mirror of
https://github.com/kuhyx/signal-bot.git
synced 2026-11-10 18:07:29 +01:00
Krzysztof kuhy Rudnicki
405541be9b
One row per command attempt, not two aggregates written only on success. command_uses holds last-use only and daily_uses counts successes; both are rate-limiter state and both are written by store.record_use, which the dispatcher calls only when a handler returns cleanly. A gate refusal and a crash returned before it, so a command that failed every time looked exactly like one nobody ran, and nothing knew which command a counted message was. command_events now takes every attempt with its outcome, its duration and the item it served; upstream_calls takes every outbound request made while serving one. The two writes stay separate on purpose -- recording a refusal as a use is what slid the cooldown window forward every time somebody was told to wait -- and test_dispatcher_outcomes asserts last_used does not move. Backfilled from journald joined against message_events, with daily_uses filling the residual so the two sources cannot double-count. Two things were wrong on the first attempt and both were found by measuring rather than by reasoning: the window has to be symmetric, because sent_at_ms comes from the sender's phone and the journal stamp from this host (194 of 299 lines landed *before* their own message, by up to 3.2 s); and matching has to be a global closest-pair assignment, because per-row greedy cannot separate two commands seconds apart. Together those took the dropped rows from 137 to zero, and recovered 418 invocations back to 29 August. Also: !analityka in the group, twelve Polish lines, naming at most three people and only ones whose display name has been seen; an Analytics tab with Nivo charts; and continuous latency measurement, which answers the open question in DOCS-latency.md -- a text send is 235 ms and an image send about 2.2 s, so the upload really is where the time goes. EPHEMERAL_TABLES was doing two unrelated jobs and is now two tuples: what a schema bump rebuilds, and what leaving a group clears. daily_uses left the first (it is a backfill source now) and stayed in the second, and _drop_reshaped took over the repair a bump used to do for it. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01T4QTp3h5Q3MvfrwGV8xqPf
74 lines
2.7 KiB
Python
74 lines
2.7 KiB
Python
# Copyright (c) 2026 Krzysztof Rudnicki. MIT licence.
|
|
"""What a command handler is allowed to touch, and how it answers.
|
|
|
|
Split out of registry.py at the 250-line cap. The division is real rather
|
|
than clerical: that file is "which word runs what", this one is "what the
|
|
thing that runs may do" -- and the reply helpers are the only place in the
|
|
codebase that decides *where* an answer goes.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
from dataclasses import dataclass
|
|
from typing import TYPE_CHECKING
|
|
|
|
from signal_bot.analytics.probe import timed_send
|
|
from signal_bot.quoting import quote_of
|
|
|
|
if TYPE_CHECKING:
|
|
from signal_bot.config import Config
|
|
from signal_bot.models import IncomingMessage
|
|
from signal_bot.store import Store
|
|
from signal_bot.transport_base import Transport
|
|
|
|
|
|
@dataclass(frozen=True, slots=True)
|
|
class CommandContext:
|
|
"""Everything a command handler is allowed to touch."""
|
|
|
|
message: IncomingMessage
|
|
group_id: str
|
|
transport: Transport
|
|
store: Store
|
|
config: Config
|
|
command_name: str
|
|
now: float
|
|
day: str
|
|
# Whatever followed the trigger, split on whitespace. Empty for every
|
|
# command that does not opt into taking any -- see Command.takes_args.
|
|
args: tuple[str, ...] = ()
|
|
|
|
async def reply(self, text: str) -> None:
|
|
"""Answer the message that triggered this command, in its group.
|
|
|
|
There is no configurable reply target. A command answers where it
|
|
was asked, which is the only routing that stays correct once the
|
|
bot watches a test group and a production group at once -- and it
|
|
quotes the asker, so a busy group can tell which "!cat" this is.
|
|
"""
|
|
async with timed_send():
|
|
await self.transport.send_text(text, self.group_id, quote_of(self.message))
|
|
|
|
async def reply_image(self, image_base64: str, caption: str = "") -> None:
|
|
"""Answer the triggering message with an image, quoting it."""
|
|
async with timed_send():
|
|
await self.transport.send_image(
|
|
image_base64, self.group_id, quote_of(self.message), caption
|
|
)
|
|
|
|
async def reply_media(self, payload_base64: str, media_type: str, filename: str) -> None:
|
|
"""Answer with an attachment that is not necessarily a photo.
|
|
|
|
Separate from reply_image because Signal decides how to render an
|
|
attachment from what it is told it is, and "a jpeg" is a lie that
|
|
makes a video arrive as a file nobody can play.
|
|
"""
|
|
async with timed_send():
|
|
await self.transport.send_media(
|
|
payload_base64,
|
|
self.group_id,
|
|
quote_of(self.message),
|
|
media_type=media_type,
|
|
filename=filename,
|
|
)
|