PKG: Declare Pillow as optional dependency

PyPDF2/_page.py

@@ -373,6 +373,8 @@ def images(self) -> List[File]:
         """
         Get a list of all images of the page.
 
+        This requires pillow. You can install it via 'pip install PyPDF2[image]'.
+
         For the moment, this does NOT include inline images. They will be added
         in future.
         """

PyPDF2/filters.py

@@ -564,7 +564,13 @@ def _xobj_to_image(x_object_obj: Dict[str, Any]) -> Tuple[Optional[str], bytes]:
 
     :return: Tuple[file extension, bytes]
     """
-    from PIL import Image
+    try:
+        from PIL import Image
+    except ImportError:
+        raise ImportError(
+            "pillow is required to do image extraction. "
+            "It can be installed via 'pip install PyPDF2[image]'"
+        )
 
     size = (x_object_obj[IA.WIDTH], x_object_obj[IA.HEIGHT])
     data = x_object_obj.get_data()  # type: ignore

docs/user/encryption-decryption.md

@@ -1,7 +1,7 @@
 # Encryption and Decryption of PDFs
 
 Please see the note in the
-[Installation doc](https://pypdf2.readthedocs.io/en/latest/user/installation.html)
+[Installation doc](installation.html)
 for installing the extra dependencies if interacting with PDFs that use AES.
 
 ## Encrypt

docs/user/extract-images.md

@@ -1,5 +1,8 @@
 # Extract Images
 
+> Please note: In order to use the following code you need to install optional
+> dependencies, see [installation guide](installation.html).
+
 Every page of a PDF document can contain an arbitrary amount of images.
 The names of the files may not be unique.
 

docs/user/installation.md

@@ -20,6 +20,20 @@ install PyPDF2 for your current user:
 pip install --user PyPDF2
 ```
 
+### Optional dependencies
+
+PyPDF2 tries to be as self-contained as possible, but for some tasks the amout
+of work to properly maintain the code would be too high. This is especially the
+case for cryptography and image formats.
+
+If you simply want to unstall all optional dependencies, please run
+
+```
+pip install PyPDF2[full]
+```
+
+Alternatively, you can install just some:
+
 If you plan to use PyPDF2 for encrypting or decrypting PDFs that use AES, you
 will need to install some extra dependencies. Encryption using RC4 is supported
 using the regular installation.
@@ -28,6 +42,11 @@ using the regular installation.
 pip install PyPDF2[crypto]
 ```
 
+If you plan to use image extraction, you need Pillow:
+
+```
+pip install PyPDF2[image]
+```
 
 ## Python Version Support
 

setup.cfg

@@ -42,7 +42,9 @@ install_requires =
     dataclasses; python_version < '3.7'
 
 [options.extras_require]
+full = PyCryptodome; Pillow
 crypto = PyCryptodome
+image = Pillow
 
 [mutmut]
 backup = False

tests/test_reader.py

@@ -5,7 +5,6 @@
 from pathlib import Path
 
 import pytest
-from PIL import Image
 
 from PyPDF2 import PdfReader
 from PyPDF2._reader import convert_to_int, convertToInt
@@ -188,6 +187,8 @@ def test_get_outline(src, outline_elements):
     ],
 )
 def test_get_images(src, expected_images):
+    from PIL import Image
+
     src_abs = RESOURCE_ROOT / src
     reader = PdfReader(src_abs)