Version 2.1.0, 2022-06-06

[MartinThoma]

What's Changed

The highlight of the 2.1.0 release is the most massive improvement to the
text extraction capabilities of PyPDF2 since 2016 🥳🎊 A very big thank you goes
to pubpub-zz who took a lot of time and
knowledge about the PDF format to finally get those improvements into PyPDF2.
Thank you 🤗💚

In case the new function causes any issues, you can use _extract_text_old
for the old functionality. Please also open a bug ticket in that case.

There were several people who have attempted to bring similar improvements to
PyPDF2. All of those were valuable. The main reason why they didn't get merged
is the big amount of open PRs / issues. pubpub-zz was the most comprehensive
PR which also incorporated the latest changes of PyPDF2 2.0.0.

Thank you to VictorCarlquist for #858 and
asabramo for #464 🤗

New Features (ENH)

• Massive text extraction improvement (Pubpub zz extract text #924). Closed many open issues:
  • Exceptions / missing spaces in extract_text() method (Exceptions / missing spaces in extract_text() method #17) 🕺
    • Whitespace issues in extract_text() (Whitespace issues in extract_text() #42) 💃
    • pypdf2 reads the hifenated words in a new line (pypdf2 reads the hifenated words in a new line #246)
  • PyPDF2 failing to read unicode character (PyPDF2 failing to read unicode character #37)
    • Unable to read bullets (Unable to read bullets  #230)
  • ExtractText yields nothing for apparently good PDF (ExtractText yields nothing for apparently good PDF #168) 🎉
  • Encoding issue in extract_text() (Encoding issue in extract_text() #235)
  • extractText() doesn't work on Chinese PDF (extractText() doesn't work on Chinese PDF #252)
  • encoding error (encoding error #260)
  • Trouble with apostophes in names in text "O'Doul" (Trouble with apostophes in names in text "O'Doul" #384)
  • extract_text works for some PDF files, but not the others (extract_text works for some PDF files, but not the others #437)
  • Euro sign not being recognized by extractText (Euro sign not being recognized by extractText #443)
  • Failed extracting text from French texts (Failed extracting text from French texts #524)
  • extract_text doesn't extract ligatures correctly (extract_text doesn't extract ligatures correctly #598)
  • reading spanish text - mark convert issue (reading spanish text - mark convert issue  #635)
  • Read PDF changed from text to random symbols (Read PDF changed from text to random symbols #654)
  • .extractText() reads / as 1. (.extractText() reads / as 1. #789)
• Update glyphlist (ENH: Update glyphlist #947) - inspired by Advanced text extraction #464
• Allow adding PageRange objects (ENH: Allow adding PageRange objects #948)

Bug Fixes (BUG)

• Delete .python-version file (Delete .python-version #944)
• Compare StreamObject.decoded_self with None (corrected the null evaluation of decoded stream objects. #931)

Robustness (ROB)

• Fix some conversion errors on non conform PDF (Fix some conversion errors on non conform PDF #932)

Documentation (DOC)

• Elaborate on PDF text extraction difficulties (DOC: Elaborate on PDF text extraction difficulties #939)
• Add logo (DOC: Add logo #942)
• rotate vs Transformation().rotate (DOC: rotate vs Transformation().rotate #937)
• Example how to use PyPDF2 with AWS S3 (DOC: Example how to use PyPDF2 with AWS S3 #938)
• How to deprecate (DOC: How to deprecate #930)
• Fix typos on robustness page (DOC: Fix typos on robustness page #935)
• Remove scripts (pdfcat) from docs (DOC: Remove scripts (pdfcat) from docs #934)

Developer Experience (DEV)

• Ignore .python-version file
• Mark deprecated code with no-cover (DEV: Mark deprecated code with no-cover #943)
• Automatically create Github releases from tags (ci: release to pypi and github on tagged pushes. #870)

Testing (TST)

• Text extraction for non-latin alphabets (TST: Text extraction for non-latin alphabets #954)
• Ignore PdfReadWarning in benchmark (TST: Ignore PdfReadWarning in benchmark #949)
• writer.remove_text (TST: writer.remove_text #946)
• Add test for Tree and _security (TST: Add test for Tree and _security #945)

Code Style (STY)

• black, isort, Flake8, splitting buildCharMap (STY: Apply black and isort #950)

Full Changelog: 2.0.0...2.1.0

New Contributors

• @chrisdonlan made their first contribution in corrected the null evaluation of decoded stream objects. #931
• @2e0byo made their first contribution in ci: release to pypi and github on tagged pushes. #870
• @HarHarLinks made their first contribution in Delete .python-version #944

Full Changelog: 2.0.0...2.1.0

---

This discussion was created from the release Version 2.1.0, 2022-06-06.

[MartinThoma]

Making a bit of advertisement that PyPDF2 is now better than ever: https://www.linkedin.com/posts/martin-thoma_github-py-pdfbenchmarks-benchmarking-activity-6939589461190062080-TBwr and https://twitter.com/_martinthoma/status/1533823440758661121 😄

@pubpub-zz pushed PyPDF2 from 86% to 96% in my benchmark - according to the benchmark, we are now be better than pdftotext and pdfminer.six in text extraction 😄