mirror of
https://github.com/python-telegram-bot/python-telegram-bot.git
synced 2024-11-22 23:27:49 +01:00
335772568f
Co-authored-by: Hinrich Mahler <22366557+Bibo-Joshi@users.noreply.github.com>
76 lines
2.6 KiB
Python
76 lines
2.6 KiB
Python
#!/usr/bin/env python
|
|
#
|
|
# A library that provides a Python interface to the Telegram Bot API
|
|
# Copyright (C) 2015-2022
|
|
# Leandro Toledo de Souza <devs@python-telegram-bot.org>
|
|
#
|
|
# This program is free software: you can redistribute it and/or modify
|
|
# it under the terms of the GNU Lesser Public License as published by
|
|
# the Free Software Foundation, either version 3 of the License, or
|
|
# (at your option) any later version.
|
|
#
|
|
# This program is distributed in the hope that it will be useful,
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
# GNU Lesser Public License for more details.
|
|
#
|
|
# You should have received a copy of the GNU Lesser Public License
|
|
# along with this program. If not, see [http://www.gnu.org/licenses/].
|
|
|
|
import pytest
|
|
from flaky import flaky
|
|
|
|
from telegram import ForceReply, ReplyKeyboardRemove
|
|
|
|
|
|
@pytest.fixture(scope='class')
|
|
def force_reply():
|
|
return ForceReply(
|
|
TestForceReply.selective,
|
|
TestForceReply.input_field_placeholder,
|
|
)
|
|
|
|
|
|
class TestForceReply:
|
|
force_reply = True
|
|
selective = True
|
|
input_field_placeholder = 'force replies can be annoying if not used properly'
|
|
|
|
def test_slot_behaviour(self, force_reply, mro_slots):
|
|
for attr in force_reply.__slots__:
|
|
assert getattr(force_reply, attr, 'err') != 'err', f"got extra slot '{attr}'"
|
|
assert len(mro_slots(force_reply)) == len(set(mro_slots(force_reply))), "duplicate slot"
|
|
|
|
@flaky(3, 1)
|
|
async def test_send_message_with_force_reply(self, bot, chat_id, force_reply):
|
|
message = await bot.send_message(chat_id, 'text', reply_markup=force_reply)
|
|
|
|
assert message.text == 'text'
|
|
|
|
def test_expected(self, force_reply):
|
|
assert force_reply.force_reply == self.force_reply
|
|
assert force_reply.selective == self.selective
|
|
assert force_reply.input_field_placeholder == self.input_field_placeholder
|
|
|
|
def test_to_dict(self, force_reply):
|
|
force_reply_dict = force_reply.to_dict()
|
|
|
|
assert isinstance(force_reply_dict, dict)
|
|
assert force_reply_dict['force_reply'] == force_reply.force_reply
|
|
assert force_reply_dict['selective'] == force_reply.selective
|
|
assert force_reply_dict['input_field_placeholder'] == force_reply.input_field_placeholder
|
|
|
|
def test_equality(self):
|
|
a = ForceReply(True, 'test')
|
|
b = ForceReply(False, 'pass')
|
|
c = ForceReply(True)
|
|
d = ReplyKeyboardRemove()
|
|
|
|
assert a != b
|
|
assert hash(a) != hash(b)
|
|
|
|
assert a == c
|
|
assert hash(a) == hash(c)
|
|
|
|
assert a != d
|
|
assert hash(a) != hash(d)
|