdetect>=0.2.3
+Requires-Dist: fast-langdetect<0.3.0,>=0.2.3
 Requires-Dist: loguru>=0.6.0
 Requires-Dist: numpy<2.0.0,>=1.21.6
 Requires-Dist: pydantic>=2.7.2
@@ -94,6 +94,10 @@ Easier to use: Just grab MinerU Desktop. No coding, no login, just a simple inte
 </div>
 
 # Changelog
+- 2025/03/03 1.2.1 released, fixed several bugs:
+  - Fixed the impact on punctuation marks during full-width to half-width conversion of letters and numbers
+  - Fixed caption matching inaccuracies in certain scenarios
+  - Fixed formula span loss issues in certain scenarios
 - 2025/02/24 1.2.0 released. This version includes several fixes and improvements to enhance parsing efficiency and accuracy:
   - Performance Optimization
     - Increased classification speed for PDF documents in auto mode.

{magic_pdf-1.2.0.dist-info → magic_pdf-1.2.2.dist-info}/RECORD

@@ -1,5 +1,5 @@
 magic_pdf/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/pdf_parse_union_core_v2.py,sha256=jIrXgU_gKL4toJ6GsCoDxByszaN8mAr5vrEy_c63ewk,38310
+magic_pdf/pdf_parse_union_core_v2.py,sha256=Pt3UtPQgOrF2YudQqrwVVC767_271E-LRg2aUsiggXg,38435
 magic_pdf/config/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/config/constants.py,sha256=fXGzANULnJWLPxwYp3BEFWx-rnorzpySMx63ffyEyq4,1272
 magic_pdf/config/drop_reason.py,sha256=CqjMzBE96Qo8OeFvhhhItY8WhyqsKhE3DmyJLoQZNCc,2248
@@ -24,7 +24,7 @@ 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=ZZTaiIn18OWuWKGbDdpoOZ3VMhe_3_JKwrKCfzDiSk0,13715
+magic_pdf/dict2md/ocr_mkcontent.py,sha256=12WeBVxnBzzruk8CfYqqsV2dpH-mDWmE4Osl1RlRoc8,13741
 magic_pdf/filter/__init__.py,sha256=_7lSez_myu4b6cdzPpQ-NfREuqeBSq_QdyBPKVLyq2U,1505
 magic_pdf/filter/pdf_classify_by_type.py,sha256=YNYXamxYgEiSujwilCNHOtrwpgJGDiQ597qJfardDVc,42354
 magic_pdf/filter/pdf_meta_scan.py,sha256=eOuM0-JgaXvHolSgepGoNDJDmv_uITWLQpH_0MfnVQw,17478
@@ -49,12 +49,13 @@ magic_pdf/libs/markdown_utils.py,sha256=86v2BmsSV4NkoRZrH4uQD1youJhYFF3vIKr_vDeg
 magic_pdf/libs/path_utils.py,sha256=Hykw_l5CU736b2egHV9P7B-qh3QNKO4nZSGCbsi0Z8E,1043
 magic_pdf/libs/pdf_check.py,sha256=7GWWvDR6g_rj_fE6XJlbTq5AFVX11ngRIzT0N18F214,3396
 magic_pdf/libs/pdf_image_tools.py,sha256=kjzSEbm7K0yiHv8kJ4VbZ9HHktM8qvAv3LhxRyDZEQk,1987
+magic_pdf/libs/performance_stats.py,sha256=BFi4NIsUYlanznYoTVq4hBpj4NOuShAlWBHzebBGVYM,1702
 magic_pdf/libs/safe_filename.py,sha256=ckwcM_eqoysTb5id8czp-tXq2G9da0-l3pshZDCHQtE,236
-magic_pdf/libs/version.py,sha256=MpAT5hgNoHnTtG1XRD_GV_A7QrHVU6vJjGSw_8qMGA4,22
+magic_pdf/libs/version.py,sha256=uuf4VNtTNA93fMhoAur9YafzaKJFnczY-H1SSCSuRVQ,22
 magic_pdf/model/__init__.py,sha256=sa-dO2k-TLy25I2gRrzjm_cQeYfzMf-pLwBJHkIxGo0,51
 magic_pdf/model/batch_analyze.py,sha256=sbrgOJWycb1Ep6e62CPi6jEyG6VSeklIxc4PmrqaLhM,11933
-magic_pdf/model/doc_analyze_by_custom_model.py,sha256=wma0aq6RyxAepEqnaiTJ9_pWWKLVBj39c6xWA85dxzA,8068
-magic_pdf/model/magic_model.py,sha256=OcKhSJ_PyAAldgpKPiPxi2uuvnj3Sf4SvXi_5Rv0a6Q,30667
+magic_pdf/model/doc_analyze_by_custom_model.py,sha256=T0-h4QmSIDXRzgF5uWO4jQrwIot221l26PXU52xeKiA,7933
+magic_pdf/model/magic_model.py,sha256=yZKWo_wRck_-YLyFGRiUHGar8sV1Y6458BFLbyBAt74,30682
 magic_pdf/model/model_list.py,sha256=aqfEJlEfbib3D3ISrxc0Coh6SbffYh8Yq2FlQN35_zA,213
 magic_pdf/model/pdf_extract_kit.py,sha256=Rd51VNZPKRA_tUbDss-b44d84K6WDG2S87a37Ax7HUA,12224
 magic_pdf/model/pp_structure_v2.py,sha256=NcqFWL4nUtjl82MFak8HX_8V3i4Aw_fK4dATrIp5uGs,3840
@@ -92,7 +93,7 @@ magic_pdf/model/sub_modules/mfd/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm
 magic_pdf/model/sub_modules/mfd/yolov8/YOLOv8.py,sha256=QfHbMr1br0pOJUu1NJEMgA6yw11G0yFImJv_AfW48_c,1008
 magic_pdf/model/sub_modules/mfd/yolov8/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/model/sub_modules/mfr/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-magic_pdf/model/sub_modules/mfr/unimernet/Unimernet.py,sha256=R05qw54QuLl2btNWdkxf4yCjDeEj8o0786e-gz_Xv8k,5290
+magic_pdf/model/sub_modules/mfr/unimernet/Unimernet.py,sha256=HPNetRfQeHoHfRTzFEaIjLSHfjrxRvS-EaApMUebZuQ,8020
 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
@@ -116,12 +117,12 @@ magic_pdf/operators/models.py,sha256=mRqbCVrxxaUVDpEBAsXaK7EL1M-goICkE1W0FYgewio
 magic_pdf/operators/pipes.py,sha256=XgBgisKQd_ruW-3Tw4v5LhqloZUHgn2aFcpi_q8LbCs,6767
 magic_pdf/post_proc/__init__.py,sha256=8CRrCQVuExa0BttRFh3Z40lFy2K5jN0sp67KWjOlj5c,50
 magic_pdf/post_proc/llm_aided.py,sha256=yzhu2cCpUZjdwf3v0swYDgSs9VWIfMAoXepYIP1EMZs,6367
-magic_pdf/post_proc/para_split_v3.py,sha256=v4SdQn4OZdHRXpWQMfQ-FGJz_tglQ88uFUqpwY542Fo,16922
+magic_pdf/post_proc/para_split_v3.py,sha256=SPN_VVGvFX5KpFMGw9OzgoE-kTZq-FF036i0cIImGH8,16975
 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=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=vrbLIzNIjxrm7PonfHaFdY6qaicc0uIly62SJwgZ5UM,5496
+magic_pdf/pre_proc/ocr_dict_merge.py,sha256=4Z3aHZ9sxzijkVpOCENslvUcpp7DXgNID4Gl3pxwIg4,5512
 magic_pdf/pre_proc/ocr_span_list_modify.py,sha256=xrgC9vR0poklZuY4Og41pZVdXzuaGFg3BnQ01X60dpo,3102
 magic_pdf/pre_proc/remove_bbox_overlap.py,sha256=mcdxAh4P56NZ3Ij8h3vW8qC_SrszfXflVWuWUuUiTNg,3089
 magic_pdf/resources/fasttext-langdetect/lid.176.ftz,sha256=jzRyz-hzintgmejpmcPL-uDc0VaWqsfXc4qAOdtgPoM,938013
@@ -138,9 +139,9 @@ magic_pdf/tools/common.py,sha256=1LfMeXBBsb3WlGeNAze_pPOYXQ8Qbfh-JgRXweojHKo,838
 magic_pdf/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 magic_pdf/utils/annotations.py,sha256=82ou3uELNbQWa9hOFFkVt0gsIskAKf5msCv5J2IJ5V0,211
 magic_pdf/utils/office_to_pdf.py,sha256=7aj-Ls2v8saD-Rgu_t3FIc-J3Ka9wnmiEH5zY-H1Vxs,729
-magic_pdf-1.2.0.dist-info/LICENSE.md,sha256=jVa0BUaKrRH4erV2P5AeJ24I2WRv9chIGxditreJ6e0,34524
-magic_pdf-1.2.0.dist-info/METADATA,sha256=7iel3MItxKhJc1Bbfh_NMbDp8a23k9G1vA8LYEw2k_U,40720
-magic_pdf-1.2.0.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
-magic_pdf-1.2.0.dist-info/entry_points.txt,sha256=wXwYke3j8fqDQTocUspL-CqDUEv3Tfcwp09fM8dZAhA,98
-magic_pdf-1.2.0.dist-info/top_level.txt,sha256=J9I0AzmHWGkp9c6DL8Oe4mEx3yYphLzkRn4H25Lg1rE,10
-magic_pdf-1.2.0.dist-info/RECORD,,
+magic_pdf-1.2.2.dist-info/LICENSE.md,sha256=jVa0BUaKrRH4erV2P5AeJ24I2WRv9chIGxditreJ6e0,34524
+magic_pdf-1.2.2.dist-info/METADATA,sha256=FYzj0yWzmFAG4mQ22DH9F4KZfqexNg7YuhgiXMHc9Ug,41001
+magic_pdf-1.2.2.dist-info/WHEEL,sha256=tZoeGjtWxWRfdplE7E3d45VPlLNQnvbKiYnx7gwAy8A,92
+magic_pdf-1.2.2.dist-info/entry_points.txt,sha256=wXwYke3j8fqDQTocUspL-CqDUEv3Tfcwp09fM8dZAhA,98
+magic_pdf-1.2.2.dist-info/top_level.txt,sha256=J9I0AzmHWGkp9c6DL8Oe4mEx3yYphLzkRn4H25Lg1rE,10
+magic_pdf-1.2.2.dist-info/RECORD,,