[WIP] add honeybadger alerts on exception

README.md

@@ -27,7 +27,7 @@ $ aws sqs create-queue --queue-name sul-speech-to-text-todo-your-username
 $ aws sqs create-queue --queue-name sul-speech-to-text-done-your-username
 ```
 
-Create a bucket: 
+Create a bucket:
 
 ```shell
 aws s3 mb s3://sul-speech-to-text-dev-your-username
@@ -238,3 +238,52 @@ pytest
 ```
 
 Note: the tests use the [moto](https://docs.getmoto.org/en/latest/) library to mock out AWS resources. If you want to test live AWS you can follow the steps above to create a job, run, and then receive the done message.
+
+### Linting and typechecking
+
+This project uses [Ruff](https://docs.astral.sh/ruff/) for typechecking.  If you've set up your virtual environment as described above, you can do:
+
+```shell
+ruff check
+```
+
+Which might get you output like
+```
+% ruff check
+tests/test_speech_to_text.py:12:27: F401 [*] `unittest.mock.MagicMock` imported but unused
+   |
+11 | import pytest
+12 | from unittest.mock import MagicMock
+   |                           ^^^^^^^^^ F401
+13 | 
+14 | BUCKET = "bucket"
+   |
+   = help: Remove unused import: `unittest.mock.MagicMock`
+
+Found 1 error.
+```
+
+or (hopefully)
+
+```
+% ruff check
+All checks passed!
+```
+
+And we use [mypy](https://www.mypy-lang.org/) for typechecking.  You can run it like so:
+```shell
+mypy .
+```
+
+Which might get you something like
+```
+% mypy .
+speech_to_text.py:89: error: Incompatible return value type (got "None", expected "dict[Any, Any]")  [return-value]
+Found 1 error in 1 file (checked 3 source files)
+```
+
+but hopefully gets you something like
+```
+% mypy .
+Success: no issues found in 3 source files
+```

env-example

@@ -5,3 +5,5 @@ AWS_ROLE_ARN=arn:aws:iam::418214828013:role/DevelopersRole
 SPEECH_TO_TEXT_S3_BUCKET=sul-speech-to-text-dev-your-username
 SPEECH_TO_TEXT_TODO_SQS_QUEUE=sul-speech-to-text-todo-dev-your-username
 SPEECH_TO_TEXT_DONE_SQS_QUEUE=sul-speech-to-text-done-dev-your-username
+HONEYBADGER_API_KEY=CHANGE_ME
+HONEYBADGER_ENV=stage

mypy.ini

@@ -1,4 +1,2 @@
 [mypy]
-
-[mypy-whisper.*]
 ignore_missing_imports = True

requirements.txt

@@ -1,8 +1,10 @@
 boto3
+boto3-stubs[s3,sqs,sts]
 ffprobe3
+honeybadger
+moto[s3,sqs,sts]
+mypy
 openai-whisper
 python-dotenv
 pytest
-moto[s3,sqs,sts]
-mypy
-boto3-stubs[s3,sqs,sts]
+ruff

speech_to_text.py

@@ -10,6 +10,7 @@
 import shutil
 import subprocess
 import traceback
+from honeybadger import honeybadger
 from functools import cache
 from pathlib import Path
 from typing import Optional, Dict
@@ -22,6 +23,11 @@
 from mypy_boto3_s3.service_resource import Bucket, S3ServiceResource
 from mypy_boto3_sqs.service_resource import SQSServiceResource, Queue
 
+honeybadger.configure(
+    api_key=os.environ.get("HONEYBADGER_API_KEY", ""),
+    environment=os.environ.get("HONEYBADGER_ENV", "stage"),
+)
+
 
 def main(daemon=True) -> None:
     # loop forever looking for jobs unless daemon option says not to
@@ -277,6 +283,9 @@ def report_error(message: str, job: Optional[Dict], traceback: str) -> None:
     """
     full_message = message + "\n" + traceback
     logging.exception(full_message)
+    honeybadger.notify(
+        "Whisper AWS process \n" + message, context={"job": job, "traceback": traceback}
+    )
 
     # it's possible that we are reporting an error without a job
     # we can only send a message to the DONE queue if we have a job!