add tests for recently fixed bugs

Author: stefan6419846

CHANGELOG.md

@@ -1,5 +1,8 @@
 # Development version
 
+* Gracefully handle hOCR files from `kraken` in `hocr_pdf`.
+* Deal with non-integer widths and heights in `hocr_pdf` which would previously lead to some white borders.
+
 # Version 1.1.0 - 2024-07-23
 
 * Fix deprecation warning from `lxml` in `hocr_wordfreq`.

pyproject.toml

@@ -49,6 +49,7 @@ dev = [
     # Tests.
     "beautifulsoup4",
     "requests",
+    "pypdf",
     "importlib-resources; python_version < '3.10'",
     # Linting.
     "flake8",

requirements.txt

@@ -5,3 +5,4 @@ Pillow==11.0.0
 lxml==5.3.0
 python-bidi==0.6.3
 reportlab==4.2.5
+pypdf==5.1.0