Run export on a cron schedule

Dockerfile

@@ -3,13 +3,20 @@ FROM dastapov/hledger:1.26 as hledger
 
 FROM library/python:3.10-slim
 
+RUN apt install cron && \
+    apt-get clean && \
+    rm -rf /var/lib/{apt,dpkg,cache,log}
+
 COPY --from=hledger /usr/bin/hledger /usr/bin/hledger
 
 COPY src /code
+COPY docker-entrypoint.sh /entrypoint.sh
 
 # TODO: Support more than one ledger file
 ENV LEDGER_FILE=/data/all.ledger
 ENV OUTPUT_FILE=/data/all.sqlite
 ENV OVERWRITE_OUTPUT=true
+ENV CRONTAB_SCHEDULE="0 15 * * *"
 
-ENTRYPOINT ["python", "/code/main.py"]
+ENTRYPOINT ["sh", "/entrypoint.sh"]
+CMD ["crond", "-f", "-l", "0"]

README.md

@@ -6,6 +6,28 @@ Exports are not designed to be consistent across iterations. It is recommended t
 
 ## Usage
 
-TBD
+The default configuration is to run in a long-lived mode where the conversion will happen on a regular schedule defined by cron. You can use [crontab guru](https://crontab.guru) to build the value for the `CRONTAB_SCHEDULE` variable.
 
-Currently planning on designing the container to be able to run as a long-lived job with a cron schedule for exporting. There will also be a method for running as a one-off.
+```
+docker run \
+    --name ledger-sqlite \
+    --env "LEDGER_FILE=/data/journal.ledger" \
+    --env "OUTPUT_FILE=/data/journal.sqlite" \
+    --env "CRONTAB_SCHEDULE=0 15 * * *" \
+    -v /path/to/data:/data \
+    ghcr.io/tgrosinger/ledger-sqlite:latest
+```
+
+Alternatively, a single conversion can be performed and the container will exit when complete. This can also be useful for debugging as the logs are easier to retrieve.
+
+```
+docker run \
+    --rm
+    --env "LEDGER_FILE=/data/journal.ledger" \
+    --env "OUTPUT_FILE=/data/journal.sqlite" \
+    -v /path/to/data:/data \
+    --entrypoint python \
+    ghcr.io/tgrosinger/ledger-sqlite:latest \
+    /code/main.py
+
+```

docker-entrypoint.sh

@@ -0,0 +1,9 @@
+#!/bin/sh
+
+# Inspired by https://github.com/dkruger/docker-cron
+
+cat << EOF > /var/spool/cron/crontabs/root
+${CRONTAB_ENTRY} python /code/main.py
+EOF
+
+exec "$@"