# What this PR does
It _appears_ like Telegram may have changed one of the error messages
they return for `telegram.error.BadRequest`. This _may_ be causing us to
infinitely retry some of these tasks.
Previously we were checking for two variants of the same type of error
message:
- "Message to reply not found"
- "Replied message not found"
_However_, if I search for the following [in the
logs](https://ops.grafana-ops.net/goto/hMgBb8CSR?orgId=1):
```logql
{namespace="amixr-prod"} |~ `(Message to be replied not found|Message to reply not found|Replied message not found)`
````
I _only_ see references to "Message to be replied not found". I have
updated references to the former to this new error log message we are
seeing.
Also:
- deduplicate some of the words we check for in
`telegram.error.BadRequest` and `telegram.error.Unauthorized` into
`apps.telegram.client.TelegramClient.BadRequestMessage` and
`apps.telegram.client.TelegramClient.UnauthorizedMessage` respectively
- deduplicate some of the wording we use in the `reason` arg passed to
`TelegramToUserConnector.create_telegram_notification_error` into
`apps.telegram.models.connectors.personal.TelegramToUserConnector.NotificationErrorReason`
- standardize how we check the `message` attribute of
`telegram.error.TelegramError`s into a new `error_message_is` static
method on `apps.telegram.client.TelegramClient`
- previously we would check these error messages in two different ways:
```python3
# style 1
if "error message to check" in e.message:
# do something
# style 2
if error.message == "error message to check":
# do something
```
## Which issue(s) this PR closes
Closes https://github.com/grafana/oncall-private/issues/2868
## Checklist
- [x] Unit, integration, and e2e (if applicable) tests updated
- [x] Documentation added (or `pr:no public docs` PR label added if not
required)
- [x] Added the relevant release notes label (see labels prefixed w/
`release:`). These labels dictate how your PR will
show up in the autogenerated release notes.
# What this PR does
Restart `send_link_to_channel_message_or_fallback_to_full_alert_group`
task without retry on the first run if alert group telegram message
doesn't exist. On the second run raise Exception
## Which issue(s) this PR fixes
https://github.com/grafana/oncall-private/issues/2492
## Checklist
- [x] Unit, integration, and e2e (if applicable) tests updated
- [x] Documentation added (or `pr:no public docs` PR label added if not
required)
- [x] `CHANGELOG.md` updated (or `pr:no changelog` PR label added if not
required)
# What this PR does
Fixes https://github.com/grafana/oncall-private/issues/2154
## Checklist
- [x] Unit, integration, and e2e (if applicable) tests updated
- [x] Documentation added (or `pr:no public docs` PR label added if not
required)
- [x] `CHANGELOG.md` updated (or `pr:no changelog` PR label added if not
required)
# What this PR does
```bash
❯ mypy .
Success: no issues found in 595 source files
```
- re-enable the mypy CI check
- fixes all `django-manager-missing` mypy errors
- disable all other rules currently giving mypy errors
- changing the approach here. rather than enforcing that backend
contributors fix >= 1 `mypy` error on their PR, lets simply disable all
the rules that're currently returning errors and slowly re-enable these
one at a time #2392
## Checklist
- [ ] Unit, integration, and e2e (if applicable) tests updated (N/A)
- [ ] Documentation added (or `pr:no public docs` PR label added if not
required) (N/A)
- [ ] `CHANGELOG.md` updated (or `pr:no changelog` PR label added if not
required) (N/A)
# What this PR does
Makes Telegram integration consistent with the rest of the system so it
uses the word "alert group" instead of "incident" when referring to
alert groups.
## Checklist
- [x] Tests updated
- [ ] Documentation added (N/A)
- [x] `CHANGELOG.md` updated
* allow posting DM messages to Telegram without connecting a channel
* fix get_channel_for_alert_group, add tests
* add docs on usage without channel connection
