magic-pdf 1.3.2__py3-none-any.whl → 1.3.3__py3-none-any.whl

magic_pdf/libs/version.py

@@ -1 +1 @@
-__version__ = "1.3.2"
+__version__ = "1.3.3"

magic_pdf/model/doc_analyze_by_custom_model.py

@@ -147,7 +147,7 @@ def doc_analyze(
             images.append(img_dict['img'])
             page_wh_list.append((img_dict['width'], img_dict['height']))
 
-    images_with_extra_info = [(images[index], ocr, dataset._lang) for index in range(len(dataset))]
+    images_with_extra_info = [(images[index], ocr, dataset._lang) for index in range(len(images))]
 
     if len(images) >= MIN_BATCH_INFERENCE_SIZE:
         batch_size = MIN_BATCH_INFERENCE_SIZE

magic_pdf/pre_proc/ocr_detect_all_bboxes.py

@@ -99,11 +99,11 @@ def ocr_prepare_bboxes_for_layout_split_v2(
     all_discarded_blocks = []
     add_bboxes(discarded_blocks, BlockType.Discarded, all_discarded_blocks)
 
-    """footnote识别：宽度超过1/3页面宽度的，高度超过10的，处于页面下半50%区域的"""
+    """footnote识别：宽度超过1/3页面宽度的，高度超过10的，处于页面下半30%区域的"""
     footnote_blocks = []
     for discarded in discarded_blocks:
         x0, y0, x1, y1 = discarded['bbox']
-        if (x1 - x0) > (page_w / 3) and (y1 - y0) > 10 and y0 > (page_h / 2):
+        if (x1 - x0) > (page_w / 3) and (y1 - y0) > 10 and y0 > (page_h * 0.7):
             footnote_blocks.append([x0, y0, x1, y1])
 
     """移除在footnote下面的任何框"""

{magic_pdf-1.3.2.dist-info → magic_pdf-1.3.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: magic-pdf
-Version: 1.3.2
+Version: 1.3.3
 Summary: A practical tool for converting PDF to Markdown
 License: AGPL-3.0
 Project-URL: Home, https://mineru.net/
@@ -29,7 +29,7 @@ Requires-Dist: tqdm >=4.67.1
 Requires-Dist: transformers !=4.51.0,<5.0.0,>=4.49.0
 Provides-Extra: full
 Requires-Dist: PyYAML <7,>=6.0.2 ; extra == 'full'
-Requires-Dist: dill <1,>=0.3.9 ; extra == 'full'
+Requires-Dist: dill <1,>=0.3.8 ; extra == 'full'
 Requires-Dist: doclayout-yolo ==0.0.2b1 ; extra == 'full'
 Requires-Dist: ftfy <7,>=6.3.1 ; extra == 'full'
 Requires-Dist: matplotlib <4,>=3.10 ; extra == 'full'
@@ -42,7 +42,7 @@ Requires-Dist: ultralytics <9,>=8.3.48 ; extra == 'full'
 Provides-Extra: full_old_linux
 Requires-Dist: PyYAML ==6.0.2 ; extra == 'full_old_linux'
 Requires-Dist: albumentations ==1.4.20 ; extra == 'full_old_linux'
-Requires-Dist: dill ==0.3.9 ; extra == 'full_old_linux'
+Requires-Dist: dill ==0.3.8 ; extra == 'full_old_linux'
 Requires-Dist: doclayout-yolo ==0.0.2b1 ; extra == 'full_old_linux'
 Requires-Dist: ftfy ==6.3.1 ; extra == 'full_old_linux'
 Requires-Dist: matplotlib <=3.10.1,>=3.10 ; extra == 'full_old_linux'

{magic_pdf-1.3.2.dist-info → magic_pdf-1.3.3.dist-info}/RECORD

@@ -52,10 +52,10 @@ magic_pdf/libs/pdf_check.py,sha256=7GWWvDR6g_rj_fE6XJlbTq5AFVX11ngRIzT0N18F214,3
 magic_pdf/libs/pdf_image_tools.py,sha256=_au7plmKKctpPKozBumSKgP8689q4vH1mU8VMLO0IbM,2260
 magic_pdf/libs/performance_stats.py,sha256=DW-c6nUTUnWKGTONRKfpucsYZm1ake016F9K7jJwbik,2136
 magic_pdf/libs/safe_filename.py,sha256=ckwcM_eqoysTb5id8czp-tXq2G9da0-l3pshZDCHQtE,236
-magic_pdf/libs/version.py,sha256=HgKA3RqZvC7slo8MgLyffCGwJbQ3cY6I7oUMFvGLWps,22
+magic_pdf/libs/version.py,sha256=Vi6om3KImlKsS_Wg5CjUgYffoi2zx7T-SRPnnGL0G7M,22
 magic_pdf/model/__init__.py,sha256=sa-dO2k-TLy25I2gRrzjm_cQeYfzMf-pLwBJHkIxGo0,51
 magic_pdf/model/batch_analyze.py,sha256=yKhKQuZTh9GG83p61bw2BRqKMbnsjsmX73gfuTRk8xE,11272
-magic_pdf/model/doc_analyze_by_custom_model.py,sha256=R9oowGvH3tN1knoSHiyECKsoW3RuKZ1y5cJd42FNurE,10318
+magic_pdf/model/doc_analyze_by_custom_model.py,sha256=-cjn7DQi6kZCqVZ0IxbXuL2kmeGhSVLzLaezIHPFzMU,10317
 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=C3sKqRkoD20Ldmo-cqGn1zRldEL-l5NYqcFvd05_fGU,10845
@@ -178,7 +178,7 @@ magic_pdf/post_proc/para_split_v3.py,sha256=SPN_VVGvFX5KpFMGw9OzgoE-kTZq-FF036i0
 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_detect_all_bboxes.py,sha256=3_bEbZC_BDwbuaBLPdCIbkxz93-g9oCtvjuXD8qbklo,9330
 magic_pdf/pre_proc/ocr_dict_merge.py,sha256=PscKGF0uJIjMxZRM69FLUs1SZO_wOswDQQV1f0M2xAo,5627
 magic_pdf/pre_proc/ocr_span_list_modify.py,sha256=bs5RLvk4kIyx9_Hqq0FU3AGPPxE8Sxs97Uwlf1sBryM,4725
 magic_pdf/pre_proc/remove_bbox_overlap.py,sha256=mcdxAh4P56NZ3Ij8h3vW8qC_SrszfXflVWuWUuUiTNg,3089
@@ -195,9 +195,9 @@ magic_pdf/tools/common.py,sha256=-x0RSFr7SNbdYq7DntaLYmQmaxyF-xKSf4xMpSUTzA0,126
 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-1.3.2.dist-info/LICENSE.md,sha256=jVa0BUaKrRH4erV2P5AeJ24I2WRv9chIGxditreJ6e0,34524
-magic_pdf-1.3.2.dist-info/METADATA,sha256=R56SxjE08VgwAabtD81lCHxjH0hho1-c3-Zeb_zkUjo,45615
-magic_pdf-1.3.2.dist-info/WHEEL,sha256=oiQVh_5PnQM0E3gPdiz09WCNmwiHDMaGer_elqB3coM,92
-magic_pdf-1.3.2.dist-info/entry_points.txt,sha256=wXwYke3j8fqDQTocUspL-CqDUEv3Tfcwp09fM8dZAhA,98
-magic_pdf-1.3.2.dist-info/top_level.txt,sha256=J9I0AzmHWGkp9c6DL8Oe4mEx3yYphLzkRn4H25Lg1rE,10
-magic_pdf-1.3.2.dist-info/RECORD,,
+magic_pdf-1.3.3.dist-info/LICENSE.md,sha256=jVa0BUaKrRH4erV2P5AeJ24I2WRv9chIGxditreJ6e0,34524
+magic_pdf-1.3.3.dist-info/METADATA,sha256=1Y-a4UouLQRhsldrhz6UZLlx4KUFOdjSk5R1gK_oYjs,45615
+magic_pdf-1.3.3.dist-info/WHEEL,sha256=oiQVh_5PnQM0E3gPdiz09WCNmwiHDMaGer_elqB3coM,92
+magic_pdf-1.3.3.dist-info/entry_points.txt,sha256=wXwYke3j8fqDQTocUspL-CqDUEv3Tfcwp09fM8dZAhA,98
+magic_pdf-1.3.3.dist-info/top_level.txt,sha256=J9I0AzmHWGkp9c6DL8Oe4mEx3yYphLzkRn4H25Lg1rE,10
+magic_pdf-1.3.3.dist-info/RECORD,,