This allows for you to create converters that can have varying
behaviour using the converter's __init__ instead of having to do a
meta-class based approach to get around the fact that __init__ is part
of the interface.
To make up for the lack of __init__, a new method Converter.prepare was
added to do the work that __init__ used to do.
If a command or a group raised a CommandError derived exception it
would be wrapped into CommandInvokeError. This fixes this case so it
raises the original exception instead.
The way the command cooldown works is using a windowed way of doing it.
That is, if we have a cooldown of 2 commands every 30 seconds then if we
do a single command, we have 30 seconds to do the second command or else
we will get rate limited. This more or less matches the common
expectations on how cooldowns should be.
These cooldowns can be bucketed up to a single dimension of depth for
a per-user, per-guild, or per-channel basis. Of course, a global bucket
is also provided. These cannot be mixed, e.g. no per-channel per-user
cooldowns.
When a command cooldown is triggered, the error handlers will receive a
an exception of type CommandOnCooldown with proper information regarding
the cooldown such as retry_after and the bucket information itself.
This allows you to strictly require a number of arguments. The default
behaviour in this case is still `True`, since it would be a breaking
change otherwise and is a sane default. However if someone would want
to set this to `False`, they would receive an exception of type
`TooManyArguments` if too many arguments are passed to a command.
Hopefully this removes the uses of `ctx.message.content == 'stuff'`
inside commands.
Global checks are checks that are executed before regular per-command
checks except done to every command that the bot has registered. This
allows you to have checks that apply to every command without having to
override `on_message` or appending the check to every single command.
This change should make it a bit more intuitive to get the original
exception without having the gotcha of checking ``isinstance`` inside
the error handler.
This happened due to not resetting the `invoked_subcommand` state
tracking. Since the `invoked_subcommand` was not reset, it would always
assume that it was valid and repeatedly call it when passed invalid
subcommands/arguments.
Provide fallback on_command_error - will only fire if no cog handlers and
no local handler.
Propagate exceptions in checks and argument parsing to bot.
Code is a little more straightforward. The Member and Channel special
cases will now work in private message contexts when given a name. The
Member special case will also work if you pass in a name and
discriminator combo.
This led to decorating a lot of things into @asyncio.coroutine.
Unfortunately there's no way to lower the amount of decoration since
coroutines spread like a virus.
