natural-pdf 0.1.10__py3-none-any.whl → 0.1.11__py3-none-any.whl

natural_pdf/exporters/original_pdf.py

@@ -0,0 +1,130 @@
+"""
+Module for exporting original PDF pages without modification.
+"""
+
+import logging
+import os
+from pathlib import Path
+from typing import TYPE_CHECKING, List, Set, Union
+
+# Lazy import for optional dependency
+try:
+    import pikepdf
+except ImportError:
+    pikepdf = None
+
+if TYPE_CHECKING:
+    from natural_pdf.core.page import Page
+    from natural_pdf.core.pdf import PDF
+    from natural_pdf.elements.collections import PageCollection
+
+logger = logging.getLogger(__name__)
+
+
+def create_original_pdf(
+    source: Union["Page", "PageCollection", "PDF"], output_path: Union[str, Path]
+):
+    """
+    Creates a new PDF file containing only the original, unmodified pages
+    specified by the source object.
+
+    Requires 'pikepdf'. Install with: pip install "natural-pdf[ocr-export]"
+
+    Args:
+        source: The Page, PageCollection, or PDF object indicating which pages to include.
+        output_path: The path to save the resulting PDF file.
+
+    Raises:
+        ImportError: If 'pikepdf' is not installed.
+        ValueError: If the source object is empty, pages are from different PDFs,
+                    or the source PDF path cannot be determined.
+        RuntimeError: If pikepdf fails to open the source or save the output.
+        pikepdf.PasswordError: If the source PDF is password-protected.
+    """
+    if pikepdf is None:
+        raise ImportError(
+            "Saving original PDF pages requires 'pikepdf'. "
+            "Install with: pip install \"natural-pdf[ocr-export]\""
+        )
+
+    output_path_str = str(output_path)
+    pages_to_extract: List["Page"] = []
+
+    # Determine the list of pages and the source PDF path
+    if hasattr(source, "pages") and isinstance(source.pages, list):  # PDF or PageCollection
+        if not source.pages:
+            raise ValueError("Cannot save an empty collection/PDF.")
+        pages_to_extract = source.pages
+    elif hasattr(source, "page") and hasattr(source, "number"): # Single Page object
+         # Check if it's a natural_pdf.core.page.Page or similar duck-typed object
+         if hasattr(source, 'pdf') and source.pdf and hasattr(source.pdf, 'path'):
+             pages_to_extract = [source]
+         else:
+            raise ValueError("Input Page object does not have a valid PDF reference with a path.")
+    else:
+        raise TypeError(f"Unsupported source type for create_original_pdf: {type(source)}")
+
+
+    if not pages_to_extract:
+         raise ValueError("No valid pages found in the source object.")
+
+    # Verify all pages come from the same PDF and get path
+    first_page_pdf_path = None
+    if hasattr(pages_to_extract[0], "pdf") and pages_to_extract[0].pdf:
+        first_page_pdf_path = getattr(pages_to_extract[0].pdf, "path", None)
+
+    if not first_page_pdf_path:
+        raise ValueError(
+            "Cannot save original pages: Source PDF path not found for the first page."
+        )
+
+    page_indices_set: Set[int] = set()
+    for page in pages_to_extract:
+        page_pdf_path = getattr(getattr(page, "pdf", None), "path", None)
+        if not page_pdf_path or page_pdf_path != first_page_pdf_path:
+            raise ValueError(
+                "Cannot save original pages: All pages must belong to the same source PDF document."
+            )
+        page_indices_set.add(page.index)  # 0-based index
+
+    sorted_indices = sorted(list(page_indices_set))
+
+    logger.info(
+        f"Extracting original pages {sorted_indices} from '{first_page_pdf_path}' to '{output_path_str}'"
+    )
+
+    try:
+        with pikepdf.Pdf.open(first_page_pdf_path) as source_pikepdf_doc:
+            target_pikepdf_doc = pikepdf.Pdf.new()
+
+            for page_index in sorted_indices:
+                if 0 <= page_index < len(source_pikepdf_doc.pages):
+                    # This correctly appends the pikepdf.Page object
+                    target_pikepdf_doc.pages.append(source_pikepdf_doc.pages[page_index])
+                else:
+                    logger.warning(
+                        f"Page index {page_index} out of bounds for source PDF '{first_page_pdf_path}'. Skipping."
+                    )
+
+            if not target_pikepdf_doc.pages:
+                raise RuntimeError(f"No valid pages found to save from source PDF.")
+
+            target_pikepdf_doc.save(output_path_str)
+            logger.info(
+                f"Successfully saved original pages PDF ({len(target_pikepdf_doc.pages)} pages) to: {output_path_str}"
+            )
+
+    except pikepdf.PasswordError:
+        logger.error(
+            f"Failed to open password-protected source PDF: {first_page_pdf_path}"
+        )
+        raise RuntimeError(
+            f"Source PDF '{first_page_pdf_path}' is password-protected."
+        ) from None # Raise specific error without chaining the generic Exception
+    except Exception as e:
+        logger.error(
+            f"Failed to save original pages PDF to '{output_path_str}': {e}",
+            exc_info=True,
+        )
+        # Re-raise as RuntimeError for consistent API error handling
+        raise RuntimeError(f"Failed to save original pages PDF: {e}") from e 

natural_pdf/exporters/searchable_pdf.py

@@ -22,7 +22,7 @@ except ImportError:
     pikepdf = None  # type: ignore
 
 try:
-    from ocrmypdf.hocrtransform import HocrTransform
+    from natural_pdf.exporters.hocr import HocrTransform
 except ImportError:
     HocrTransform = None  # type: ignore
 
@@ -310,7 +310,7 @@ def create_searchable_pdf(
     """
     Creates a searchable PDF from a natural_pdf.PDF object using OCR results.
 
-    Relies on ocrmypdf for hOCR transformation. Requires optional dependencies.
+    Relies on pikepdf for saving the PDF.
 
     Args:
         source: The natural_pdf.PDF, PageCollection, or Page object
@@ -323,7 +323,7 @@ def create_searchable_pdf(
         # This should ideally not happen if dependencies are in main install,
         # but serves as a safeguard during development or if install is broken.
         raise ImportError(
-            "Required dependencies (Pillow, pikepdf, ocrmypdf) are missing. "
+            "Required dependencies (Pillow, pikepdf) are missing. "
             "Please ensure natural-pdf is installed correctly with all dependencies."
         )
     # --- End Safeguard Check ---

natural_pdf/ocr/engine_surya.py

@@ -59,7 +59,7 @@ class SuryaOCREngine(OCREngine):
 
         # Store languages instance variable during initialization to use here
         langs = (
-            [[lang] for lang in self._langs]
+            [self._langs]  # Send all languages together in one list per image
             if hasattr(self, "_langs")
             else [[self.DEFAULT_LANGUAGES[0]]]
         )

{natural_pdf-0.1.10.dist-info → natural_pdf-0.1.11.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: natural-pdf
-Version: 0.1.10
+Version: 0.1.11
 Summary: A more intuitive interface for working with PDFs
 Author-email: Jonathan Soma <jonathan.soma@gmail.com>
 License-Expression: MIT
@@ -91,7 +91,6 @@ 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"

{natural_pdf-0.1.10.dist-info → natural_pdf-0.1.11.dist-info}/RECORD

@@ -1,4 +1,4 @@
-natural_pdf/__init__.py,sha256=LBrQcFOGooaUsTSAk6zrPCQqu0IM-ClvJLasexEk64k,2728
+natural_pdf/__init__.py,sha256=HIYdzHD7QBRssIseUX_oDJYvVJs646tNSYhKHqk0HeA,2495
 natural_pdf/analyzers/__init__.py,sha256=dIXjsMqoxKmd9OOnSBzn12wvdIz7D7YNQRAnXslpJSM,142
 natural_pdf/analyzers/text_options.py,sha256=nE2E1pp4psDPpxmtarvNtEQsgozPkyFRjv0TVP2HTyU,2865
 natural_pdf/analyzers/text_structure.py,sha256=Uhxc7aYB1jddkiwRTEPOg_Te2HfOua4z_OtgP1m3org,12794
@@ -23,20 +23,23 @@ natural_pdf/collections/pdf_collection.py,sha256=obHizc2KR4ZiAspodaPOeMgfpoW3aKg
 natural_pdf/core/__init__.py,sha256=QC8H4M3KbXwMFiQORZ0pdPlzx1Ix6oKKQSS7Ib2KEaA,38
 natural_pdf/core/element_manager.py,sha256=knRN6qXxV-6KZCj2GUOyiqRi83DjJzL77TmKGeiD08Y,25144
 natural_pdf/core/highlighting_service.py,sha256=wINdRxq63_CYYA81EwuCRqhNKimn0dNKyoKWuzkirc0,31959
-natural_pdf/core/page.py,sha256=icJLu6jRbkD3iOE8r60XPkQZ8FN3ZcKo5TT5MVGkGl0,105122
-natural_pdf/core/pdf.py,sha256=gOvLumJZaHXdDwpxbX9HcC_Rea4HaYMemBdYg5GX7gQ,63837
+natural_pdf/core/page.py,sha256=S7Uj3DVksX7o3Qg7hpNulYuxHmqzSJIJ0yXVytPhFqY,105158
+natural_pdf/core/pdf.py,sha256=qpZx5LXZ5Oq1fZ4mzDXBDOIcsApRinMEH0CjVY6jNvM,69273
 natural_pdf/elements/__init__.py,sha256=S8XeiNWJ1WcgnyYKdYV1yxQlAxCCO3FfITT8MQwNbyk,41
 natural_pdf/elements/base.py,sha256=7vVCPQyEHifh4LyBuv0kLTqr_gNbbEMc4SoiJmLfEUQ,37585
-natural_pdf/elements/collections.py,sha256=AN0WrrQYfCmcRS0-PHP4RQHxxdpcWnDuH2cWnmqtDE0,97184
+natural_pdf/elements/collections.py,sha256=HsNt_4x-yqNI_bDGeNEiih3hotAfrbppmp_O7rq9HGs,107141
 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=LfyB_9DCw5Tzn_G9xsjFz2FfKBOHRqGIND4DQWoA7KM,97324
+natural_pdf/elements/region.py,sha256=XYWUym7hgkzMMfmXw0hEz_iGJ6Sdyf6DRz6XjgMVwN0,97250
 natural_pdf/elements/text.py,sha256=13HvVZGinj2Vm_fFCAnqi7hohtoKvnpCp3VCfkpeAbc,11146
 natural_pdf/export/mixin.py,sha256=L1q3MIEFWuvie4j4_EmW7GT3NerbZ1as0XMUoqTS7gM,5083
 natural_pdf/exporters/__init__.py,sha256=7MnvRLLQdwtg-ULu-8uK8C84GsKiJamyhRw_GgWhw7k,151
 natural_pdf/exporters/base.py,sha256=XhR1xlkHOh7suOuX7mWbsj1h2o1pZNet-OAS5YCJyeI,2115
+natural_pdf/exporters/hocr.py,sha256=wilmVyBgmBNp2ZEdbKijk9ag8E1AGMMl6rBtsAOzp-Y,20201
+natural_pdf/exporters/hocr_font.py,sha256=e9QdxeCExxpY_dpzwGxFlT_3TcvNejw9qpkNc1NVa4Y,4612
+natural_pdf/exporters/original_pdf.py,sha256=vZeqBsCZh3JRRWwtfHzM78fxvhKkAI4QK3LLkeXidUM,5082
 natural_pdf/exporters/paddleocr.py,sha256=BYpdtJI7S8rBkI2dkRESx2epVAZOTfzqU-rjJnUQ5jQ,16249
-natural_pdf/exporters/searchable_pdf.py,sha256=qsaPsnbOOaZHA_aplfZbwQnBoK9KghWm-wzbyRRomeY,16859
+natural_pdf/exporters/searchable_pdf.py,sha256=-sbjjM4oV2YCiJaVKcUIPXjAs94ouXSyOSlAzv_qM7I,16815
 natural_pdf/extraction/manager.py,sha256=mUBbfgLG5Pl31wmajXwyipdEJb_dZ5I-y8GnWw7IzGo,4969
 natural_pdf/extraction/mixin.py,sha256=eKbr70VibpbtfjvCE80lTFuYHzq_BoVtOHjznL_GMRA,11719
 natural_pdf/extraction/result.py,sha256=c1vLguCR6l95cvg-BJJmZvL_MPg2McJaczge55bKZMg,934
@@ -45,7 +48,7 @@ natural_pdf/ocr/engine.py,sha256=ZBC1tZNM5EDbGDJJmZI9mNHr4nCMLEZvUFhiJq8GdF4,874
 natural_pdf/ocr/engine_doctr.py,sha256=519WpvSHgwP6Hv24tci_YHFX7XPlaxOnlREN_YG-Yys,16331
 natural_pdf/ocr/engine_easyocr.py,sha256=9TbxJjmhWFrzM8mcNnZjoRtIDr6gwpuwKm4-Zfub2-8,9281
 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/engine_surya.py,sha256=CQHpPecCYsJsr7pEvEyubAf5FJFs7vFHAm_0cGGr-A4,4839
 natural_pdf/ocr/ocr_factory.py,sha256=gBFXdFs7E4aCynHz06sQsAhaO3s8yhgoFgN5nyxtg9c,5221
 natural_pdf/ocr/ocr_manager.py,sha256=f0q68ynGYVPkF4D3WnufxmHWD5R1jW5Z_1czTEi9JVU,13931
 natural_pdf/ocr/ocr_options.py,sha256=ZvtnFn1kPkFEoWveQ13uy6B-ofquP0gHEi4tBHrjqCE,6438
@@ -73,8 +76,8 @@ natural_pdf/utils/tqdm_utils.py,sha256=wV3RXvqog26eWEFEqjt2LkGnLswmO1GXaVGSqgS7t
 natural_pdf/utils/visualization.py,sha256=30pRWQdsRJh2pSObh-brKVsFgC1n8tHmSrta_UDnVPw,8989
 natural_pdf/widgets/__init__.py,sha256=O2fSDo604wDAP6UwUkmBq3eT91RSqHwBpAOQXq92S8s,214
 natural_pdf/widgets/viewer.py,sha256=dC_hlPlosc08gsDc3bdAa8chOKtAoH9QFU6mrGOG9vE,39532
-natural_pdf-0.1.10.dist-info/licenses/LICENSE,sha256=9zfwINwJlarbDmdh6iJV4QUG54QSJlSAUcnC1YiC_Ns,1074
-natural_pdf-0.1.10.dist-info/METADATA,sha256=gjUsfmnbqrdiHcaH6L1qiw6VX4MBlWjVj5HqlDnhuQY,7401
-natural_pdf-0.1.10.dist-info/WHEEL,sha256=wXxTzcEDnjrTwFYjLPcsW_7_XihufBwmpiBeiXNBGEA,91
-natural_pdf-0.1.10.dist-info/top_level.txt,sha256=Cyw1zmNDlUZfb5moU-WUWGprrwH7ln_8LDGdmMHF1xI,17
-natural_pdf-0.1.10.dist-info/RECORD,,
+natural_pdf-0.1.11.dist-info/licenses/LICENSE,sha256=9zfwINwJlarbDmdh6iJV4QUG54QSJlSAUcnC1YiC_Ns,1074
+natural_pdf-0.1.11.dist-info/METADATA,sha256=HBEH41sOW2opbRoN_yUq8iw3jB2fvdOXEDj0ZGfmw8g,7354
+natural_pdf-0.1.11.dist-info/WHEEL,sha256=0CuiUZ_p9E4cD6NyLD6UG80LBXYyiSYZOKDm5lp32xk,91
+natural_pdf-0.1.11.dist-info/top_level.txt,sha256=Cyw1zmNDlUZfb5moU-WUWGprrwH7ln_8LDGdmMHF1xI,17
+natural_pdf-0.1.11.dist-info/RECORD,,

{natural_pdf-0.1.10.dist-info → natural_pdf-0.1.11.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (80.1.0)
+Generator: setuptools (80.3.1)
 Root-Is-Purelib: true
 Tag: py3-none-any