optimization: use slim python image variant

- hiredis compilation fixed

Dockerfile

@@ -3,7 +3,7 @@
 ############
 
 ARG NODE_VERSION=lts
-ARG PYTHON_VERSION=3.12
+ARG PYTHON_VERSION=3.12-slim
 FROM node:${NODE_VERSION} AS build-ui
 
 # env setup
@@ -57,6 +57,9 @@ RUN set -ex; \
   export DEBIAN_FRONTEND=noninteractive; \
   apt-get update; apt-get install --yes --no-install-recommends \
     libmagic1 \
+    # need to build hiredis
+    gcc \
+    libc-dev \
   ; rm -rf /var/lib/apt/lists/*; \
   \
   # install ovdashboard_api

README.md

@@ -54,12 +54,12 @@ It is also heavily advisable that you log into your device using SSH, so you sho
 
 ### Install Base System
 
-`OVDashboard` is designed to run on a `DietPi` installation. Full installation documentation is available [at dietpi.com](https://dietpi.com/docs/deploy/). To quickly get up and running:
+OVDashboard is designed to run on a `DietPi` installation. Full installation documentation is available [at dietpi.com](https://dietpi.com/docs/deploy/). To quickly get up and running:
 
 1. Download Image from [dietpi.com/#download](https://dietpi.com/#download)
 1. Uncompress Image and flash onto SD card – you might need "7zip", "balenaEtcher" and/or other tools
 1. Check the SD card, open "dietpi.txt" and change some options (full documentation [here](https://dietpi.com/docs/usage/#options-within-the-file)):
-   - For WiFi, use `AUTO_SETUP_NET_WIFI_ENABLED=1` and `AUTO_SETUP_NET_WIFI_COUNTRY_CODE=DE`, and also check "dietpi-wifi.txt"
+   - For WiFi, use `AUTO_SETUP_NET_WIFI_ENABLED=1` and `AUTO_SETUP_NET_WIFI_COUNTRY_CODE=DE`, and also edit "dietpi-wifi.txt" like `aWIFI_SSID[0]='OV WLAN'` and `aWIFI_KEY[0]='Strong_pa55w0rd'`
    - System options, e.g. `AUTO_SETUP_AUTOMATED=1`, `AUTO_SETUP_NET_HOSTNAME=OVDashboard`, `AUTO_SETUP_GLOBAL_PASSWORD=dietpi`, `AUTO_SETUP_LOCALE=de_DE.UTF-8`, `AUTO_SETUP_KEYBOARD_LAYOUT=de`, `AUTO_SETUP_TIMEZONE=Europe/Berlin`, `CONFIG_SERIAL_CONSOLE_ENABLE=0`
 1. Be sure to at least change the password (and remember it 🙂️), then put the SD card into your device and boot it. Let the first time setup finish, it will take a bit. It will let you know when it is done!
 1. Log into your device using SSH. By default, that's user name `root` and password `dietpi`
@@ -83,19 +83,32 @@ This can all be done after logging into your prepared device:
 > - The installer will ask for the connected screen's resolution. The default values should be fine, you can likely just hit **Return** here.
 > - The installer will ask for "display languages". This will affect some details on the connected screen. <br />
     For German, enter `de-DE,de,en-US,en`.
+> - The installer will ask you to "review the Docker Compose file" before starting the services. You will want to edit `WEBDAV__HOST`, `WEBDAV__USERNAME` and `WEBDAV__PASSWORD` at least. <br />
+    Refer to [the "Settings"](TODO) for the full list of options.
 
-Afterwards, reboot your device (`reboot` in the terminal). Your OVDashboard should be working now.
+Afterwards, reboot your device (`reboot` in the terminal). 
+
+If the install was successful, your OVDashboard should be showing up on the connected screen after rebooting.
+
+> You will also be able to view your OVDashboard on any webbrowser in your network using `http://<device-ip>`. <br />
+> The device IP is displayed in the lower right region of the OVDashboard.
 
 For a better understanding of your newly created OVDashboard, refer to the [about section](#about-the-default-ovdashboard-deployment).
 
-## Updates, upgrades
-
-<!-- TODO `/opt/ovdashboard` -->
-
 ## Configuration
 
+### "Config" in your WebDAV share: `config.txt`
+
 <!-- TODO -->
 
+### "Settings" on your Device: `/opt/ovdashboard/docker-compose.yml`
+
+<!-- TODO -->
+
+## Updating your Device
+
+<!-- TODO `/opt/ovdashboard` -->
+
 ## Setup for development and contribution
 
 Refer to the specific README files for [the API](./api/README.md) and [the UI](./ui/README.md) to contribute to one of those sub-projects.
@@ -103,17 +116,9 @@ Refer to the specific README files for [the API](./api/README.md) and [the UI](.
 
 ## About the "default" OVDashboard deployment
 
-The default deployment is made up of three parts: The **installer** runs once, prepares your device for OVDashboard usage and deploys the **server**.
+Running the installer script carries out the following actions:
 
-The **server** will then run continuously and make OVDashboard available as a web application in your network.
-
-Also, a browser is installed to display the OVDashboard using your device.
-
-
-### The OVDashboard installer will:
-
-- install and set up Docker, Docker Compose and Chromium-Browser
-- create the OVDashboard Compose project at `/opt/ovdashboard`
-- get a `docker` image from [`code.yavook.de`](https://code.yavook.de/OEKZident.de/-/packages/container/ovdashboard) containing the OVDashboard server
-- set up your device to auto-run Chromium in "kiosk" mode to display the local OVDashboard
-- auto-hide your mouse cursor
+- install Chromium-Browser, Docker and Docker Compose
+- create the OVDashboard project at `/opt/ovdashboard`
+- start the OVDashboard project, this deploys the `ovdashboard` service from [`code.yavook.de`](https://code.yavook.de/OEKZident.de/-/packages/container/ovdashboard) and a [`redis` instance](https://redis.io/) to your device
+- set up your device to auto-boot into Chromium "kiosk" mode to display the local OVDashboard

api/ovdashboard_api/routers/v1/image.py

@@ -63,13 +63,23 @@ async def find_images_by_prefix(
     response_class=StreamingResponse,
 )
 async def get_image_by_prefix(
+    cfg: Config = Depends(get_config),
     remote_path: str = Depends(RP_IMAGE),
     name: str = Depends(LM_IMAGE.getter.func),
 ) -> StreamingResponse:
-    cfg = await get_config()
     img = Image.open(BytesIO(await WebDAV.read_bytes(f"{remote_path}/{name}")))
 
     img_buffer = BytesIO()
+
+    width, height = img.size
+    target_height = cfg.image.height
+    target_width = int(width * target_height / height)
+
+    img = img.resize(
+        size=(target_width, target_height),
+        resample=Image.LANCZOS,
+    )
+
     img.save(img_buffer, **cfg.image.save_params)
     img_buffer.seek(0)
 

deploy/chores/docker_buildx

@@ -6,13 +6,12 @@ script_dir="$( dirname "${script}" )"
 # shellcheck disable=SC1091
 . "${script_dir}/check_version"
 
-install_version="$( \
-    grep '^ovd_version' "${script_dir}/../install.sh" \
-    | sed -E 's/^ovd_version[^0-9]*((0|[1-9][0-9]*)[0-9\.]*[0-9]).*$/\1/'
-)"
+# defined in `check_version` script
+# shellcheck disable=SC2154
+echo "${git_version}" >/dev/null
 
 docker buildx build \
     --pull --push \
-    --tag "code.yavook.de/oekzident.de/ovdashboard:${install_version}" \
+    --tag "code.yavook.de/oekzident.de/ovdashboard:${git_version}" \
     --platform "linux/amd64,linux/arm64" \
     "${script_dir}/../.."

deploy/install.sh

@@ -67,18 +67,32 @@ echo '{"intl":{"selected_languages":"'"${display_lang}"'"}}' \
 
 mkdir -p /opt/ovdashboard
 
-# compose file
+# prepare compose project
 curl \
     --proto "=https" --tlsv1.2 -sSf \
     --output "/opt/ovdashboard/docker-compose.yml" \
     "https://code.yavook.de/OEKZident.de/ovdashboard/raw/tag/v${ovd_version}/deploy/docker-compose.yml"
-echo "Please review the Docker Compose file before continuing! [press Enter]"
-read -r _ENTER
+docker compose \
+    --project-directory "/opt/ovdashboard" \
+    pull
+
+# review compose file
+echo "Please review the Docker Compose file before continuing! [hit Return]"
+read -r _RETURN
 nano "/opt/ovdashboard/docker-compose.yml"
 
 # start server
 docker compose \
-    --file "/opt/ovdashboard/docker-compose.yml" \
     --project-directory "/opt/ovdashboard" \
-    --project-name "ovdashboard" \
     up --detach
+
+############
+# finalize #
+############
+
+echo ""
+echo "#########################"
+echo "# OVDashboard Installed #"
+echo "#########################"
+echo ""
+echo "You can now reboot your device."