natural-pdf 0.1.5__py3-none-any.whl → 0.1.7__py3-none-any.whl

natural_pdf/utils/visualization.py

@@ -192,7 +192,7 @@ def merge_images_with_legend(
     if not legend:
         return image  # Return original image if legend is None or empty
 
-    bg_color = (255, 255, 255, 255) # Always use white for the merged background
+    bg_color = (255, 255, 255, 255)  # Always use white for the merged background
 
     if position == "right":
         # Create a new image with extra width for the legend

{natural_pdf-0.1.5.dist-info → natural_pdf-0.1.7.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: natural-pdf
-Version: 0.1.5
+Version: 0.1.7
 Summary: A more intuitive interface for working with PDFs
 Author-email: Jonathan Soma <jonathan.soma@gmail.com>
 License-Expression: MIT
@@ -16,29 +16,34 @@ Requires-Dist: Pillow
 Requires-Dist: colour
 Requires-Dist: numpy
 Requires-Dist: urllib3
-Requires-Dist: torch
-Requires-Dist: torchvision
-Requires-Dist: transformers
-Requires-Dist: huggingface_hub
-Requires-Dist: ocrmypdf
-Requires-Dist: pikepdf
+Requires-Dist: tqdm
 Provides-Extra: interactive
 Requires-Dist: ipywidgets<9.0.0,>=7.0.0; extra == "interactive"
 Provides-Extra: haystack
 Requires-Dist: haystack-ai; extra == "haystack"
 Requires-Dist: chroma-haystack; extra == "haystack"
 Requires-Dist: sentence-transformers; extra == "haystack"
-Requires-Dist: protobuf<4; extra == "haystack"
+Requires-Dist: natural-pdf[core-ml]; extra == "haystack"
 Provides-Extra: easyocr
 Requires-Dist: easyocr; extra == "easyocr"
+Requires-Dist: natural-pdf[core-ml]; extra == "easyocr"
 Provides-Extra: paddle
 Requires-Dist: paddlepaddle; extra == "paddle"
 Requires-Dist: paddleocr; extra == "paddle"
 Provides-Extra: layout-yolo
 Requires-Dist: doclayout_yolo; extra == "layout-yolo"
+Requires-Dist: natural-pdf[core-ml]; extra == "layout-yolo"
 Provides-Extra: surya
 Requires-Dist: surya-ocr; extra == "surya"
+Requires-Dist: natural-pdf[core-ml]; extra == "surya"
 Provides-Extra: qa
+Requires-Dist: natural-pdf[core-ml]; extra == "qa"
+Provides-Extra: docling
+Requires-Dist: docling; extra == "docling"
+Requires-Dist: natural-pdf[core-ml]; extra == "docling"
+Provides-Extra: llm
+Requires-Dist: openai>=1.0; extra == "llm"
+Requires-Dist: pydantic; extra == "llm"
 Provides-Extra: test
 Requires-Dist: pytest; extra == "test"
 Provides-Extra: dev
@@ -50,18 +55,33 @@ Requires-Dist: nox; extra == "dev"
 Requires-Dist: nox-uv; extra == "dev"
 Requires-Dist: build; extra == "dev"
 Requires-Dist: uv; extra == "dev"
+Requires-Dist: pipdeptree; extra == "dev"
+Requires-Dist: nbformat; extra == "dev"
+Requires-Dist: jupytext; extra == "dev"
+Requires-Dist: nbclient; extra == "dev"
 Provides-Extra: all
-Requires-Dist: ipywidgets<9.0.0,>=7.0.0; extra == "all"
-Requires-Dist: easyocr; extra == "all"
-Requires-Dist: paddlepaddle; extra == "all"
-Requires-Dist: paddleocr; extra == "all"
-Requires-Dist: doclayout_yolo; extra == "all"
-Requires-Dist: surya-ocr; extra == "all"
-Requires-Dist: haystack-ai; extra == "all"
-Requires-Dist: chroma-haystack; extra == "all"
-Requires-Dist: sentence-transformers; extra == "all"
-Requires-Dist: protobuf<4; extra == "all"
-Requires-Dist: pytest; extra == "all"
+Requires-Dist: natural-pdf[interactive]; extra == "all"
+Requires-Dist: natural-pdf[haystack]; extra == "all"
+Requires-Dist: natural-pdf[easyocr]; extra == "all"
+Requires-Dist: natural-pdf[paddle]; extra == "all"
+Requires-Dist: natural-pdf[layout_yolo]; extra == "all"
+Requires-Dist: natural-pdf[surya]; extra == "all"
+Requires-Dist: natural-pdf[qa]; extra == "all"
+Requires-Dist: natural-pdf[ocr-export]; extra == "all"
+Requires-Dist: natural-pdf[docling]; extra == "all"
+Requires-Dist: natural-pdf[llm]; extra == "all"
+Requires-Dist: natural-pdf[test]; extra == "all"
+Provides-Extra: core-ml
+Requires-Dist: torch; extra == "core-ml"
+Requires-Dist: torchvision; extra == "core-ml"
+Requires-Dist: transformers[sentencepiece]; extra == "core-ml"
+Requires-Dist: huggingface_hub; extra == "core-ml"
+Provides-Extra: ocr-export
+Requires-Dist: ocrmypdf; extra == "ocr-export"
+Requires-Dist: pikepdf; extra == "ocr-export"
+Provides-Extra: export-extras
+Requires-Dist: jupytext; extra == "export-extras"
+Requires-Dist: nbformat; extra == "export-extras"
 Dynamic: license-file
 
 # Natural PDF
@@ -89,6 +109,10 @@ pip install natural-pdf[easyocr]
 pip install natural-pdf[surya]
 pip install natural-pdf[paddle]
 
+# Example: Install support for features using Large Language Models (e.g., via OpenAI-compatible APIs)
+pip install natural-pdf[llm]
+# (May require setting API key environment variables, e.g., GOOGLE_API_KEY for Gemini)
+
 # Example: Install with interactive viewer support
 pip install natural-pdf[interactive]
 
@@ -141,7 +165,7 @@ Natural PDF offers a range of features for working with PDFs:
 *   **Spatial Navigation:** Select content relative to other elements (`heading.below()`, `element.select_until(...)`).
 *   **Text & Table Extraction:** Get clean text or structured table data, automatically handling exclusions.
 *   **OCR Integration:** Extract text from scanned documents using engines like EasyOCR, PaddleOCR, or Surya.
-*   **Layout Analysis:** Detect document structures (titles, paragraphs, tables) using AI models.
+*   **Layout Analysis:** Detect document structures (titles, paragraphs, tables) using various engines (e.g., YOLO, Paddle, LLM via API).
 *   **Document QA:** Ask natural language questions about your document's content.
 *   **Semantic Search:** Index PDFs and find relevant pages or documents based on semantic meaning using Haystack.
 *   **Visual Debugging:** Highlight elements and use an interactive viewer or save images to understand your selections.

{natural_pdf-0.1.5.dist-info → natural_pdf-0.1.7.dist-info}/RECORD

@@ -12,12 +12,13 @@ docs/document-qa/index.ipynb,sha256=MXJoFhi8TUKK6ZnRFiUBglLGpMbzwdb7LJYfzw8Gp48,
 docs/document-qa/index.md,sha256=mwuO4tothg0OzBXewnj73QEJu46Udq7f1pQBYrKOHwM,2131
 docs/element-selection/index.ipynb,sha256=-7PwKw1RbPlZ4stzN1Rd1GJ8mwjOD4ySsLcpqVX7chc,1193628
 docs/element-selection/index.md,sha256=_1P8vI64Y0aSVwUzdRJD4ayb80BJWBLED9TvVpveFx8,6979
+docs/finetuning/index.md,sha256=Ur3zqSaR0X8PvBCSyI7cFiDv5qZ6Jtv4omBKXCKAzEk,9200
 docs/installation/index.md,sha256=nd4RZrQFR8_vv7Xm3xAzp7z-CQQr9ffAcGa7yuEYn2U,1594
 docs/interactive-widget/index.ipynb,sha256=zY1rz5N34OUW-OtgcbI6iiOjlIJqXjVcx9OoNWMjuyU,293111
 docs/interactive-widget/index.md,sha256=tZbq0uYI7Zwo9mLbhXpqeBriuAjazkIyEJeP-jasJ-Q,259
 docs/layout-analysis/index.ipynb,sha256=dkS_-cu-KGir5G2LGRcxBThKnW0dfA5nPPnwpoYGFtU,1869093
 docs/layout-analysis/index.md,sha256=ZnH5yd7B_eOLgGxW_4rNlzQs4Tn3Xx1cK3jX43CSpSM,5390
-docs/ocr/index.md,sha256=e5E9wqY6ehX7sJtrrie6gr6HtQb2neiYEoDKu4Qcjs4,8931
+docs/ocr/index.md,sha256=uuzTqcAgUmMN7jZVq8VkVcbRDHn8Yg2nJVvHJ-bDK-Y,8177
 docs/pdf-navigation/index.ipynb,sha256=h6yew0HePXK1_c5FmETqzjBQceUBT0MU-vnXx_y91mo,8018
 docs/pdf-navigation/index.md,sha256=P3b3tsmOcmRtnfRxpsMeTgwm7vApnH_4le_QIwJd51M,2391
 docs/regions/index.ipynb,sha256=5A-N5A4v4lcXNptOAeI4i7i9Gx66To-Yus8B816dHBk,1303347
@@ -28,38 +29,38 @@ docs/text-analysis/index.ipynb,sha256=iaup8pcQXGp0ZK3IWi-HHssQLdIzWYGYfvZK5i8yjj
 docs/text-analysis/index.md,sha256=02pfZemOgV37izV7H-XzKmHu7AedDKLidQ-sKhYaMVw,3527
 docs/text-extraction/index.ipynb,sha256=809y9ZamXT3bc3GhwwFyoDnlyEpO-kUZ3tIsZZWyrj8,2537087
 docs/text-extraction/index.md,sha256=b1KfQpvIEelc8cPbFETUnK92az7iB4b7-LqK2DRH8vw,6985
-docs/tutorials/01-loading-and-extraction.ipynb,sha256=tB1TLios1FaieMUE4RuY_H6fVYpmDwFiMMxW6sillbs,541071
+docs/tutorials/01-loading-and-extraction.ipynb,sha256=SCW26hxW9PhOspiR-2X5CD6L1EiJRfXouO-OF_Nc718,4548
 docs/tutorials/01-loading-and-extraction.md,sha256=g40J8GhKz-ikM2URj5MqIatKKj4l5kTFozHeVjxDJQA,2191
-docs/tutorials/02-finding-elements.ipynb,sha256=oEdkN20PXGM1oH9p0QnSsq8yjQJJ9SU9eQhKO_g4RVQ,524374
+docs/tutorials/02-finding-elements.ipynb,sha256=k1CSz47_atA9D6DXfQzVS64t5-L-KjssU2VuFvdy7oU,524374
 docs/tutorials/02-finding-elements.md,sha256=qOkjcWUzem05of54aKzKvy-MMzRX_S4CyZisVV-73QM,4162
-docs/tutorials/03-extracting-blocks.ipynb,sha256=2e7fc9t_46x0DM5RLI9aUUfLeVzfFZzzzKphBA8G5lY,260729
+docs/tutorials/03-extracting-blocks.ipynb,sha256=1UjdP0j3kPCE3aU8p1jBCBqflG-xRLli2Ltx80DhOVk,260729
 docs/tutorials/03-extracting-blocks.md,sha256=_kqvhk6rSL7cGp2MSwTJk8LYlJGbK_r_umnCSBdR8XU,1665
-docs/tutorials/04-table-extraction.ipynb,sha256=o0LdALyko01oHJbMuuqZkVnoF2pfFnMNgtg1IgpSnRI,3973
+docs/tutorials/04-table-extraction.ipynb,sha256=u92Wppw1qHG__Mx3ZKtETm4AWuGF8X-Ln3kvmF8zCSo,3973
 docs/tutorials/04-table-extraction.md,sha256=4q4v17VX8K-ZBtWYy0nbWPccyqB_ybd5Vl_IROmxz6Q,2130
-docs/tutorials/05-excluding-content.ipynb,sha256=6ZLFm3L_Odr4NJD2iW6mL81y5e3xymQvEVyv-VY5O6U,336197
+docs/tutorials/05-excluding-content.ipynb,sha256=oSg8ll_nuWOfQHGLp0fNKVeyYyn_L8a-F7HJADjjdq8,336857
 docs/tutorials/05-excluding-content.md,sha256=U52SPlc5knqxiyhRokmxrj06T54r2ENyTfP7BMGykhY,3907
-docs/tutorials/06-document-qa.ipynb,sha256=BdFQNRqg6U4trACM8lmYSs9RYHJlxcd6DafXdw5mWR8,10303
+docs/tutorials/06-document-qa.ipynb,sha256=Facyqns8jw2bTvsOSbNnsLskFH8kg1JTz4kmJ16dpcE,10303
 docs/tutorials/06-document-qa.md,sha256=PzPPgw0Rkkfe6sfz3XyKD9S9JbQ40qf4bDzCBvwH1P0,3026
-docs/tutorials/07-layout-analysis.ipynb,sha256=A5HMljUq7AaDSg_-vFywIQCyjKW2tjMmSPyPdaKFAE4,554523
+docs/tutorials/07-layout-analysis.ipynb,sha256=tdNnMro1V66YPx0h96HZnujSm-zDpy7o78euQix4lyU,559517
 docs/tutorials/07-layout-analysis.md,sha256=NAYVzJTecDnXjo_isbPCSUBSn3c-xM1tELct1Zn5GmI,2533
-docs/tutorials/07-working-with-regions.ipynb,sha256=cRkr9VRho7J-dx9aIINO253Uz8io3PhD2mjNrASxql4,69510
-docs/tutorials/07-working-with-regions.md,sha256=Hi18sZhiHV1NDYE-EQ82OPMwrz-j1Krjw_ipT9cTkSI,4379
-docs/tutorials/08-spatial-navigation.ipynb,sha256=7HAAaK80R82Fy09heZ9WKwijY50DS89qGt_Xf2lB0Vo,193515
+docs/tutorials/07-working-with-regions.ipynb,sha256=s4BFKKbKUemmURCpg6j91rNI8eFFOJUgxY4QN4alK4I,69584
+docs/tutorials/07-working-with-regions.md,sha256=oanbTFSQ-topAVd9kjfkaPiMjHcx6Y8cqyxVbmxLhgs,4365
+docs/tutorials/08-spatial-navigation.ipynb,sha256=jfwF6OHLvrMvaaknp-9AfUvr-pPXjPljUyGnFKF9wsw,194523
 docs/tutorials/08-spatial-navigation.md,sha256=IMbOYBjayXKE7pHfBjApTxOoKRD8WYj7opf8fsJCtzA,4855
-docs/tutorials/09-section-extraction.ipynb,sha256=JqkcPDXaifJSYJjbBB3LxB8XCMhbrWs-y5GcuOIvoNA,1100632
+docs/tutorials/09-section-extraction.ipynb,sha256=Aqcy08oXTJ1pkJCmVVumndje-4WXnbkl_QfJPhps7f8,1100736
 docs/tutorials/09-section-extraction.md,sha256=Jy_be8ftAl_VPBWl5nEv7_5sKSZPx22DLUcBVHMD3Nc,7832
-docs/tutorials/10-form-field-extraction.ipynb,sha256=azOE7nDz-rYm-AqXF1NvO41CthR9DTwA_rbXHtobDZ4,280125
+docs/tutorials/10-form-field-extraction.ipynb,sha256=yyopvBoS5vkKKtUQ6rZ4Kyo5E0Olp2WYnmunhfzSQkQ,281491
 docs/tutorials/10-form-field-extraction.md,sha256=t9tPlW36vJEhDrKIsHGg_f3P_MK62DT4-ZK1thKFs4Y,5494
-docs/tutorials/11-enhanced-table-processing.ipynb,sha256=GWH3xn2LTQztOTvkqjbqsUc6IbmuA0hJVFEPZ_O7Jew,1278
+docs/tutorials/11-enhanced-table-processing.ipynb,sha256=BWpVUhtjaAX7r4OOdiy5gQgrSqREaoB0L5TuHqoHEn8,1278
 docs/tutorials/11-enhanced-table-processing.md,sha256=2HK-r1UwU7FLn7zWr_pMG7iLk-i0L4U4-t6ubOEeduc,282
-docs/tutorials/12-ocr-integration.ipynb,sha256=HeahYziw6aEIzMdTCN8F3XPPBmyVmZ0NU11ZT9JiMy0,23897
-docs/tutorials/12-ocr-integration.md,sha256=8FYgRciCkAPFF-tW1rkl5CrMGmvCR6oVWT1-f_tJ5as,4831
-docs/tutorials/13-semantic-search.ipynb,sha256=LhqelW0jxcAW1hpvBrEcCeM6gb5AKD10PJ439ywlHrw,73920
+docs/tutorials/12-ocr-integration.ipynb,sha256=xurkoPwgk2p6mhmPdCehy9ccuYHrAhBCb1zGnjRbZ7Y,26724
+docs/tutorials/12-ocr-integration.md,sha256=wU90sfnm1R6BoMFq-orbGpl8OUVcm-wEBTlK0bLgJC4,4572
+docs/tutorials/13-semantic-search.ipynb,sha256=5h806AIal3EwXPVuXJESbXwdUImCx7fo0mo5-f3Dj44,42817
 docs/tutorials/13-semantic-search.md,sha256=nsNjv0ipYUC3YPSqT5d6dga9ZjObEc04Mc8c0-gsRnU,2914
 docs/visual-debugging/index.ipynb,sha256=MJ92u3Q9sfRCyDAQM4KWmCrs4QhKwIagbn6ytPF83L4,2175800
 docs/visual-debugging/index.md,sha256=ueGD2kNFhEAgIHt7qxCfrLRLjHcR7NTD3AU9okBhX9k,4176
 docs/visual-debugging/region.png,sha256=ULAJs3ZTxMjpD9F4w1DKaZXmhxga3KRq3NrUsXgw28s,67835
-natural_pdf/__init__.py,sha256=A3Bc-K2F_LtG08IjkJGngZraLsAT2FSm35Yic7i4Tuk,2913
+natural_pdf/__init__.py,sha256=UdS-I3d7MzSvpxL-QMQUSUO5IGhh8c5of34BIs49TaU,2670
 natural_pdf/analyzers/__init__.py,sha256=dIXjsMqoxKmd9OOnSBzn12wvdIz7D7YNQRAnXslpJSM,142
 natural_pdf/analyzers/text_options.py,sha256=nE2E1pp4psDPpxmtarvNtEQsgozPkyFRjv0TVP2HTyU,2865
 natural_pdf/analyzers/text_structure.py,sha256=9h8hKRz0JWnr13xQr3b4FFr_-hDIjue07WvG7LmT8nc,12827
@@ -67,35 +68,40 @@ natural_pdf/analyzers/utils.py,sha256=Lgub1kYSTOnNxeLO1klStHLwH-GIuT4vpdqyVRF-Mc
 natural_pdf/analyzers/layout/__init__.py,sha256=oq1uJ5UkGGMbBKGirV1aRKK3hxAUyjTLywYkPCQH1f0,33
 natural_pdf/analyzers/layout/base.py,sha256=9dCR758mAuz7ExlHJ-gwnPnETaM4GZV3W1IRei_t13s,6815
 natural_pdf/analyzers/layout/docling.py,sha256=4BJYyNVR6VegZGxyisvNIBBRvVk6YKPyDVs7ZdVfzEU,12676
+natural_pdf/analyzers/layout/gemini.py,sha256=CzJPWyyEghuCNpu2CMb6OA6FtBGdGhXspHjsjy6I4JE,11195
 natural_pdf/analyzers/layout/layout_analyzer.py,sha256=6aed1qz5jpndOiakXCBRZAcnyG_waeXi3WPuP5fRvh4,14046
-natural_pdf/analyzers/layout/layout_manager.py,sha256=kVBPQ8Ex33SYzzm1fhQOtP4qmHOc92dn4BEfff66Qx4,10053
-natural_pdf/analyzers/layout/layout_options.py,sha256=1u8RVdiRwaq5hhGUpVLIdYXCH6TqEq0UxCPdm6JrdTI,3369
+natural_pdf/analyzers/layout/layout_manager.py,sha256=Vh8EKiszKqjELofxQ1eiVLKVjibyjBsZpLFzTf0_21E,11179
+natural_pdf/analyzers/layout/layout_options.py,sha256=s7xr4brE3OutE6aYNAi2PniRy1p2w8a342C2xGpvX2s,3777
 natural_pdf/analyzers/layout/paddle.py,sha256=gTI9ZqNd5-t4H5IByGfL32WgcE6JrdchW6jRiGI6ulM,13375
 natural_pdf/analyzers/layout/surya.py,sha256=vhji6ynHPMyQLHuYRPQcplNi7m_lG4P4NYtWv6MzcME,13556
 natural_pdf/analyzers/layout/tatr.py,sha256=-GJhMy4d0yx6egkO9-ULAIdQkkQRyAKExoIta-b256U,12971
-natural_pdf/analyzers/layout/yolo.py,sha256=gy_1DY4sG7jU5rQ7Rb6FUGYI9FFMaozAWiWuxRH5yNw,8294
-natural_pdf/collections/pdf_collection.py,sha256=E9GVEgGjTBGpNkuSO_f4GMrSB7Tmi60wnkD7pgvBVOM,12175
+natural_pdf/analyzers/layout/yolo.py,sha256=ANo2U4EZgeN2eYKM1bZIuysiuJLgwl4JeQchrRxOKwA,8388
+natural_pdf/collections/pdf_collection.py,sha256=afE0tNIfwA7IRCc8g0EGgiBgJz3TuJbEzZ5meDNAnQw,13272
 natural_pdf/core/__init__.py,sha256=QC8H4M3KbXwMFiQORZ0pdPlzx1Ix6oKKQSS7Ib2KEaA,38
-natural_pdf/core/element_manager.py,sha256=R2vY7nYbqrtL1FLRbrJvpUyCpf97zUui-2CaEV2CFQc,21858
+natural_pdf/core/element_manager.py,sha256=RjLCzeHDRJCoCx1W_6jGg8KsiCTuXz7Uc2BoSY4M7mE,22144
 natural_pdf/core/highlighting_service.py,sha256=CTVd7y-fpIreFSe70cTpMu1Pwl6HKMtTHp0bh2U7VXk,32609
-natural_pdf/core/page.py,sha256=CQy3zgHT6VBmo7n6cZ5RITSUURIEPzPsWqWQAUGFOZc,78302
-natural_pdf/core/pdf.py,sha256=yaShN4vHJ1BXxWAj4lRk7udTcl6F9ddicBFCV6kd--w,41146
+natural_pdf/core/page.py,sha256=emS6jJdb-J7xnK8Uo8Hs1n0plbIAGA_YH6kmp36wVgM,84955
+natural_pdf/core/pdf.py,sha256=hOR1i3bJjfJCBCI2m4pBNAMEYpmbtG905QbFe-l8gZU,46525
 natural_pdf/elements/__init__.py,sha256=S8XeiNWJ1WcgnyYKdYV1yxQlAxCCO3FfITT8MQwNbyk,41
-natural_pdf/elements/base.py,sha256=vUga2Nm8DWoRfKMWVTt5N8UMh1q-YzUNbTfDouHzS2U,35698
-natural_pdf/elements/collections.py,sha256=2kwOF_-5TePvLbZLVyeEZRt4Im3KlmX8j46giVdxcUE,66000
+natural_pdf/elements/base.py,sha256=UtoSD-c_s0yiLpWZrIIJjeJ9MgGz_4R0UHYcsFWH6bc,35157
+natural_pdf/elements/collections.py,sha256=w0JqLwn57Je00Aq4Ay8SeYmxPjPJvUOtkLbgfGM2-nM,68882
 natural_pdf/elements/line.py,sha256=7cow3xMUKhAj7zoQz7OaB1eIH2_a8B__LB7iGJ4Mb0o,4612
 natural_pdf/elements/rect.py,sha256=kiVa3e377ZnqIOXc89d9ZSY4EcmDxtccdtUw-HOQzpw,3796
-natural_pdf/elements/region.py,sha256=GOHnq4j4GL-UUQyLdnCLPb0YhBq_YrHn6anecX03t30,67714
-natural_pdf/elements/text.py,sha256=Q4hKlXyGhz7njnr_-sON1p8Uqqc8qZBLAqu0VUkT-OE,10958
-natural_pdf/exporters/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+natural_pdf/elements/region.py,sha256=9E21LYQWB98coi_73Kpf9mQ60p9ElzGOJzxdtgOUfh4,69662
+natural_pdf/elements/text.py,sha256=8PNKSLUgXUhEu9IFfbNbSSpuu0Slm11T6UH8jn4O6hQ,11078
+natural_pdf/exporters/__init__.py,sha256=7MnvRLLQdwtg-ULu-8uK8C84GsKiJamyhRw_GgWhw7k,151
+natural_pdf/exporters/base.py,sha256=s-NpHoH81x80GQxs0oqjdhPGrzbUa8npjnK8apKOsHQ,2115
+natural_pdf/exporters/paddleocr.py,sha256=1G2bS2-CcuAtS78JZYRczO3r5k8fdO9jrExH0Kr9r7M,16249
 natural_pdf/exporters/searchable_pdf.py,sha256=qsaPsnbOOaZHA_aplfZbwQnBoK9KghWm-wzbyRRomeY,16859
-natural_pdf/ocr/__init__.py,sha256=8ytKCg4VmiX1LkehnCbBNnL-zLc95CmsLJKDfGFtklE,1916
-natural_pdf/ocr/engine.py,sha256=D8B8QHF_8E68JaklyHoNUA3hGn5ld7TGKHD7Ho6mJMg,4316
-natural_pdf/ocr/engine_easyocr.py,sha256=aE9alo9rKxZddQAt3liWvlGssH8b2jRtmLwpndttlRM,8617
-natural_pdf/ocr/engine_paddle.py,sha256=4ELH9P9-FOTYj_CbUOKiW2gf5U9v2Tscx0fX0nW8j84,9248
-natural_pdf/ocr/engine_surya.py,sha256=pzv4CKneXlRvCXsdImAEou81MOTwMYCNhNIVG6Fg3rU,7922
-natural_pdf/ocr/ocr_manager.py,sha256=S2ndzKdB-nmK9glbSmn7srotIgKweCByeGIX6SOoEY4,10465
-natural_pdf/ocr/ocr_options.py,sha256=JZXRxjsQuKf9GJMt56YikcOqsTQ7SvXOv2XZ7z1qnB8,3794
+natural_pdf/ocr/__init__.py,sha256=jKaDbo13CdCDcas1WiBmg5gjBvVeG-Z9uaeYxyzvaNY,2464
+natural_pdf/ocr/engine.py,sha256=ZBC1tZNM5EDbGDJJmZI9mNHr4nCMLEZvUFhiJq8GdF4,8741
+natural_pdf/ocr/engine_easyocr.py,sha256=rnDXLNa-keymonR3qbLEbbxA6bqk4QUAVCHKUDixqKg,9045
+natural_pdf/ocr/engine_paddle.py,sha256=2nIrvLBBAiZG1BxVo3eFVJulA6YGoOTXw_RN98p_BUk,6184
+natural_pdf/ocr/engine_surya.py,sha256=iySjG-Dahgh0cLICfbMtOcwUpRFcZjo-5Ed5Zwz-o5Y,4805
+natural_pdf/ocr/ocr_factory.py,sha256=IFccj0BB75YGV4hjcy4ECtGQX_JQzdptpvDFfeGxxgI,4391
+natural_pdf/ocr/ocr_manager.py,sha256=PqF1z1ET8emSw19r7jtEkC9_LZJXY7C5zK5cFklo57I,9238
+natural_pdf/ocr/ocr_options.py,sha256=MIH7cOe8esuiGcVe4AtArSeQdaIpUu9RaUZbuwwvKQw,3294
+natural_pdf/ocr/utils.py,sha256=kdO4sCBqCb5qB-9iPqdPN8_5t1jWwijpT-ci5UHnz6A,3867
 natural_pdf/qa/__init__.py,sha256=Pjo62JTnUNEjGNsC437mvsS5KQ5m7X_BibGvavR9AW0,108
 natural_pdf/qa/document_qa.py,sha256=W4E4vS_Eox_IBsYpVb0ifQbJb0FP-PYEIG93CU3rUkE,15246
 natural_pdf/search/__init__.py,sha256=EB_HRwlktJn5WGPVtSaRbOQNjLAZTxujeYf_eN-zd2U,4191
@@ -107,16 +113,23 @@ natural_pdf/search/searchable_mixin.py,sha256=M2a6FaFVM0vcfh7FgjDH6BLhS-7ggeVpcf
 natural_pdf/selectors/__init__.py,sha256=oZGeqSv53EqmIZOhcnawuaGGlRg1h79vArXuZCWKm4A,123
 natural_pdf/selectors/parser.py,sha256=59_GSsTApM6MFvtqhrrmbKaBfODPbGXMluvvQJcrqhE,15754
 natural_pdf/templates/__init__.py,sha256=jYBxzfi73vew0f6yhIh1MlRxw4F_TVN2hKQR0YXOFe0,20
-natural_pdf/templates/ocr_debug.html,sha256=Zy9StzBeHFQU8ity6cjFSZLe3TY0QOabUux4c5WQUzs,19171
+natural_pdf/templates/finetune/fine_tune_paddleocr.md,sha256=AGt6kQWSTJZ8F28iN1D4p_Q6f1bvFML9gyUk6QcSHDc,14517
+natural_pdf/templates/spa/index.html,sha256=6hLTp07OeV5Q4jUMp5Sgl-dwfBs3oPzBxqphG4kEs24,787
+natural_pdf/templates/spa/words.txt,sha256=vkGtl5Y7-Nq-3Vhx1daRWWF1Jp1UCVaw-ZZaiFwrurk,2493885
+natural_pdf/templates/spa/css/style.css,sha256=Qdl0U3L5HMyhBDNzyRPklfb3OxW6rMxCfQbzO8i8IW4,7643
+natural_pdf/templates/spa/js/app.js,sha256=Efb7NmcTN9RLdLwKpDcU6CG5Ix0laHtzRHmfUlDMJXw,19679
 natural_pdf/utils/__init__.py,sha256=s3M8FggaK1P3EBYn6R_-HgSDjNc9C73gyKe1hihtNWg,43
+natural_pdf/utils/debug.py,sha256=lk_6qzxan8NagjEtJEZpZ2MS30SO8ce6iznBxmA0xgk,995
 natural_pdf/utils/highlighting.py,sha256=EIY6ihVGtUTS_DjWyxpnr_UXpcR4btC1KhSGQ9VUfKg,698
+natural_pdf/utils/identifiers.py,sha256=n61viCQiMlf5-E_jsPLe-FkPBdKkMKv-gfs5tGqlKiw,1117
+natural_pdf/utils/packaging.py,sha256=HSgpubpHICU75L4ZAZPU8iOjium055XWnklV9_YqoCA,21579
 natural_pdf/utils/reading_order.py,sha256=s3DsYq_3g_1YA07qhd4BGEjeIRTeyGtnwc_hNtSzwBY,7290
-natural_pdf/utils/text_extraction.py,sha256=VlbkXg14GlvwYTjRJWa8FVUigETY3Hq0v8NlIRnzYkM,8619
-natural_pdf/utils/visualization.py,sha256=ir5PgpptRuVuVeRT9IcdTsNeEpdOYD_69rByjHQ7JhI,8592
+natural_pdf/utils/text_extraction.py,sha256=ujhqU2C9y2YwzGDBfT9oiGPUvSz6mVqq72ttd3Ksskg,7712
+natural_pdf/utils/visualization.py,sha256=5GbhxtvZW-77ONVnICupg-s2D-OaxLZNqkKlOrQESK4,8593
 natural_pdf/widgets/__init__.py,sha256=O2fSDo604wDAP6UwUkmBq3eT91RSqHwBpAOQXq92S8s,214
 natural_pdf/widgets/viewer.py,sha256=Aiw6kuBc0WkhcZrPNKyLNzzWbmtmU6rvOmHV0IuXCBk,40862
 natural_pdf/widgets/frontend/viewer.js,sha256=w8ywfz_IOAAv2nP_qaf2VBUkF1KhjT3zorhJxM1-CfU,4371
-natural_pdf-0.1.5.dist-info/licenses/LICENSE,sha256=9zfwINwJlarbDmdh6iJV4QUG54QSJlSAUcnC1YiC_Ns,1074
+natural_pdf-0.1.7.dist-info/licenses/LICENSE,sha256=9zfwINwJlarbDmdh6iJV4QUG54QSJlSAUcnC1YiC_Ns,1074
 notebooks/Examples.ipynb,sha256=l4YMtMEx_DWBzWIjl9CmBkWTo0g_nK8l_XWOyzYooQM,4275170
 pdfs/.gitkeep,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 pdfs/01-practice.pdf,sha256=dxWyJIa2cm7bALE3BWDJ2dg3inyFlo1n8ntVyy0hkTo,7906
@@ -126,9 +139,7 @@ pdfs/2014 Statistics.pdf,sha256=B-30OQVjqj_3718-G9cGUefNddnz-MosPdHAzfGfkcc,9559
 pdfs/2019 Statistics.pdf,sha256=reuSJxvAlx9_P-pW7IPqzox0jFCxSPbK1i1-WFu-uGA,511439
 pdfs/Atlanta_Public_Schools_GA_sample.pdf,sha256=PLBh_uWJQH0MnBaSm5ng5Ima63_m6Mi11CjdravB_S8,137689
 pdfs/needs-ocr.pdf,sha256=vusKiLxSOlELUTetfZfaotNU54RtMj9PCzGfLc2cuNs,139305
-tests/test_loading.py,sha256=AHjnIKqEAdtQa28kEAhFQTJ0Nnu49AmxnPM8YE8_EP0,1770
-tests/test_optional_deps.py,sha256=e9H3ylLsB4cnyC3TVMgUbBMzmSbq6MlH8jn_pqh4Hus,12111
-natural_pdf-0.1.5.dist-info/METADATA,sha256=7lBaQX1e1ibQibz2ZRYt0DsDy84k3sk6dGJqT9lQvWg,5466
-natural_pdf-0.1.5.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
-natural_pdf-0.1.5.dist-info/top_level.txt,sha256=N44f8aOLMpI6MzrNHsCD8MzElkir_H1nPUGZ4QToWqI,38
-natural_pdf-0.1.5.dist-info/RECORD,,
+natural_pdf-0.1.7.dist-info/METADATA,sha256=BMzSroqVMlbJrti_56ilNFZkSEH2-hJc8vUVrjk3OZU,6766
+natural_pdf-0.1.7.dist-info/WHEEL,sha256=pxyMxgL8-pra_rKaQ4drOZAegBVuX-G_4nRHjjgWbmo,91
+natural_pdf-0.1.7.dist-info/top_level.txt,sha256=7nDKUnpkN7B8cBI7DEpW5JM8S7OcOgHw3jXH-1iCX2o,32
+natural_pdf-0.1.7.dist-info/RECORD,,

{natural_pdf-0.1.5.dist-info → natural_pdf-0.1.7.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (78.1.0)
+Generator: setuptools (79.0.0)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

{natural_pdf-0.1.5.dist-info → natural_pdf-0.1.7.dist-info}/top_level.txt

@@ -2,4 +2,3 @@ docs
 natural_pdf
 notebooks
 pdfs
-tests