+ From a266f068c03fc404316f6f7d9c7607b1365efdf2 Mon Sep 17 00:00:00 2001

+ From: =?UTF-8?q?Miro=20Hron=C4=8Dok?= <miro@hroncok.cz>

+ Date: Sat, 16 Jul 2022 07:55:47 +0200

+ Subject: [PATCH] Low effort port to charset_normalizer

+ 

+ See https://github.com/PyYoshi/cChardet/issues/77

+ ---

+  src/calibre/ebooks/chardet.py | 2 +-

+  src/calibre/test_build.py     | 2 +-

+  2 files changed, 2 insertions(+), 2 deletions(-)

+ 

+ diff --git a/src/calibre/ebooks/chardet.py b/src/calibre/ebooks/chardet.py

+ index 53fe6c51087..92a8accdafb 100644

+ --- a/src/calibre/ebooks/chardet.py

+ +++ b/src/calibre/ebooks/chardet.py

+ @@ -103,7 +103,7 @@ def substitute_entites(raw):

+  

+  

+  def detect(bytestring):

+ -    from cchardet import detect as implementation

+ +    from charset_normalizer import detect as implementation

+      ans = implementation(bytestring)

+      enc = ans.get('encoding')

+      if enc:

+ diff --git a/src/calibre/test_build.py b/src/calibre/test_build.py

+ index c0f50292160..ba89dea28d0 100644

+ --- a/src/calibre/test_build.py

+ +++ b/src/calibre/test_build.py

+ @@ -73,7 +73,7 @@ def test_pychm(self):

+          del CHMFile, chmlib

+  

+      def test_chardet(self):

+ -        from cchardet import detect

+ +        from charset_normalizer import detect

+          raw = 'mūsi Füße'.encode()

+          data = detect(raw)

+          self.assertEqual(data['encoding'].lower(), 'utf-8')

+ # Use charset_normalizer instead of cchardet

+ # Downstream only, upstream plans to fork and maintain cchardet instead

+ # see https://bugzilla.redhat.com/show_bug.cgi?id=2021804

+ Patch4:         https://github.com/kovidgoyal/calibre/pull/1690.patch

+ 

+ BuildRequires:  python3dist(charset-normalizer)

+ Requires:       python3dist(charset-normalizer)