mirror of
https://github.com/Cog-Creators/Red-DiscordBot.git
synced 2025-11-06 03:08:55 -05:00
3d6020b9cf
* basic caching layer * bit more work, now with an upper size to the cache * cache fix * smarter cache invalidation * One more cache case * Put in a bare skeleton of something else still needed * more logic handling improvements * more work, still not finished * mass-resolve is done in theory, but needs testing * small bugfixin + comments * add note about before/after hooks * LRU-dict fix * when making comments about optimizations, provide historical context * fmt pass
103 lines
3.4 KiB
Python
103 lines
3.4 KiB
Python
from redbot.core import commands
|
|
from redbot.core.config import Config
|
|
from .resolvers import entries_from_ctx, resolve_lists
|
|
|
|
# This has optimizations in it that may not hold True if other parts of the permission
|
|
# model are changed from the state they are in currently.
|
|
# (commit hash ~ 3bcf375204c22271ad3ed1fc059b598b751aa03f)
|
|
#
|
|
# This is primarily to help with the performance of the help formatter
|
|
|
|
# This is less efficient if only checking one command,
|
|
# but is much faster for checking all of them.
|
|
|
|
|
|
async def mass_resolve(*, ctx: commands.Context, config: Config):
|
|
"""
|
|
Get's all the permission cog interactions for all loaded commands
|
|
in the given context.
|
|
"""
|
|
|
|
owner_settings = await config.owner_models()
|
|
guild_owner_settings = await config.guild(ctx.guild).owner_models() if ctx.guild else None
|
|
|
|
ret = {"allowed": [], "denied": [], "default": []}
|
|
|
|
for cogname, cog in ctx.bot.cogs.items():
|
|
|
|
cog_setting = resolve_cog_or_command(
|
|
objname=cogname, models=owner_settings, ctx=ctx, typ="cogs"
|
|
)
|
|
if cog_setting is None and guild_owner_settings:
|
|
cog_setting = resolve_cog_or_command(
|
|
objname=cogname, models=guild_owner_settings, ctx=ctx, typ="cogs"
|
|
)
|
|
|
|
for command in [c for c in ctx.bot.all_commands.values() if c.instance is cog]:
|
|
resolution = recursively_resolve(
|
|
com_or_group=command,
|
|
o_models=owner_settings,
|
|
g_models=guild_owner_settings,
|
|
ctx=ctx,
|
|
)
|
|
|
|
for com, resolved in resolution:
|
|
if resolved is None:
|
|
resolved = cog_setting
|
|
if resolved is True:
|
|
ret["allowed"].append(com)
|
|
elif resolved is False:
|
|
ret["denied"].append(com)
|
|
else:
|
|
ret["default"].append(com)
|
|
|
|
ret = {k: set(v) for k, v in ret.items()}
|
|
|
|
return ret
|
|
|
|
|
|
def recursively_resolve(*, com_or_group, o_models, g_models, ctx, override=False):
|
|
ret = []
|
|
if override:
|
|
current = False
|
|
else:
|
|
current = resolve_cog_or_command(
|
|
typ="commands", objname=com_or_group.qualified_name, ctx=ctx, models=o_models
|
|
)
|
|
if current is None and g_models:
|
|
current = resolve_cog_or_command(
|
|
typ="commands", objname=com_or_group.qualified_name, ctx=ctx, models=o_models
|
|
)
|
|
ret.append((com_or_group, current))
|
|
if isinstance(com_or_group, commands.Group):
|
|
for com in com_or_group.commands:
|
|
ret.extend(
|
|
recursively_resolve(
|
|
com_or_group=com,
|
|
o_models=o_models,
|
|
g_models=g_models,
|
|
ctx=ctx,
|
|
override=(current is False),
|
|
)
|
|
)
|
|
return ret
|
|
|
|
|
|
def resolve_cog_or_command(*, typ, ctx, objname, models: dict) -> bool:
|
|
"""
|
|
Resolves models in order.
|
|
"""
|
|
|
|
resolved = None
|
|
|
|
if objname in models.get(typ, {}):
|
|
blacklist = models[typ][objname].get("deny", [])
|
|
whitelist = models[typ][objname].get("allow", [])
|
|
resolved = resolve_lists(ctx=ctx, whitelist=whitelist, blacklist=blacklist)
|
|
if resolved is not None:
|
|
return resolved
|
|
resolved = models[typ][objname].get("default", None)
|
|
if resolved is not None:
|
|
return resolved
|
|
return None
|