Files
dpy_webhooks/README.md
T
dsql ce48f2aea8 fix: skip token-less webhooks in pick(); self-heal deletes the dead pick, not the managed webhook
dpywebhooks-1 (HIGH): pick()/send_any could select token-less channel-follower
(type 2) or application-owned (type 3) webhooks returned by channel.webhooks().
Webhook.send raises a raw ValueError on those before any HTTP call, and
send_any only caught discord.HTTPException, so the ValueError bypassed
fallback and self-heal entirely. strategy='first' picks the lowest id, so any
channel following an announcement channel could be permanently broken. Fix:
pick() now filters candidates to w.token is not None.

dpywebhooks-2 (HIGH): send_any's self-heal always called clear(), which
targets the managed webhook (cache/store), not the dead webhook that was
actually picked. When the dead pick and the managed webhook differed, this
deleted a healthy managed webhook and churned delete+create toward the
10-cap on every send. Fix: self-heal now deletes the dead picked webhook
directly (bot-auth webhook.delete() authorizes on it without its own token);
clear() is only used when the pick IS the managed webhook.

dpywebhooks-8 (nit): count() and _create() now read live webhooks through
list() instead of calling channel.webhooks() directly, so the token filter
and any future list() change reach all three call sites from one source.

Version 0.1.0 -> 0.1.1. README + module docstrings updated to document the
token filter and the corrected self-heal delete target.

Signed-off-by: disqualifier <dev@disqualifier.me>
2026-07-03 15:31:06 -04:00

97 lines
4.4 KiB
Markdown

# dpy_webhooks
Per-channel Discord webhook management for discord.py: get-or-create one webhook per channel,
persist its `{id, token}` so it survives restarts, enforce Discord's 10-per-channel cap, and
send through it. One webhook per channel, reused across sends — instead of creating a new one
each time (wasteful, rate-limited, and it leaks toward the cap).
## Install
```
dpy_webhooks @ git+ssh://git@git.rethinkstudios.io/rethink-public/dpy_webhooks.git@v0.1.1
```
## Usage
```python
from dpy_webhooks import DPYWebhooks
hooks = DPYWebhooks(bot) # store optional; in-memory if omitted
await hooks.send(channel, content="hi", embeds=[e]) # get-or-create + send
wh = await hooks.get_or_create(channel) # reuse the webhook directly
```
With a persistence store so webhooks survive a restart:
```python
hooks = DPYWebhooks(bot, store=my_store)
```
Work with **all** of a channel's existing webhooks (read live, no store, no create) — for
discovery, load-spreading, or rotation:
```python
await hooks.list(channel) # every live webhook on the channel
await hooks.pick(channel, strategy="round_robin") # select one existing (None if none)
await hooks.send_any(channel, strategy="round_robin", content="hi") # send via a picked one
```
`pick` strategies are `"first"` (oldest by id), `"random"`, and `"round_robin"` (per-channel
in-process index that wraps on the live count and rebuilds if the count changed); an unknown
strategy raises `ValueError`. `pick` only considers webhooks that have a token — channel-
follower and application-owned webhooks have none and can't be sent through, so they're
skipped rather than picked. `send_any` sends through a picked existing webhook, **falling
back to `get_or_create`** when the channel has none. On a dead pick, self-heal deletes that
specific dead webhook (not the managed one) and recreates + retries once — unless the pick
IS the managed webhook, in which case it goes through the normal `clear()` path. None of
these three create or touch the store — they read live from the channel.
## What you inject
- **`client`** — your `discord.Client` / `commands.Bot`. Used to rebuild persisted webhooks
via `Webhook.partial(id, token, client=...)`, so discord.py supplies its own session and
state. Requires **`discord.py>=2.2`**.
- **`store`** (optional) — any object matching the `WebhookStore` protocol. Omit it and the
lib uses a non-durable in-memory store (records lost on restart).
### The store protocol
```python
class WebhookStore(Protocol):
async def get(self, channel_id: int) -> dict | None: ... # {"webhook_id", "token", "guild_id"} or None
async def set(self, channel_id: int, record: dict) -> None: ...
async def delete(self, channel_id: int) -> None: ...
```
A file- or mongo-backed store satisfies this — the lib depends only on the protocol, not on
any concrete store class. `InMemoryWebhookStore` is the bundled default.
## API & contract
The module docstring (`help(dpy_webhooks)` / IDE hover) is the source of truth. In short:
- `get_or_create(channel, *, evict_oldest=False) -> discord.Webhook`
- `get(channel) -> discord.Webhook | None` — cache then store; never creates
- `send(channel, **kwargs) -> discord.WebhookMessage` — get-or-create then send, with
self-heal
- `clear(channel) -> None` — delete server-side + store record (idempotent)
- `count(channel) -> int` — webhooks currently on the channel
- `list(channel) -> list[discord.Webhook]` — all live webhooks on the channel (no create)
- `pick(channel, *, strategy="first") -> discord.Webhook | None` — select an existing one
- `send_any(channel, *, strategy="round_robin", **kwargs) -> discord.WebhookMessage` — send
via a picked existing webhook, falling back to `get_or_create`
**Self-heal.** If `send` hits a dead webhook (deleted server-side / invalid token), the lib
clears the record, recreates the webhook, and retries the send **once**; a second failure
raises loud.
**Fail-loud.** Nothing is swallowed to `None`. A channel already at 10 webhooks raises
`WebhookCapacityError` (pass `evict_oldest=True` to reclaim the oldest instead). Missing
Manage Webhooks perms propagate discord's own `Forbidden` unwrapped. Store failures propagate
— the store owns its durability.
## Versioning
Tagged `vX.Y.Z`; pin a tag in your install line. Targets `discord.py>=2.2` (not
`discord.py-self`).