patches for ebuild from versions in fedora package

mail-filter/pyzor/files/65.patch

@@ -0,0 +1,42 @@
+From 6332a429ed415187599ecce7d8a169ee19f0bbe5 Mon Sep 17 00:00:00 2001
+From: Michael Orlitzky <michael@orlitzky.com>
+Date: Sun, 4 Mar 2018 17:34:33 -0500
+Subject: [PATCH] scripts/pyzor: read stdin as binary in _get_input_msg().
+
+Reading stdin in python-3.x is done as text, with a best-guess
+encoding. But this can go awry: for example, if an iso-8859-1 message
+is passed in and if python guesses the "utf-8" encoding, then read()
+will fail with a UnicodeDecodeError on non-ASCII characters. For
+example, the "copyright" symbol is a single byte 0xa9 in iso-8859-1,
+and the utf-8 decoder can't handle it:
+
+  UnicodeDecodeError: 'utf-8' codec can't decode byte 0xa9... invalid
+  start byte
+
+Instead -- and as was done in python-2.x -- we can read stdin as
+binary using the new get_binary_stdin() function. Afterwards, we use
+email.message_from_bytes() instead of the email.message_from_file()
+constructor to parse the byte data. The resulting function is able to
+correctly parse these messages.
+
+Closes: https://github.com/SpamExperts/pyzor/issues/64
+---
+ scripts/pyzor | 5 ++++-
+ 1 file changed, 4 insertions(+), 1 deletion(-)
+
+diff --git a/scripts/pyzor b/scripts/pyzor
+index 567a7f9..1ba632f 100755
+--- a/scripts/pyzor
++++ b/scripts/pyzor
+@@ -171,7 +171,10 @@ def _get_input_digests(dummy):
+ 
+ 
+ def _get_input_msg(digester):
+-    msg = email.message_from_file(sys.stdin)
++    # Read and process stdin as bytes because we don't know its
++    # encoding. Python-3.x will try to guess -- and can sometimes
++    # guess wrong -- leading to decoding errors in read().
++    msg = email.message_from_bytes(get_binary_stdin().read())
+     digested = digester(msg).value
+     yield digested
+ 

mail-filter/pyzor/pyzor-9999.ebuild

@@ -0,0 +1,69 @@
+# Copyright 1999-2020 Gentoo Authors
+# Distributed under the terms of the GNU General Public License v2
+
+EAPI=6
+
+PYTHON_COMPAT=( python3_6 python3_7 python3_8 )
+inherit distutils-r1 git-r3
+
+DESCRIPTION="A distributed, collaborative spam detection and filtering network"
+HOMEPAGE="https://github.com/SpamExperts/pyzor"
+EGIT_REPO_URI="https://github.com/majn/telegram-purple"
+EGIT_BRANCH="master"
+
+LICENSE="GPL-2"
+SLOT="0"
+KEYWORDS="~alpha ~amd64 ~hppa ~ia64 ~ppc ~ppc64 ~sparc ~x86 ~amd64-linux ~x86-linux"
+
+IUSE="doc gdbm gevent pyzord redis test"
+# The test suite is py2-only
+RESTRICT="test"
+
+# The mysql-python library is always required for the MySQL engine. We
+# depend on it conditionally here because otherwise repoman will balk at
+# the potential conflict between PYTHON_TARGETS and USE=mysql. But as a
+# result, if you try to use the MySQL engine with python-3.x, it just
+# won't work because you'll be missing the library.
+RDEPEND="
+	pyzord? (
+		gdbm? ( $(python_gen_impl_dep 'gdbm') )
+		redis? ( dev-python/redis-py[${PYTHON_USEDEP}] )
+		gevent? ( dev-python/gevent[${PYTHON_USEDEP}] )
+	)"
+DEPEND="dev-python/setuptools[${PYTHON_USEDEP}]
+	doc? ( dev-python/sphinx[${PYTHON_USEDEP}] )
+	test? ( ${RDEPEND} )"
+
+# TODO: maybe upstream would support skipping tests for which the
+# dependencies are missing?
+REQUIRED_USE="pyzord? ( || ( gdbm redis ) )
+	test? ( gdbm redis )"
+
+#PATCHES=(
+#	"${FILESDIR}/65.patch"
+#)
+
+python_test() {
+	PYTHONPATH=. "${PYTHON}" ./tests/unit/__init__.py
+}
+
+python_compile_all() {
+	use doc && emake -C docs html
+}
+
+python_install_all() {
+	use doc && HTML_DOCS=( docs/.build/html/. )
+	distutils-r1_python_install_all
+}
+
+src_install() {
+	distutils-r1_src_install
+
+	if use pyzord; then
+		dodir /usr/sbin
+		mv "${D}"usr/bin/pyzord* "${ED}usr/sbin" \
+		   || die "failed to relocate pyzord"
+	else
+		rm "${D}"usr/bin/pyzord* || die "failed to remove pyzord"
+	fi
+}