import copy
import logging
import os
import tempfile
import uuid
import weakref

import numpy as np
import pytest
from numpy.testing import assert_array_equal, assert_equal

from brian2 import (
    BrianLogger,
    BrianObject,
    Clock,
    Hz,
    MagicError,
    MagicNetwork,
    Network,
    NetworkOperation,
    NeuronGroup,
    PoissonGroup,
    PopulationRateMonitor,
    Quantity,
    SpikeGeneratorGroup,
    SpikeMonitor,
    StateMonitor,
    Synapses,
    TimedArray,
    collect,
    defaultclock,
    magic_network,
    ms,
    network_operation,
    prefs,
    profiling_summary,
    restore,
    run,
    second,
    start_scope,
    stop,
    store,
    us,
)
from brian2.core.network import schedule_propagation_offset, scheduling_summary
from brian2.devices.device import (
    Device,
    RuntimeDevice,
    all_devices,
    device,
    get_device,
    reinit_and_delete,
    reset_device,
    set_device,
)
from brian2.tests.utils import assert_allclose
from brian2.utils.logger import catch_logs


@pytest.mark.codegen_independent
def test_incorrect_network_use():
    """Test some wrong uses of `Network` and `MagicNetwork`"""
    with pytest.raises(TypeError):
        Network(name="mynet", anotherkwd="does not exist")
    with pytest.raises(TypeError):
        Network("not a BrianObject")
    net = Network()
    with pytest.raises(TypeError):
        net.add("not a BrianObject")
    with pytest.raises(ValueError):
        MagicNetwork()
    G = NeuronGroup(10, "v:1")
    net.add(G)
    with pytest.raises(TypeError):
        net.remove(object())
    with pytest.raises(MagicError):
        magic_network.add(G)
    with pytest.raises(MagicError):
        magic_network.remove(G)


@pytest.mark.codegen_independent
def test_network_contains():
    """
    Test `Network.__contains__`.
    """
    G = NeuronGroup(1, "v:1", name="mygroup")
    net = Network(G)
    assert "mygroup" in net
    assert "neurongroup" not in net


@pytest.mark.codegen_independent
def test_empty_network():
    # Check that an empty network functions correctly
    net = Network()
    net.run(1 * second)


class Counter(BrianObject):
    add_to_magic_network = True

    def __init__(self, **kwds):
        super().__init__(**kwds)
        self.count = 0
        self.state = {"state": 0}

    def get_states(self, *args, **kwds):
        return dict(self.state)

    def set_states(self, values, *args, **kwds):
        for k, v in values.items():
            self.state[k] = v

    def run(self):
        self.count += 1


class CounterWithContained(Counter):
    add_to_magic_network = True

    def __init__(self, **kwds):
        super().__init__(**kwds)
        self.sub_counter = Counter()
        self.contained_objects.append(self.sub_counter)


@pytest.mark.codegen_independent
def test_network_single_object():
    # Check that a network with a single object functions correctly
    x = Counter()
    net = Network(x)
    net.run(1 * ms)
    assert_equal(x.count, 10)


@pytest.mark.codegen_independent
def test_network_two_objects():
    # Check that a network with two objects and the same clock function correctly
    x = Counter(order=5)
    y = Counter(order=6)
    net = Network()
    net.add([x, [y]])  # check that a funky way of adding objects work correctly
    net.run(1 * ms)
    assert_equal(len(net.objects), 2)
    assert_equal(x.count, 10)
    assert_equal(y.count, 10)


@pytest.mark.codegen_independent
def test_network_from_dict():
    # Check that a network from a dictionary works
    x = Counter()
    y = Counter()
    d = dict(a=x, b=y)
    net = Network()
    net.add(d)
    net.run(1 * ms)
    assert_equal(len(net.objects), 2)
    assert_equal(x.count, 10)
    assert_equal(y.count, 10)


class NameLister(BrianObject):
    add_to_magic_network = True
    updates = []

    def __init__(self, **kwds):
        super().__init__(**kwds)

    def run(self):
        NameLister.updates.append(self.name)


@pytest.mark.codegen_independent
def test_network_different_clocks():
    NameLister.updates[:] = []
    # Check that a network with two different clocks functions correctly
    x = NameLister(name="x", dt=0.1 * ms, order=0)
    y = NameLister(name="y", dt=1 * ms, order=1)
    net = Network(x, y)
    net.run(100 * second + defaultclock.dt, report="text")
    updates = "".join(NameLister.updates)[2:]  # ignore the first time step
    assert updates == ("xxxxxxxxxxy" * 100000)


@pytest.mark.codegen_independent
def test_network_different_when():
    # Check that a network with different when attributes functions correctly
    NameLister.updates[:] = []
    x = NameLister(name="x", when="start")
    y = NameLister(name="y", when="end")
    net = Network(x, y)
    net.run(0.3 * ms)
    assert_equal("".join(NameLister.updates), "xyxyxy")


@pytest.mark.codegen_independent
def test_network_default_schedule():
    net = Network()
    assert net.schedule == [
        "start",
        "groups",
        "thresholds",
        "synapses",
        "resets",
        "end",
    ]
    # Set the preference and check that the change is taken into account
    prefs.core.network.default_schedule = list(
        reversed(["start", "groups", "thresholds", "synapses", "resets", "end"])
    )
    assert net.schedule == list(
        reversed(["start", "groups", "thresholds", "synapses", "resets", "end"])
    )


@pytest.mark.codegen_independent
def test_network_schedule_change():
    # Check that a changed schedule is taken into account correctly
    NameLister.updates[:] = []
    x = NameLister(name="x", when="thresholds")
    y = NameLister(name="y", when="resets")
    net = Network(x, y)
    net.run(0.3 * ms)
    assert_equal("".join(NameLister.updates), "xyxyxy")
    NameLister.updates[:] = []
    net.schedule = ["start", "groups", "synapses", "resets", "thresholds", "end"]
    net.run(0.3 * ms)
    assert_equal("".join(NameLister.updates), "yxyxyx")


@pytest.mark.codegen_independent
def test_network_before_after_schedule():
    # Test that before... and after... slot names can be used
    NameLister.updates[:] = []
    x = NameLister(name="x", when="before_resets")
    y = NameLister(name="y", when="after_thresholds")
    net = Network(x, y)
    net.schedule = ["thresholds", "resets", "end"]
    net.run(0.3 * ms)
    assert_equal("".join(NameLister.updates), "yxyxyx")


@pytest.mark.codegen_independent
def test_network_custom_slots():
    # Check that custom slots can be inserted into the schedule
    NameLister.updates[:] = []
    x = NameLister(name="x", when="thresholds")
    y = NameLister(name="y", when="in_between")
    z = NameLister(name="z", when="resets")
    net = Network(x, y, z)
    net.schedule = [
        "start",
        "groups",
        "thresholds",
        "in_between",
        "synapses",
        "resets",
        "end",
    ]
    net.run(0.3 * ms)
    assert_equal("".join(NameLister.updates), "xyzxyzxyz")


@pytest.mark.codegen_independent
def test_network_incorrect_schedule():
    # Test that incorrect arguments provided to schedule raise errors
    net = Network()
    # net.schedule = object()
    with pytest.raises(TypeError):
        setattr(net, "schedule", object())
    # net.schedule = 1
    with pytest.raises(TypeError):
        setattr(net, "schedule", 1)
    # net.schedule = {'slot1', 'slot2'}
    with pytest.raises(TypeError):
        setattr(net, "schedule", {"slot1", "slot2"})
    # net.schedule = ['slot', 1]
    with pytest.raises(TypeError):
        setattr(net, "schedule", ["slot", 1])
    # net.schedule = ['start', 'after_start']
    with pytest.raises(ValueError):
        setattr(net, "schedule", ["start", "after_start"])
    # net.schedule = ['before_start', 'start']
    with pytest.raises(ValueError):
        setattr(net, "schedule", ["before_start", "start"])


@pytest.mark.codegen_independent
def test_schedule_warning():
    previous_device = get_device()
    from uuid import uuid4

    # TestDevice1 supports arbitrary schedules, TestDevice2 does not
    class TestDevice1(Device):
        # These functions are needed during the setup of the defaultclock
        def get_value(self, var):
            return np.array([0.0001])

        def add_array(self, var):
            pass

        def init_with_zeros(self, var, dtype):
            pass

        def fill_with_array(self, var, arr):
            pass

    class TestDevice2(TestDevice1):
        def __init__(self):
            super().__init__()
            self.network_schedule = [
                "start",
                "groups",
                "synapses",
                "thresholds",
                "resets",
                "end",
            ]

    # Unique names are important for getting the warnings again for multiple
    # runs of the test suite
    name1 = f"testdevice_{str(uuid4())}"
    name2 = f"testdevice_{str(uuid4())}"
    all_devices[name1] = TestDevice1()
    all_devices[name2] = TestDevice2()

    set_device(name1)
    assert schedule_propagation_offset() == 0 * ms
    net = Network()
    assert schedule_propagation_offset(net) == 0 * ms

    # Any schedule should work
    net.schedule = list(reversed(net.schedule))
    with catch_logs() as l:
        net.run(0 * ms)
        assert len(l) == 0, "did not expect a warning"

    assert schedule_propagation_offset(net) == defaultclock.dt

    set_device(name2)
    assert schedule_propagation_offset() == defaultclock.dt

    # Using the correct schedule should work
    net.schedule = ["start", "groups", "synapses", "thresholds", "resets", "end"]
    with catch_logs() as l:
        net.run(0 * ms)
        assert len(l) == 0, "did not expect a warning"
    assert schedule_propagation_offset(net) == defaultclock.dt

    # Using another (e.g. the default) schedule should raise a warning
    net.schedule = None
    with catch_logs() as l:
        net.run(0 * ms)
        assert len(l) == 1 and l[0][1].endswith("schedule_conflict")
    reset_device(previous_device)


@pytest.mark.codegen_independent
def test_scheduling_summary_magic():
    basename = f"name{str(uuid.uuid4()).replace('-', '_')}"
    group = NeuronGroup(
        10, "dv/dt = -v/(10*ms) : 1", threshold="v>1", reset="v=1", name=basename
    )
    group.run_regularly("v = rand()", dt=defaultclock.dt * 10, when="end")
    state_mon = StateMonitor(group, "v", record=True, name=f"{basename}_sm")
    inactive_state_mon = StateMonitor(
        group, "v", record=True, name=f"{basename}_sm_ia", when="after_end"
    )
    inactive_state_mon.active = False
    summary_before = scheduling_summary()

    assert [entry.name for entry in summary_before.entries] == [
        f"{basename}_sm",
        f"{basename}_stateupdater",
        f"{basename}_spike_thresholder",
        f"{basename}_spike_resetter",
        f"{basename}_run_regularly",
        f"{basename}_sm_ia",
    ]
    assert [entry.when for entry in summary_before.entries] == [
        "start",
        "groups",
        "thresholds",
        "resets",
        "end",
        "after_end",
    ]
    assert [entry.dt for entry in summary_before.entries] == [
        defaultclock.dt,
        defaultclock.dt,
        defaultclock.dt,
        defaultclock.dt,
        defaultclock.dt * 10,
        defaultclock.dt,
    ]
    assert [entry.active for entry in summary_before.entries] == [
        True,
        True,
        True,
        True,
        True,
        False,
    ]
    assert len(str(summary_before))
    assert len(summary_before._repr_html_())
    run(defaultclock.dt)
    summary_after = scheduling_summary()
    assert str(summary_after) == str(summary_before)
    assert summary_after._repr_html_() == summary_before._repr_html_()


@pytest.mark.codegen_independent
def test_scheduling_summary():
    basename = f"name{str(uuid.uuid4()).replace('-', '_')}"
    group = NeuronGroup(
        10, "dv/dt = -v/(10*ms) : 1", threshold="v>1", reset="v=1", name=basename
    )
    group.run_regularly("v = rand()", dt=defaultclock.dt * 10, when="end")
    state_mon = StateMonitor(group, "v", record=True, name=f"{basename}_sm")
    inactive_state_mon = StateMonitor(
        group, "v", record=True, name=f"{basename}_sm_ia", when="after_end"
    )
    inactive_state_mon.active = False

    @network_operation(name=f"{basename}_net_op", when="before_end")
    def foo():
        pass

    net = Network(group, state_mon, inactive_state_mon, foo)
    summary_before = scheduling_summary(net)
    assert [entry.name for entry in summary_before.entries] == [
        f"{basename}_sm",
        f"{basename}_stateupdater",
        f"{basename}_spike_thresholder",
        f"{basename}_spike_resetter",
        f"{basename}_net_op",
        f"{basename}_run_regularly",
        f"{basename}_sm_ia",
    ]
    assert [entry.when for entry in summary_before.entries] == [
        "start",
        "groups",
        "thresholds",
        "resets",
        "before_end",
        "end",
        "after_end",
    ]
    assert [entry.dt for entry in summary_before.entries] == [
        defaultclock.dt,
        defaultclock.dt,
        defaultclock.dt,
        defaultclock.dt,
        defaultclock.dt,
        defaultclock.dt * 10,
        defaultclock.dt,
    ]
    assert [entry.active for entry in summary_before.entries] == [
        True,
        True,
        True,
        True,
        True,
        True,
        False,
    ]
    assert len(str(summary_before))
    assert len(summary_before._repr_html_())
    run(defaultclock.dt)
    summary_after = scheduling_summary(net)
    assert str(summary_after) == str(summary_before)
    assert summary_after._repr_html_() == summary_before._repr_html_()


class Preparer(BrianObject):
    add_to_magic_network = True

    def __init__(self, **kwds):
        super().__init__(**kwds)
        self.did_reinit = False
        self.did_pre_run = False
        self.did_post_run = False

    def reinit(self, level=0):
        self.did_reinit = True

    def before_run(self, namespace=None, level=0):
        self.did_pre_run = True

    def after_run(self):
        self.did_post_run = True


@pytest.mark.codegen_independent
def test_magic_network():
    # test that magic network functions correctly
    x = Counter()
    y = Counter()
    run(10 * ms)
    assert_equal(x.count, 100)
    assert_equal(y.count, 100)

    assert len(repr(magic_network))  # very basic test...
    assert len(str(magic_network))  # very basic test...


class Stopper(BrianObject):
    add_to_magic_network = True

    def __init__(self, stoptime, stopfunc, **kwds):
        super().__init__(**kwds)
        self.stoptime = stoptime
        self.stopfunc = stopfunc

    def run(self):
        self.stoptime -= 1
        if self.stoptime <= 0:
            self.stopfunc()


@pytest.mark.codegen_independent
def test_network_stop():
    # test that Network.stop and global stop() work correctly
    net = Network()
    x = Stopper(10, net.stop)
    net.add(x)
    net.run(10 * ms)
    assert_equal(defaultclock.t, 1 * ms)

    x = Stopper(10, stop)
    net = Network(x)
    net.run(10 * ms)
    assert_equal(defaultclock.t, 1 * ms)


@pytest.mark.codegen_independent
def test_network_operations():
    # test NetworkOperation and network_operation
    seq = []

    def f1():
        seq.append("a")

    op1 = NetworkOperation(f1, when="start", order=1)

    @network_operation
    def f2():
        seq.append("b")

    @network_operation(when="end", order=1)
    def f3():
        seq.append("c")

    # In complex frameworks, network operations might be object methods that
    # access some common data
    class Container:
        def __init__(self):
            self.g1_data = "B"
            self.g2_data = "C"

        def g1(self):
            seq.append(self.g1_data)

        def g2(self):
            seq.append(self.g2_data)

    c = Container()
    c_op1 = NetworkOperation(c.g1)
    c_op2 = NetworkOperation(c.g2, when="end", order=1)
    net = Network(op1, f2, f3, c_op1, c_op2)
    net.run(1 * ms)

    assert_equal("".join(seq), "bBacC" * 10)


@pytest.mark.codegen_independent
def test_incorrect_network_operations():
    # Network operations with more than one argument are not allowed
    def func(x, y):
        pass

    class Container:
        def func(self, x, y):
            pass

    c = Container()

    with pytest.raises(TypeError):
        NetworkOperation(func)
    with pytest.raises(TypeError):
        NetworkOperation(c.func)

    # Incorrect use of @network_operation -- it does not work on an instance
    # method
    try:

        class Container:
            @network_operation
            def func(self):
                pass

        raise AssertionError("expected a TypeError")
    except TypeError:
        pass  # this is what we expected


@pytest.mark.codegen_independent
def test_network_operations_name():
    # test NetworkOperation name input
    seq = []

    def f1():
        seq.append("a")

    def f2():
        seq.append("b")

    def x():
        pass

    op = NetworkOperation(lambda: x)
    assert_equal(op.name, "networkoperation")

    op0 = NetworkOperation(lambda: x, name="named_network")
    assert_equal(op0.name, "named_network")

    op1 = NetworkOperation(f1, name="networkoperation_1")
    op2 = NetworkOperation(f1, name="networkoperation_3")
    op3 = NetworkOperation(f2, name="networkoperation_2")

    net = Network(op1, op2, op3)
    net.run(1 * ms)

    assert_equal("".join(seq), "aba" * 10)


@pytest.mark.codegen_independent
def test_network_active_flag():
    # test that the BrianObject.active flag is recognised by Network.run
    x = Counter()
    y = Counter()
    y.active = False
    run(1 * ms)
    assert_equal(x.count, 10)
    assert_equal(y.count, 0)


@pytest.mark.standalone_compatible
@pytest.mark.multiple_runs
def test_spikes_after_deactivating():
    # Make sure that a spike in the last time step gets cleared. See #1319
    always_spike = NeuronGroup(1, "", threshold="True", reset="")
    spike_mon = SpikeMonitor(always_spike)
    run(defaultclock.dt)
    always_spike.active = False
    run(defaultclock.dt)
    device.build(direct_call=False, **device.build_options)
    assert_equal(spike_mon.t[:], [0] * second)


@pytest.mark.codegen_independent
def test_network_t():
    # test that Network.t works as expected
    x = Counter(dt=1 * ms)
    y = Counter(dt=2 * ms)
    net = Network(x, y)
    net.run(4 * ms)
    assert_equal(net.t, 4 * ms)
    net.run(1 * ms)
    assert_equal(net.t, 5 * ms)
    assert_equal(x.count, 5)
    assert_equal(y.count, 3)
    net.run(0.5 * ms)  # should only update x
    assert_equal(net.t, 5.5 * ms)
    assert_equal(x.count, 6)
    assert_equal(y.count, 3)
    net.run(0.5 * ms)  # shouldn't do anything
    assert_equal(net.t, 6 * ms)
    assert_equal(x.count, 6)
    assert_equal(y.count, 3)
    net.run(0.5 * ms)  # should update x and y
    assert_equal(net.t, 6.5 * ms)
    assert_equal(x.count, 7)
    assert_equal(y.count, 4)

    del x, y, net

    # now test with magic run
    x = Counter(dt=1 * ms)
    y = Counter(dt=2 * ms)
    run(4 * ms)
    assert_equal(magic_network.t, 4 * ms)
    assert_equal(x.count, 4)
    assert_equal(y.count, 2)
    run(4 * ms)
    assert_equal(magic_network.t, 8 * ms)
    assert_equal(x.count, 8)
    assert_equal(y.count, 4)
    run(1 * ms)
    assert_equal(magic_network.t, 9 * ms)
    assert_equal(x.count, 9)
    assert_equal(y.count, 5)


@pytest.mark.codegen_independent
def test_incorrect_dt_defaultclock():
    defaultclock.dt = 0.5 * ms
    G = NeuronGroup(1, "dv/dt = -v / (10*ms) : 1")
    net = Network(G)
    net.run(0.5 * ms)
    defaultclock.dt = 1 * ms
    with pytest.raises(ValueError):
        net.run(0 * ms)


@pytest.mark.codegen_independent
def test_incorrect_dt_custom_clock():
    clock = Clock(dt=0.5 * ms)
    G = NeuronGroup(1, "dv/dt = -v / (10*ms) : 1", clock=clock)
    net = Network(G)
    net.run(0.5 * ms)
    clock.dt = 1 * ms
    with pytest.raises(ValueError):
        net.run(0 * ms)


@pytest.mark.codegen_independent
def test_network_remove():
    x = Counter()
    y = Counter()
    net = Network(x, y)
    net.remove(y)
    net.run(1 * ms)
    assert_equal(x.count, 10)
    assert_equal(y.count, 0)
    # the relevance of this test is when we use weakref.proxy objects in
    # Network.objects, we should be able to add and remove these from
    # the Network just as much as the original objects
    # TODO: Does this test make sense now that Network does not store weak
    #       references by default?
    for obj in copy.copy(net.objects):
        net.remove(obj)
    net.run(1 * ms)
    assert_equal(x.count, 10)
    assert_equal(y.count, 0)


@pytest.mark.codegen_independent
def test_contained_objects():
    obj = CounterWithContained()
    net = Network(obj)
    # The contained object should not be stored explicitly
    assert len(net.objects) == 1
    # It should be accessible via the network interface, though
    assert len(net) == 2
    net.run(defaultclock.dt)

    # The contained object should be executed during the run
    assert obj.count == 1
    assert obj.sub_counter.count == 1

    # contained objects should be accessible via get_states/set_states
    states = net.get_states()
    assert len(states) == 2
    assert set(states.keys()) == {obj.name, obj.sub_counter.name}
    assert set(states[obj.name].keys()) == {"state"}
    assert set(states[obj.sub_counter.name].keys()) == {"state"}
    net[obj.name].set_states({"state": 1})
    net[obj.sub_counter.name].set_states({"state": 2})

    net.remove(obj)
    assert len(net.objects) == 0
    assert len(net) == 0
    assert len(net.get_states()) == 0
    net.run(defaultclock.dt)
    assert obj.count == 1
    assert obj.sub_counter.count == 1


class NoninvalidatingCounter(Counter):
    add_to_magic_network = True
    invalidates_magic_network = False


@pytest.mark.codegen_independent
def test_invalid_magic_network():
    x = Counter()
    run(1 * ms)
    assert_equal(x.count, 10)
    y = Counter()
    try:
        run(1 * ms)
        raise AssertionError("Expected a MagicError")
    except MagicError:
        pass  # this is expected
    del x, y
    x = Counter()
    run(1 * ms)
    y = NoninvalidatingCounter()
    run(1 * ms)
    assert_equal(x.count, 20)
    assert_equal(y.count, 10)
    del y
    run(1 * ms)
    assert_equal(x.count, 30)
    del x
    x = Counter()
    run(1 * ms)
    assert_equal(magic_network.t, 1 * ms)
    del x
    x = Counter()
    y = Counter()
    run(1 * ms)
    assert_equal(x.count, 10)
    assert_equal(y.count, 10)


@pytest.mark.codegen_independent
def test_multiple_networks_invalid():
    x = Counter()
    net = Network(x)
    net.run(1 * ms)
    try:
        run(1 * ms)
        raise AssertionError("Expected a RuntimeError")
    except RuntimeError:
        pass  # this is expected

    try:
        net2 = Network(x)
        raise AssertionError("Expected a RuntimeError")
    except RuntimeError:
        pass  # this is expected


@pytest.mark.codegen_independent
def test_magic_weak_reference():
    """
    Test that holding a weak reference to an object does not make it get
    simulated."""

    G1 = NeuronGroup(1, "v:1")

    # this object should not be included
    G2 = weakref.ref(NeuronGroup(1, "v:1"))

    with catch_logs(log_level=logging.DEBUG) as l:
        run(1 * ms)
        # Check the debug messages for the number of included objects
        magic_objects = [
            msg[2] for msg in l if msg[1] == "brian2.core.magic.magic_objects"
        ][0]
        assert "2 objects" in magic_objects, f"Unexpected log message: {magic_objects}"


@pytest.mark.codegen_independent
def test_magic_unused_object():
    """Test that creating unused objects does not affect the magic system."""

    def create_group():
        # Produce two objects but return only one
        G1 = NeuronGroup(1, "v:1")  # no Thresholder or Resetter
        G2 = NeuronGroup(1, "v:1")  # This object should be garbage collected
        return G1

    G = create_group()
    with catch_logs(log_level=logging.DEBUG) as l:
        run(1 * ms)

        # Check the debug messages for the number of included objects
        magic_objects = [
            msg[2] for msg in l if msg[1] == "brian2.core.magic.magic_objects"
        ][0]
        assert "2 objects" in magic_objects, f"Unexpected log message: {magic_objects}"


@pytest.mark.codegen_independent
def test_network_access():
    x = Counter(name="counter")
    net = Network(x)
    assert len(net) == 1
    assert len(repr(net))  # very basic test...
    assert len(str(net))  # very basic test...

    # accessing objects
    assert net["counter"] is x
    with pytest.raises(TypeError):
        net[123]
    with pytest.raises(TypeError):
        net[1:3]
    with pytest.raises(KeyError):
        net["non-existing"]

    objects = [obj for obj in net]
    assert set(objects) == set(net.objects)

    # deleting objects
    del net["counter"]
    with pytest.raises(TypeError):
        net.__delitem__(123)
    with pytest.raises(TypeError):
        net.__delitem__(slice(1, 3))
    with pytest.raises(KeyError):
        net.__delitem__("counter")


@pytest.mark.codegen_independent
def test_dependency_check():
    def create_net():
        G = NeuronGroup(10, "v: 1", threshold="False")
        dependent_objects = [
            StateMonitor(G, "v", record=True),
            SpikeMonitor(G),
            PopulationRateMonitor(G),
            Synapses(G, G, on_pre="v+=1"),
        ]
        return dependent_objects

    dependent_objects = create_net()
    # Trying to simulate the monitors/synapses without the group should fail
    for obj in dependent_objects:
        with pytest.raises(ValueError):
            Network(obj).run(0 * ms)

    # simulation with a magic network should work when we have an explicit
    # reference to one of the objects, but the object should be inactive and
    # we should get a warning
    assert all(obj.active for obj in dependent_objects)
    for obj in dependent_objects:  # obj is our explicit reference
        with catch_logs() as l:
            run(0 * ms)
            dependency_warnings = [
                msg[2] for msg in l if msg[1] == "brian2.core.magic.dependency_warning"
            ]
            assert len(dependency_warnings) == 1
        assert not obj.active


def test_loop():
    """
    Somewhat realistic test with a loop of magic networks
    """

    def run_simulation():
        G = NeuronGroup(10, "dv/dt = -v / (10*ms) : 1", reset="v=0", threshold="v>1")
        G.v = np.linspace(0, 1, 10)
        run(1 * ms)
        # We return potentially problematic references to a VariableView
        return G.v

    # First run
    with catch_logs(log_level=logging.DEBUG) as l:
        v = run_simulation()
        assert v[0] == 0 and 0 < v[-1] < 1
        # Check the debug messages for the number of included objects
        magic_objects = [
            msg[2] for msg in l if msg[1] == "brian2.core.magic.magic_objects"
        ][0]
        assert "4 objects" in magic_objects

    # Second run
    with catch_logs(log_level=logging.DEBUG) as l:
        v = run_simulation()
        assert v[0] == 0 and 0 < v[-1] < 1
        # Check the debug messages for the number of included objects
        magic_objects = [
            msg[2] for msg in l if msg[1] == "brian2.core.magic.magic_objects"
        ][0]
        assert "4 objects" in magic_objects


@pytest.mark.codegen_independent
def test_magic_collect():
    """
    Make sure all expected objects are collected in a magic network
    """
    P = PoissonGroup(10, rates=100 * Hz)
    G = NeuronGroup(10, "v:1", threshold="False")
    S = Synapses(G, G, "")

    state_mon = StateMonitor(G, "v", record=True)
    spike_mon = SpikeMonitor(G)
    rate_mon = PopulationRateMonitor(G)

    objects = collect()

    assert len(objects) == 6, f"expected {int(6)} objects, got {len(objects)}"


import sys
from contextlib import contextmanager
from io import BytesIO, StringIO


@contextmanager
def captured_output():
    new_out, new_err = StringIO(), StringIO()
    old_out, old_err = sys.stdout, sys.stderr
    try:
        sys.stdout, sys.stderr = new_out, new_err
        yield sys.stdout, sys.stderr
    finally:
        sys.stdout, sys.stderr = old_out, old_err


@pytest.mark.codegen_independent
def test_progress_report():
    """
    Very basic test of progress reporting
    """
    G = NeuronGroup(1, "")
    net = Network(G)

    # No output
    with captured_output() as (out, err):
        net.run(1 * ms, report=None)
    # There should be at least two lines of output
    out, err = out.getvalue(), err.getvalue()
    assert len(out) == 0 and len(err) == 0

    with captured_output() as (out, err):
        net.run(1 * ms)
    # There should be at least two lines of output
    out, err = out.getvalue(), err.getvalue()
    assert len(out) == 0 and len(err) == 0

    # Progress should go to stdout
    with captured_output() as (out, err):
        net.run(1 * ms, report="text")
    # There should be at least two lines of output
    out, err = out.getvalue(), err.getvalue()
    assert len(out.split("\n")) >= 2 and len(err) == 0

    with captured_output() as (out, err):
        net.run(1 * ms, report="stdout")
    # There should be at least two lines of output
    out, err = out.getvalue(), err.getvalue()
    assert len(out.split("\n")) >= 2 and len(err) == 0

    # Progress should go to stderr
    with captured_output() as (out, err):
        net.run(1 * ms, report="stderr")
    # There should be at least two lines of output
    out, err = out.getvalue(), err.getvalue()
    assert len(err.split("\n")) >= 2 and len(out) == 0

    # Custom function
    calls = []

    def capture_progress(elapsed, complete, start, duration):
        calls.append((elapsed, complete, start, duration))

    with captured_output() as (out, err):
        net.run(1 * ms, report=capture_progress)
    out, err = out.getvalue(), err.getvalue()

    assert len(err) == 0 and len(out) == 0
    # There should be at least a call for the start and the end
    assert len(calls) >= 2 and calls[0][1] == 0.0 and calls[-1][1] == 1.0


@pytest.mark.codegen_independent
def test_progress_report_incorrect():
    """
    Test wrong use of the report option
    """
    G = NeuronGroup(1, "")
    net = Network(G)
    with pytest.raises(ValueError):
        net.run(1 * ms, report="unknown")
    with pytest.raises(TypeError):
        net.run(1 * ms, report=object())


@pytest.mark.standalone_compatible
@pytest.mark.multiple_runs
def test_multiple_runs_report_standalone():
    group = NeuronGroup(1, "dv/dt = 1*Hz : 1")
    run(1 * ms, report="text")
    run(1 * ms)
    device.build(direct_call=False, **device.build_options)


@pytest.mark.standalone_compatible
@pytest.mark.multiple_runs
def test_multiple_runs_report_standalone_2():
    group = NeuronGroup(1, "dv/dt = 1*Hz : 1")
    run(1 * ms)
    run(1 * ms, report="text")
    device.build(direct_call=False, **device.build_options)


@pytest.mark.standalone_compatible
@pytest.mark.multiple_runs
def test_multiple_runs_report_standalone_3():
    group = NeuronGroup(1, "dv/dt = 1*Hz : 1")
    run(1 * ms, report="text")
    run(1 * ms, report="text")
    device.build(direct_call=False, **device.build_options)


# This tests a specific limitation of the C++ standalone mode (cannot mix
# multiple report methods)
@pytest.mark.cpp_standalone
@pytest.mark.standalone_only
def test_multiple_runs_report_standalone_incorrect():
    set_device("cpp_standalone", build_on_run=False)
    group = NeuronGroup(1, "dv/dt = 1*Hz : 1")
    run(1 * ms, report="text")
    with pytest.raises(NotImplementedError):
        run(1 * ms, report="stderr")


@pytest.mark.codegen_independent
def test_store_restore():
    source = NeuronGroup(
        10,
        """dv/dt = rates : 1
                                rates : Hz""",
        threshold="v>1",
        reset="v=0",
    )
    source.rates = "i*100*Hz"
    target = NeuronGroup(10, "v:1")
    synapses = Synapses(source, target, model="w:1", on_pre="v+=w")
    synapses.connect(j="i")
    synapses.w = "i*1.0"
    synapses.delay = "i*ms"
    state_mon = StateMonitor(target, "v", record=True)
    spike_mon = SpikeMonitor(source)
    net = Network(source, target, synapses, state_mon, spike_mon)
    net.store()  # default time slot
    net.run(10 * ms)
    net.store("second")
    net.run(10 * ms)
    v_values = state_mon.v[:, :]
    spike_indices, spike_times = spike_mon.it_
    net.restore()  # Go back to beginning
    assert defaultclock.t == 0 * ms
    assert net.t == 0 * ms
    net.run(20 * ms)
    assert_equal(v_values, state_mon.v[:, :])
    assert_equal(spike_indices, spike_mon.i[:])
    assert_equal(spike_times, spike_mon.t_[:])

    # Go back to middle
    net.restore("second")
    assert defaultclock.t == 10 * ms
    assert net.t == 10 * ms
    net.run(10 * ms)
    assert_equal(v_values, state_mon.v[:, :])
    assert_equal(spike_indices, spike_mon.i[:])
    assert_equal(spike_times, spike_mon.t_[:])

    # Go back again (see github issue #681)
    net.restore("second")
    assert defaultclock.t == 10 * ms
    assert net.t == 10 * ms


@pytest.mark.codegen_independent
def test_store_restore_to_file():
    filename = tempfile.mktemp(suffix="state", prefix="brian_test")
    source = NeuronGroup(
        10,
        """
        dv/dt = rates : 1
        rates : Hz
        """,
        threshold="v>1",
        reset="v=0",
    )
    source.rates = "i*100*Hz"
    target = NeuronGroup(10, "v:1")
    synapses = Synapses(source, target, model="w:1", on_pre="v+=w")
    synapses.connect(j="i")
    synapses.w = "i*1.0"
    synapses.delay = "i*ms"
    state_mon = StateMonitor(target, "v", record=True)
    spike_mon = SpikeMonitor(source)
    net = Network(source, target, synapses, state_mon, spike_mon)
    net.store(filename=filename)  # default time slot
    net.run(10 * ms)
    net.store("second", filename=filename)
    net.run(10 * ms)
    v_values = state_mon.v[:, :]
    spike_indices, spike_times = spike_mon.it_

    net.restore(filename=filename)  # Go back to beginning
    assert defaultclock.t == 0 * ms
    assert net.t == 0 * ms
    net.run(20 * ms)
    assert_equal(v_values, state_mon.v[:, :])
    assert_equal(spike_indices, spike_mon.i[:])
    assert_equal(spike_times, spike_mon.t_[:])

    # Go back to middle
    net.restore("second", filename=filename)
    assert defaultclock.t == 10 * ms
    assert net.t == 10 * ms
    net.run(10 * ms)
    assert_equal(v_values, state_mon.v[:, :])
    assert_equal(spike_indices, spike_mon.i[:])
    assert_equal(spike_times, spike_mon.t_[:])
    try:
        os.remove(filename)
    except OSError:
        pass


@pytest.mark.codegen_independent
def test_store_restore_to_file_new_objects():
    # A more realistic test where the objects are completely re-created
    filename = tempfile.mktemp(suffix="state", prefix="brian_test")

    def create_net():
        # Use a bit of a complicated spike and connection pattern with
        # heterogeneous delays

        # Note: it is important that all objects have the same name, this would
        # be the case if we were running this in a new process but to not rely
        # on garbage collection we will assign explicit names here
        source = SpikeGeneratorGroup(
            5,
            np.arange(5).repeat(3),
            [3, 4, 1, 2, 3, 7, 5, 4, 1, 0, 5, 9, 7, 8, 9] * ms,
            name="source",
        )
        target = NeuronGroup(10, "v:1", name="target")
        synapses = Synapses(source, target, model="w:1", on_pre="v+=w", name="synapses")
        synapses.connect("j>=i")
        synapses.w = "i*1.0 + j*2.0"
        synapses.delay = "(5-i)*ms"
        state_mon = StateMonitor(target, "v", record=True, name="statemonitor")
        input_spikes = SpikeMonitor(source, name="input_spikes")
        net = Network(source, target, synapses, state_mon, input_spikes)
        return net

    net = create_net()
    net.store(filename=filename)  # default time slot
    net.run(5 * ms)
    net.store("second", filename=filename)
    net.run(5 * ms)
    input_spike_indices = np.array(net["input_spikes"].i)
    input_spike_times = Quantity(net["input_spikes"].t, copy=True)
    v_values_full_sim = Quantity(net["statemonitor"].v[:, :], copy=True)

    net = create_net()
    net.restore(filename=filename)  # Go back to beginning
    net.run(10 * ms)
    assert_equal(input_spike_indices, net["input_spikes"].i)
    assert_equal(input_spike_times, net["input_spikes"].t)
    assert_equal(v_values_full_sim, net["statemonitor"].v[:, :])

    net = create_net()
    net.restore("second", filename=filename)  # Go back to middle
    net.run(5 * ms)
    assert_equal(input_spike_indices, net["input_spikes"].i)
    assert_equal(input_spike_times, net["input_spikes"].t)
    assert_equal(v_values_full_sim, net["statemonitor"].v[:, :])

    try:
        os.remove(filename)
    except OSError:
        pass


@pytest.mark.codegen_independent
def test_store_restore_to_file_differing_nets():
    # Check that the store/restore mechanism is not used with differing
    # networks
    filename = tempfile.mktemp(suffix="state", prefix="brian_test")

    source = SpikeGeneratorGroup(
        5, [0, 1, 2, 3, 4], [0, 1, 2, 3, 4] * ms, name="source_1"
    )
    mon = SpikeMonitor(source, name="monitor")
    net = Network(source, mon)
    net.store(filename=filename)

    source_2 = SpikeGeneratorGroup(
        5, [0, 1, 2, 3, 4], [0, 1, 2, 3, 4] * ms, name="source_2"
    )
    mon = SpikeMonitor(source_2, name="monitor")
    net = Network(source_2, mon)
    with pytest.raises(KeyError):
        net.restore(filename=filename)

    net = Network(source)  # Without the monitor
    with pytest.raises(KeyError):
        net.restore(filename=filename)


@pytest.mark.codegen_independent
def test_store_restore_magic():
    source = NeuronGroup(
        10,
        """
        dv/dt = rates : 1
        rates : Hz
        """,
        threshold="v>1",
        reset="v=0",
    )
    source.rates = "i*100*Hz"
    target = NeuronGroup(10, "v:1")
    synapses = Synapses(source, target, model="w:1", on_pre="v+=w")
    synapses.connect(j="i")
    synapses.w = "i*1.0"
    synapses.delay = "i*ms"
    state_mon = StateMonitor(target, "v", record=True)
    spike_mon = SpikeMonitor(source)
    store()  # default time slot
    run(10 * ms)
    store("second")
    run(10 * ms)
    v_values = state_mon.v[:, :]
    spike_indices, spike_times = spike_mon.it_

    restore()  # Go back to beginning
    assert magic_network.t == 0 * ms
    run(20 * ms)
    assert defaultclock.t == 20 * ms
    assert_equal(v_values, state_mon.v[:, :])
    assert_equal(spike_indices, spike_mon.i[:])
    assert_equal(spike_times, spike_mon.t_[:])

    # Go back to middle
    restore("second")
    assert magic_network.t == 10 * ms
    run(10 * ms)
    assert defaultclock.t == 20 * ms
    assert_equal(v_values, state_mon.v[:, :])
    assert_equal(spike_indices, spike_mon.i[:])
    assert_equal(spike_times, spike_mon.t_[:])


@pytest.mark.codegen_independent
def test_store_restore_magic_to_file():
    filename = tempfile.mktemp(suffix="state", prefix="brian_test")
    source = NeuronGroup(
        10,
        """
        dv/dt = rates : 1
        rates : Hz
        """,
        threshold="v>1",
        reset="v=0",
    )
    source.rates = "i*100*Hz"
    target = NeuronGroup(10, "v:1")
    synapses = Synapses(source, target, model="w:1", on_pre="v+=w")
    synapses.connect(j="i")
    synapses.w = "i*1.0"
    synapses.delay = "i*ms"
    state_mon = StateMonitor(target, "v", record=True)
    spike_mon = SpikeMonitor(source)
    store(filename=filename)  # default time slot
    run(10 * ms)
    store("second", filename=filename)
    run(10 * ms)
    v_values = state_mon.v[:, :]
    spike_indices, spike_times = spike_mon.it_

    restore(filename=filename)  # Go back to beginning
    assert magic_network.t == 0 * ms
    run(20 * ms)
    assert defaultclock.t == 20 * ms
    assert_equal(v_values, state_mon.v[:, :])
    assert_equal(spike_indices, spike_mon.i[:])
    assert_equal(spike_times, spike_mon.t_[:])

    # Go back to middle
    restore("second", filename=filename)
    assert magic_network.t == 10 * ms
    run(10 * ms)
    assert defaultclock.t == 20 * ms
    assert_equal(v_values, state_mon.v[:, :])
    assert_equal(spike_indices, spike_mon.i[:])
    assert_equal(spike_times, spike_mon.t_[:])
    try:
        os.remove(filename)
    except OSError:
        pass


@pytest.mark.codegen_independent
def test_store_restore_spikequeue():
    # See github issue #938
    source = SpikeGeneratorGroup(1, [0], [0] * ms)
    target = NeuronGroup(1, "v : 1")
    conn = Synapses(source, target, on_pre="v += 1", delay=2 * defaultclock.dt)
    conn.connect()
    run(defaultclock.dt)  # Spike is not yet delivered
    store()
    run(2 * defaultclock.dt)
    assert target.v[0] == 1
    restore()
    run(2 * defaultclock.dt)
    assert target.v[0] == 1
    restore()
    run(2 * defaultclock.dt)
    assert target.v[0] == 1


@pytest.mark.skipif(
    not isinstance(get_device(), RuntimeDevice),
    reason="Getting/setting random number state only supported for runtime device.",
)
def test_restore_with_random_state():
    group = NeuronGroup(10, "dv/dt = -v/(10*ms) + (10*ms)**-0.5*xi : 1", method="euler")
    group.v = "rand()"
    mon = StateMonitor(group, "v", record=True)
    store()
    run(10 * ms)
    old_v = np.array(group.v)

    restore()  # Random state is not restored
    run(10 * ms)
    assert np.var(old_v - group.v) > 0  # very basic test for difference

    restore(restore_random_state=True)  # Random state is restored
    run(10 * ms)
    assert_equal(old_v, group.v)


@pytest.mark.codegen_independent
def test_store_restore_restore_synapses():
    group = NeuronGroup(10, "x : 1", threshold="False", reset="", name="group")
    synapses = Synapses(group, group, on_pre="x += 1", name="synapses")
    synapses.connect(i=[1, 3, 5], j=[6, 4, 2])
    net = Network(group, synapses)

    tmp_file = tempfile.mktemp()
    net.store(filename=tmp_file)

    # clear up
    del net
    del synapses
    del group

    # Recreate the network without connecting the synapses
    group = NeuronGroup(10, "x: 1", threshold="False", reset="", name="group")
    synapses = Synapses(group, group, "", on_pre="x += 1", name="synapses")
    net = Network(group, synapses)
    try:
        net.restore(filename=tmp_file)

        assert len(synapses) == 3
        assert_array_equal(synapses.i, [1, 3, 5])
        assert_array_equal(synapses.j, [6, 4, 2])

        # Tunning the network should not raise an error, despite the lack
        # of Synapses.connect
        net.run(0 * ms)
    finally:
        os.remove(tmp_file)


@pytest.mark.codegen_independent
def test_defaultclock_dt_changes():
    BrianLogger.suppress_name("resolution_conflict")
    for dt in [0.1 * ms, 0.01 * ms, 0.5 * ms, 1 * ms, 3.3 * ms]:
        defaultclock.dt = dt
        G = NeuronGroup(1, "v:1")
        mon = StateMonitor(G, "v", record=True)
        net = Network(G, mon)
        net.run(2 * dt)
        assert_equal(mon.t[:], [0, dt / ms] * ms)


@pytest.mark.standalone_compatible
@pytest.mark.multiple_runs
def test_dt_changes_between_runs():
    defaultclock.dt = 0.1 * ms
    G = NeuronGroup(1, "v:1")
    mon = StateMonitor(G, "v", record=True)
    run(0.5 * ms)
    defaultclock.dt = 0.5 * ms
    run(0.5 * ms)
    defaultclock.dt = 0.1 * ms
    run(0.5 * ms)
    device.build(direct_call=False, **device.build_options)
    assert len(mon.t[:]) == 5 + 1 + 5
    assert_allclose(
        mon.t[:], [0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 1.0, 1.1, 1.2, 1.3, 1.4] * ms
    )


@pytest.mark.codegen_independent
def test_dt_restore():
    defaultclock.dt = 0.5 * ms
    G = NeuronGroup(1, "dv/dt = -v/(10*ms) : 1")
    mon = StateMonitor(G, "v", record=True)
    net = Network(G, mon)
    net.store()

    net.run(1 * ms)
    assert_equal(mon.t[:], [0, 0.5] * ms)
    defaultclock.dt = 1 * ms
    net.run(2 * ms)
    assert_equal(mon.t[:], [0, 0.5, 1, 2] * ms)
    net.restore()
    assert_equal(mon.t[:], [])
    net.run(1 * ms)
    assert defaultclock.dt == 0.5 * ms
    assert_equal(mon.t[:], [0, 0.5] * ms)


@pytest.mark.codegen_independent
def test_continuation():
    defaultclock.dt = 1 * ms
    G = NeuronGroup(1, "dv/dt = -v / (10*ms) : 1")
    G.v = 1
    mon = StateMonitor(G, "v", record=True)
    net = Network(G, mon)
    net.run(2 * ms)

    # Run the same simulation but with two runs that use sub-dt run times
    G2 = NeuronGroup(1, "dv/dt = -v / (10*ms) : 1")
    G2.v = 1
    mon2 = StateMonitor(G2, "v", record=True)
    net2 = Network(G2, mon2)
    net2.run(0.5 * ms)
    net2.run(1.5 * ms)

    assert_equal(mon.t[:], mon2.t[:])
    assert_equal(mon.v[:], mon2.v[:])


@pytest.mark.codegen_independent
def test_get_set_states():
    G = NeuronGroup(10, "v:1", name="a_neurongroup")
    G.v = "i"
    net = Network(G)
    states1 = net.get_states()
    states2 = magic_network.get_states()
    states3 = net.get_states(read_only_variables=False)
    assert (
        set(states1.keys())
        == set(states2.keys())
        == set(states3.keys())
        == {"a_neurongroup"}
    )
    assert (
        set(states1["a_neurongroup"].keys())
        == set(states2["a_neurongroup"].keys())
        == {"i", "dt", "N", "t", "v", "t_in_timesteps"}
    )
    assert set(states3["a_neurongroup"]) == {"v"}

    # Try re-setting the state
    G.v = 0
    net.set_states(states3)
    assert_equal(G.v, np.arange(10))


@pytest.mark.codegen_independent
def test_multiple_runs_defaultclock():
    defaultclock.dt = 0.1 * ms
    G = NeuronGroup(1, "dv/dt = -v / (10*ms) : 1")
    net = Network(G)
    net.run(0.5 * ms)

    # The new dt is not compatible with the previous time but it should not
    # raise an error because we start a new simulation at time 0
    defaultclock.dt = 1 * ms
    G = NeuronGroup(1, "dv/dt = -v / (10*ms) : 1")
    net = Network(G)
    net.run(1 * ms)


@pytest.mark.codegen_independent
def test_multiple_runs_defaultclock_incorrect():
    defaultclock.dt = 0.1 * ms
    G = NeuronGroup(1, "dv/dt = -v / (10*ms) : 1")
    net = Network(G)
    net.run(0.5 * ms)

    # The new dt is not compatible with the previous time since we cannot
    # continue at 0.5ms with a dt of 1ms
    defaultclock.dt = 1 * ms
    with pytest.raises(ValueError):
        net.run(1 * ms)


@pytest.mark.standalone_compatible
def test_profile():
    G = NeuronGroup(
        10,
        "dv/dt = -v / (10*ms) : 1",
        threshold="v>1",
        reset="v=0",
        name="profile_test",
    )
    G.v = 1.1
    net = Network(G)
    net.run(1 * ms, profile=True)
    # The should be four simulated CodeObjects, one for the group and one each
    # for state update, threshold and reset + 1 for the clock
    info = net.profiling_info
    info_dict = dict(info)
    # Standalone does not include the NeuronGroup object (which is not doing
    # anything during the run) in the profiling information, while runtime
    # does
    assert 3 <= len(info) <= 4
    assert len(info) == 3 or "profile_test" in info_dict
    for obj in ["stateupdater", "spike_thresholder", "spike_resetter"]:
        name = f"profile_test_{obj}"
        assert name in info_dict or f"{name}_codeobject" in info_dict
    assert all([t >= 0 * second for _, t in info])


@pytest.mark.standalone_compatible
def test_profile_off():
    G = NeuronGroup(
        10,
        "dv/dt = -v / (10*ms) : 1",
        threshold="v>1",
        reset="v=0",
        name="profile_test",
    )
    net = Network(G)
    net.run(1 * ms, profile=False)
    with pytest.raises(ValueError):
        profiling_summary(net)


@pytest.mark.codegen_independent
def test_profile_ipython_html():
    G = NeuronGroup(
        10,
        "dv/dt = -v / (10*ms) : 1",
        threshold="v>1",
        reset="v=0",
        name="profile_test",
    )
    G.v = 1.1
    net = Network(G)
    net.run(1 * ms, profile=True)
    summary = profiling_summary(net)
    assert len(summary._repr_html_())


@pytest.mark.codegen_independent
def test_magic_scope():
    """
    Check that `start_scope` works as expected.
    """
    G1 = NeuronGroup(1, "v:1", name="G1")
    G2 = NeuronGroup(1, "v:1", name="G2")
    objs1 = {obj.name for obj in collect()}
    start_scope()
    G3 = NeuronGroup(1, "v:1", name="G3")
    G4 = NeuronGroup(1, "v:1", name="G4")
    objs2 = {obj.name for obj in collect()}
    assert objs1 == {"G1", "G2"}
    assert objs2 == {"G3", "G4"}


@pytest.mark.standalone_compatible
def test_runtime_rounding():
    # Test that runtime and standalone round in the same way, see github issue
    # #695 for details
    defaultclock.dt = 20.000000000020002 * us
    G = NeuronGroup(1, "v:1")
    mon = StateMonitor(G, "v", record=True)
    run(defaultclock.dt * 250)
    assert len(mon.t) == 250


@pytest.mark.codegen_independent
def test_small_runs():
    # One long run and multiple small runs should give the same results
    group_1 = NeuronGroup(10, "dv/dt = -v / (10*ms) : 1")
    group_1.v = "(i + 1) / N"
    mon_1 = StateMonitor(group_1, "v", record=True)
    net_1 = Network(group_1, mon_1)
    net_1.run(1 * second)

    group_2 = NeuronGroup(10, "dv/dt = -v / (10*ms) : 1")
    group_2.v = "(i + 1) / N"
    mon_2 = StateMonitor(group_2, "v", record=True)
    net_2 = Network(group_2, mon_2)
    runtime = 1 * ms
    while True:
        runtime *= 3
        runtime = min([runtime, 1 * second - net_2.t])
        net_2.run(runtime)
        if net_2.t >= 1 * second:
            break

    assert_allclose(mon_1.t_[:], mon_2.t_[:])
    assert_allclose(mon_1.v_[:], mon_2.v_[:])


@pytest.mark.codegen_independent
def test_both_equal():
    # check all objects added by Network.add() also have their contained_objects added to 'Network'
    tau = 10 * ms
    diff_eqn = """dv/dt = (1-v)/tau : 1"""
    chg_code = """v = 2*v"""

    Ng = NeuronGroup(1, diff_eqn, method="exact")
    M1 = StateMonitor(Ng, "v", record=True)
    netObj = Network(Ng, M1)
    Ng.run_regularly(chg_code, dt=20 * ms)
    netObj.run(100 * ms)

    start_scope()
    Ng = NeuronGroup(1, diff_eqn, method="exact")
    M2 = StateMonitor(Ng, "v", record=True)
    Ng.run_regularly(chg_code, dt=20 * ms)
    run(100 * ms)

    assert (M1.v == M2.v).all()


@pytest.mark.standalone_compatible
@pytest.mark.multiple_runs
def test_long_run():
    defaultclock.dt = 0.1 * ms
    group = NeuronGroup(1, "x : 1")
    group.run_regularly("x += 1")
    # Timesteps are internally stored as 64bit integers, but previous versions
    # converted them into 32bit integers along the way. We'll make sure that
    # this is not the case and everything runs fine. To not actually run such a
    # long simulation we run a single huge time step
    start_step = 2**31 - 5
    defaultclock.dt = 0.1 * ms
    start_time = start_step * defaultclock.dt
    defaultclock.dt = start_time
    run(start_time)  # A single, *very* long time step
    defaultclock.dt = 0.1 * ms
    run(6 * defaultclock.dt)
    device.build(direct_call=False, **device.build_options)
    assert group.x == 7


@pytest.mark.codegen_independent
def test_long_run_dt_change():
    # Check that the dt check is not too restrictive, see issue #730 for details
    group = NeuronGroup(1, "")  # does nothing...
    defaultclock.dt = 0.1 * ms
    run(100 * second)
    # print profiling_summary()
    defaultclock.dt = 0.01 * ms
    run(1 * second)


@pytest.mark.standalone_compatible
@pytest.mark.multiple_runs
def test_multiple_runs_constant_change():
    const_v = 1
    group = NeuronGroup(1, "v = const_v : 1")
    mon = StateMonitor(group, "v", record=0)
    run(defaultclock.dt)
    const_v = 2
    run(defaultclock.dt)
    device.build(direct_call=False, **device.build_options)
    assert_equal(mon.v[0], [1, 2])


@pytest.mark.standalone_compatible
@pytest.mark.multiple_runs
def test_multiple_runs_function_change():
    inp = TimedArray([1, 2], dt=defaultclock.dt)
    group = NeuronGroup(1, "v = inp(t) : 1")
    mon = StateMonitor(group, "v", record=0)
    run(2 * defaultclock.dt)
    inp = TimedArray([0, 0, 3, 4], dt=defaultclock.dt)
    run(2 * defaultclock.dt)
    device.build(direct_call=False, **device.build_options)
    assert_equal(mon.v[0], [1, 2, 3, 4])


@pytest.mark.codegen_independent
def test_unused_object_warning():
    with catch_logs() as logs:
        # Create a NeuronGroup that is not used in the network
        NeuronGroup(1, "v:1", name="never_used")
        # Make sure that it gets garbage collected
        import gc

        gc.collect()
    assert len(logs) == 1
    assert logs[0][0] == "WARNING"
    assert logs[0][1].endswith("unused_brian_object")
    assert "never_used" in logs[0][2]


@pytest.mark.codegen_independent
def test_negative_duration_in_run():
    G = NeuronGroup(1, "v:1")
    with pytest.raises(ValueError):
        run(-1 * second)


@pytest.mark.codegen_independent
def test_negative_duration_in_net_run():
    G = NeuronGroup(1, "v:1")
    net = Network(G)
    with pytest.raises(ValueError):
        net.run(-1 * second)


if __name__ == "__main__":
    BrianLogger.log_level_warn()
    for t in [
        test_incorrect_network_use,
        test_network_contains,
        test_empty_network,
        test_network_single_object,
        test_network_two_objects,
        test_network_from_dict,
        test_network_different_clocks,
        test_network_different_when,
        test_network_default_schedule,
        test_network_schedule_change,
        test_network_before_after_schedule,
        test_network_custom_slots,
        test_network_incorrect_schedule,
        test_schedule_warning,
        test_scheduling_summary_magic,
        test_scheduling_summary,
        test_magic_network,
        test_network_stop,
        test_network_operations,
        test_incorrect_network_operations,
        test_network_operations_name,
        test_network_active_flag,
        test_network_t,
        test_incorrect_dt_defaultclock,
        test_incorrect_dt_custom_clock,
        test_network_remove,
        test_magic_weak_reference,
        test_magic_unused_object,
        test_invalid_magic_network,
        test_multiple_networks_invalid,
        test_network_access,
        test_loop,
        test_magic_collect,
        test_progress_report,
        test_progress_report_incorrect,
        test_multiple_runs_report_standalone,
        test_multiple_runs_report_standalone_2,
        test_multiple_runs_report_standalone_3,
        test_multiple_runs_report_standalone_incorrect,
        test_store_restore,
        test_store_restore_to_file,
        test_store_restore_to_file_new_objects,
        test_store_restore_to_file_differing_nets,
        test_store_restore_magic,
        test_store_restore_magic_to_file,
        test_store_restore_spikequeue,
        test_store_restore_restore_synapses,
        test_defaultclock_dt_changes,
        test_dt_changes_between_runs,
        test_dt_restore,
        test_continuation,
        test_get_set_states,
        test_multiple_runs_defaultclock,
        test_multiple_runs_defaultclock_incorrect,
        test_profile,
        test_profile_off,
        test_profile_ipython_html,
        test_magic_scope,
        test_runtime_rounding,
        test_small_runs,
        test_both_equal,
        test_long_run,
        test_long_run_dt_change,
        test_multiple_runs_constant_change,
        test_multiple_runs_function_change,
    ]:
        set_device(all_devices["runtime"])
        t()
        reinit_and_delete()