Full credit goes to exiledkingcc

Co-authored-by: exiledkingcc <exiledkingcc@gmail.com>

This makes PR #1816 smaller

Co-authored-by: exiledkingcc <exiledkingcc@gmail.com>

Pass it through to compute_U_value / compute_O_value

Also fix a bug I (Martin) just introduced in
compute_O_value by setting the token_bytes=4 and not 16.

All credit for refactoring goes to exiledkingcc

Co-authored-by: exiledkingcc <exiledkingcc@gmail.com>

* Rename parameters of Encryption class
* Rename attribute key_size to Length
* Add P attribute explicitly instead of using "entry"
* Add EncryptMetadata attribute
* Add values attribute

Co-authored-by: exiledkingcc <exiledkingcc@gmail.com>

Fixes #1814

Fixes #1809

…1844)

Closes #1832

/Type is mandatory in page tree nodes according to the the PDF specification. Hence dealing with such files is a robustness improvements.
Acrobat Reader can open such PDF documents as well.

Fixes #500

Closes #1786

Closes #1846

Closes #1183

…1828)

Fixes #1825

New Features (ENH)
-  Simplify metadata input (Document Information Dictionary) (#1851)
-  Extend cmap compatibilty to GBK_EUC_H/V (#1812)

Bug Fixes (BUG)
-  Prevent infinite loop when no character follows after a comment (#1828)
-  get_contents does not return ContentStream (#1847)
-  Accept XYZ destination with zoom missing (default to zoom=0.0) (#1844)
-  Cope with 1 Bit images (#1815)

Robustness (ROB)
-  Handle missing /Type entry in Page tree (#1845)

Documentation (DOC)
-  Expand file size explanations (#1835)
-  Add comparison with pdfplumber (#1837)
-  Clarify that PyPDF2 is dead (#1827)
-  Add Hunter King as Contributor for #1806

Maintenance (MAINT)
-  Refactor internal Encryption class (#1821)
-  Add R parameter to generate_values (#1820)
-  Make encryption_key parameter of write_to_stream optional (#1819)
-  Prepare for adding AES enryption support (#1818)

Code Style (STY):
-  Iterate directly over the list instead of using range (#1839)
-  Minor refactorings in _encryption.py (#1822)

[Full Changelog](3.8.1...3.9.0)