diff: Use bytes as much as possible, to prevent possible encoding issues
In practice, this prevents weird codepoints to behave like linebreaks of
sorts.
And while I believe it also decreses memory consume, especially for
larger diffs.
Signed-off-by: Mattia Rizzolo <mattia@debian.org>
Showing
- diffoscope/diff.py 33 additions, 28 deletionsdiffoscope/diff.py
- tests/comparators/test_pdf.py 6 additions, 0 deletionstests/comparators/test_pdf.py
- tests/data/test_weird_non_unicode_chars1.pdf 0 additions, 0 deletionstests/data/test_weird_non_unicode_chars1.pdf
- tests/data/test_weird_non_unicode_chars2.pdf 0 additions, 0 deletionstests/data/test_weird_non_unicode_chars2.pdf
tests/data/test_weird_non_unicode_chars1.pdf
0 → 100644
File added
tests/data/test_weird_non_unicode_chars2.pdf
0 → 100644
File added
Please register or sign in to comment