Chore: Import pikepdf and img2pdf lazily so startup does not load them

src/documents/barcodes.py

@@ -10,9 +10,6 @@ from typing import TYPE_CHECKING
 import regex as regex_mod
 from django.conf import settings
 from pdf2image import convert_from_path
-from pikepdf import Page
-from pikepdf import PasswordError
-from pikepdf import Pdf
 
 from documents.converters import convert_from_tiff_to_pdf
 from documents.data_models import ConsumableDocument
@@ -34,6 +31,7 @@ from documents.utils import maybe_override_pixel_limit
 from paperless.config import BarcodeConfig
 
 if TYPE_CHECKING:
+    from pikepdf import Page
     from PIL import Image
 
 logger = logging.getLogger("paperless.barcodes")
@@ -282,6 +280,8 @@ class BarcodePlugin(ConsumeTaskPlugin):
         # No op if not a TIFF
         self.convert_from_tiff_to_pdf()
 
+        from pikepdf import PasswordError
+
         try:
             self.barcodes = scan_pdf(
                 self.pdf_file,
@@ -433,6 +433,8 @@ class BarcodePlugin(ConsumeTaskPlugin):
 
         document_paths = []
         fname: str = self.input_doc.original_file.stem
+        from pikepdf import Pdf
+
         with Pdf.open(self.pdf_file) as input_pdf:
             # Start with an empty document
             current_document: list[Page] = []
@@ -516,6 +518,8 @@ def scan_pdf(pdf_path: Path, settings: BarcodeConfig, work_dir: Path) -> list[Ba
     """
     barcodes: list[Barcode] = []
 
+    from pikepdf import Pdf
+
     with Pdf.open(pdf_path) as pdf:
         num_of_pages = len(pdf.pages)
     logger.debug(f"PDF has {num_of_pages} pages")

src/documents/converters.py

@@ -1,6 +1,5 @@
 from pathlib import Path
 
-import img2pdf
 from PIL import Image
 
 from documents.utils import copy_basic_file_stats
@@ -17,6 +16,8 @@ def convert_from_tiff_to_pdf(tiff_path: Path, target_directory: Path) -> Path:
 
     Returns the path of the PDF created.
     """
+    import img2pdf
+
     # override pixel setting if needed
     maybe_override_pixel_limit()
 

src/documents/double_sided.py

@@ -6,7 +6,6 @@ from pathlib import Path
 from typing import Final
 
 from django.conf import settings
-from pikepdf import Pdf
 
 from documents.consumer import ConsumerError
 from documents.converters import convert_from_tiff_to_pdf
@@ -85,6 +84,8 @@ class CollatePlugin(NoCleanupPluginMixin, NoSetupPluginMixin, ConsumeTaskPlugin)
                 valid_staging_exists = True
 
         if valid_staging_exists:
+            from pikepdf import Pdf
+
             try:
                 # Collate pages from second PDF in reverse order
                 with Pdf.open(staging) as pdf1, Pdf.open(pdf_file) as pdf2:

src/documents/pdf_ops.py

@@ -14,8 +14,6 @@ from __future__ import annotations
 from typing import TYPE_CHECKING
 from typing import NamedTuple
 
-import pikepdf
-
 if TYPE_CHECKING:
     from collections.abc import Callable
     from collections.abc import Iterable
@@ -43,6 +41,8 @@ def rotate_pdf(src: Path, dst: Path, degrees: int) -> None:
     Rotate every page relatively on the opened document, not a rebuild, so Info,
     XMP and outlines are kept. ``src`` is not modified.
     """
+    import pikepdf
+
     with pikepdf.open(src) as pdf:
         for page in pdf.pages:
             page.rotate(degrees, relative=True)
@@ -58,6 +58,8 @@ def remove_pages(src: Path, dst: Path, pages: Iterable[int]) -> None:
     Duplicates are ignored. Pages are removed highest first so earlier removals
     never shift the index of later ones.
     """
+    import pikepdf
+
     unique = sorted(set(pages))
     _require_positive(unique)
     with pikepdf.open(src) as pdf:
@@ -81,6 +83,8 @@ def build_pdfs(
     (Info, XMP, outlines) is not carried over. Returns the written paths in output
     order.
     """
+    import pikepdf
+
     for specs, _ in outputs:
         _require_positive(spec.page for spec in specs)
 
@@ -140,6 +144,8 @@ def needs_decrypt(src: Path) -> bool:
     True if ``src`` is encrypted. A PDF that needs a password to open at all
     counts as encrypted.
     """
+    import pikepdf
+
     try:
         with pikepdf.open(src) as pdf:
             return bool(pdf.is_encrypted)
@@ -154,6 +160,8 @@ def decrypt_pdf(src: Path, make_dst: Callable[[], Path], password: str) -> Path:
     ``make_dst`` is only called once the password has been accepted, so a wrong
     password never leaves a destination behind.
     """
+    import pikepdf
+
     with pikepdf.open(src, password=password) as pdf:
         pdf.remove_unreferenced_resources()
         dst = make_dst()
@@ -170,6 +178,8 @@ class PdfMerger:
     """
 
     def __init__(self) -> None:
+        import pikepdf
+
         self._pdf = pikepdf.new()
         self._version: str = self._pdf.pdf_version
 
@@ -185,6 +195,8 @@ class PdfMerger:
         self._pdf.close()
 
     def add(self, path: Path) -> None:
+        import pikepdf
+
         with pikepdf.open(str(path)) as pdf:
             self._version = max(self._version, pdf.pdf_version)
             self._pdf.pages.extend(pdf.pages)