[enh] container: tidy builds (#5086)

Building the container currently does not work properly.
When rebuilding several times with `make container`, `version_frozen.py`
is recreated, which wouldn't be an issue if the file’s timestamp was constant.
Now, when creating `version_frozen.py`, it will have the same timestamp as the
commit when it was created. (`version_frozen.py` is moved to a dedicated layer).

Reusing "builder" cache when building "dist" could be slow
(CD reports 2 seconds, but locally I've seen it take up to 10 seconds),
so the Dockerfile is now split and we save a couple steps
by importing the "builder" image directly.

The last changes made it possible to remove the layer cache in "builder",
since the overhead is now greater than building the layers from scratch.

Until now, all "dist" layers were squashed into a single layer,
which in most cases is a good idea
(except for storage/delivery pricing/overhead), but in our case,
since we manage the entire pipeline, we can ignore this
and share layers between builds.
This means (for example) that if we change files unrelated to the container
in several consecutive commits (documentation changes), we don't have to push
the entire image to registry, but only the different layers
(`version_frozen.py` in this example).
The same applies when pulling, as only the layers that have changed
compared to the local layers will be downloaded (that's the theory,
we'll see if this works as expected or if we need to tweak something else).

container/Dockerfile

@@ -1,61 +0,0 @@
-FROM ghcr.io/searxng/base:searxng-builder AS builder
-
-COPY ./requirements*.txt ./
-
-RUN --mount=type=cache,id=pip,target=/root/.cache/pip python -m venv ./venv \
- && . ./venv/bin/activate \
- && pip install -r requirements.txt \
- && pip install -r requirements-server.txt
-
-COPY ./searx/ ./searx/
-
-ARG TIMESTAMP_SETTINGS="0"
-
-RUN python -m compileall -q searx \
- && touch -c --date=@$TIMESTAMP_SETTINGS ./searx/settings.yml \
- && find ./searx/static \
-    \( -name "*.html" -o -name "*.css" -o -name "*.js" -o -name "*.svg" -o -name "*.ttf" -o -name "*.eot" \) \
-    -type f -exec gzip -9 -k {} + -exec brotli --best {} +
-
-FROM ghcr.io/searxng/base:searxng AS dist
-
-ARG LABEL_DATE="0001-01-01T00:00:00Z"
-ARG GIT_URL="unspecified"
-ARG SEARXNG_GIT_VERSION="unspecified"
-ARG LABEL_VCS_REF="unspecified"
-ARG LABEL_VCS_URL="unspecified"
-
-COPY --chown=searxng:searxng --from=builder /usr/local/searxng/venv/ ./venv/
-COPY --chown=searxng:searxng --from=builder /usr/local/searxng/searx/ ./searx/
-COPY --chown=searxng:searxng ./container/template/ ./.template/
-COPY --chown=searxng:searxng ./container/entrypoint.sh ./entrypoint.sh
-
-LABEL org.opencontainers.image.authors="searxng <$GIT_URL>" \
-      org.opencontainers.image.created="$LABEL_DATE" \
-      org.opencontainers.image.description="A privacy-respecting, hackable metasearch engine" \
-      org.opencontainers.image.documentation="https://github.com/searxng/searxng-docker" \
-      org.opencontainers.image.licenses="AGPL-3.0-or-later" \
-      org.opencontainers.image.revision="$LABEL_VCS_REF" \
-      org.opencontainers.image.source="$LABEL_VCS_URL" \
-      org.opencontainers.image.title="searxng" \
-      org.opencontainers.image.url="$LABEL_VCS_URL" \
-      org.opencontainers.image.version="$SEARXNG_GIT_VERSION"
-
-ENV SEARXNG_VERSION="$SEARXNG_GIT_VERSION" \
-    SEARXNG_SETTINGS_PATH="$CONFIG_PATH/settings.yml" \
-    GRANIAN_PROCESS_NAME="searxng" \
-    GRANIAN_INTERFACE="wsgi" \
-    GRANIAN_HOST="::" \
-    GRANIAN_PORT="8080" \
-    GRANIAN_WEBSOCKETS="false" \
-    GRANIAN_LOOP="uvloop" \
-    GRANIAN_BLOCKING_THREADS="4" \
-    GRANIAN_WORKERS_KILL_TIMEOUT="30s" \
-    GRANIAN_BLOCKING_THREADS_IDLE_TIMEOUT="5m"
-
-VOLUME $CONFIG_PATH
-VOLUME $DATA_PATH
-
-EXPOSE 8080
-
-ENTRYPOINT ["/usr/local/searxng/entrypoint.sh"]

container/builder.dockerfile

@@ -0,0 +1,24 @@
+FROM ghcr.io/searxng/base:searxng-builder AS builder
+
+COPY ./requirements*.txt ./
+
+RUN --mount=type=cache,id=pip,target=/root/.cache/pip set -eux; \
+    python -m venv ./.venv/; \
+    . ./.venv/bin/activate; \
+    pip install -r ./requirements.txt -r ./requirements-server.txt
+
+COPY ./searx/ ./searx/
+
+ARG TIMESTAMP_SETTINGS="0"
+
+RUN set -eux; \
+    python -m compileall -q ./searx/; \
+    touch -c --date=@$TIMESTAMP_SETTINGS ./searx/settings.yml; \
+    find ./searx/static/ -type f \
+        \( -name "*.html" -o -name "*.css" -o -name "*.js" -o -name "*.svg" \) \
+        -exec gzip -9 -k {} + \
+        -exec brotli -9 -k {} + \
+        -exec gzip --test {}.gz + \
+        -exec brotli --test {}.br +; \
+    # Move always changing files to /usr/local/searxng/
+    mv ./searx/version_frozen.py ./

container/dist.dockerfile

@@ -0,0 +1,44 @@
+FROM ghcr.io/searxng/base:searxng AS dist
+
+ARG CONTAINER_IMAGE_ORGANIZATION="searxng"
+ARG CONTAINER_IMAGE_NAME="searxng"
+
+COPY --chown=searxng:searxng --from=localhost/$CONTAINER_IMAGE_ORGANIZATION/$CONTAINER_IMAGE_NAME:builder /usr/local/searxng/.venv/ ./.venv/
+COPY --chown=searxng:searxng --from=localhost/$CONTAINER_IMAGE_ORGANIZATION/$CONTAINER_IMAGE_NAME:builder /usr/local/searxng/searx/ ./searx/
+COPY --chown=searxng:searxng ./container/ ./
+COPY --chown=searxng:searxng --from=localhost/$CONTAINER_IMAGE_ORGANIZATION/$CONTAINER_IMAGE_NAME:builder /usr/local/searxng/version_frozen.py ./searx/
+
+ARG CREATED="0001-01-01T00:00:00Z"
+ARG VERSION="unknown"
+ARG VCS_URL="unknown"
+ARG VCS_REVISION="unknown"
+
+LABEL org.opencontainers.image.created="$CREATED" \
+      org.opencontainers.image.description="SearXNG is a metasearch engine. Users are neither tracked nor profiled." \
+      org.opencontainers.image.documentation="https://docs.searxng.org/admin/installation-docker" \
+      org.opencontainers.image.licenses="AGPL-3.0-or-later" \
+      org.opencontainers.image.revision="$VCS_REVISION" \
+      org.opencontainers.image.source="$VCS_URL" \
+      org.opencontainers.image.title="SearXNG" \
+      org.opencontainers.image.url="https://searxng.org" \
+      org.opencontainers.image.version="$VERSION"
+
+ENV SEARXNG_VERSION="$VERSION" \
+    SEARXNG_SETTINGS_PATH="$CONFIG_PATH/settings.yml" \
+    GRANIAN_PROCESS_NAME="searxng" \
+    GRANIAN_INTERFACE="wsgi" \
+    GRANIAN_HOST="::" \
+    GRANIAN_PORT="8080" \
+    GRANIAN_WEBSOCKETS="false" \
+    GRANIAN_LOOP="uvloop" \
+    GRANIAN_BLOCKING_THREADS="4" \
+    GRANIAN_WORKERS_KILL_TIMEOUT="30s" \
+    GRANIAN_BLOCKING_THREADS_IDLE_TIMEOUT="5m"
+
+# "*_PATH" ENVs are defined in base images
+VOLUME $CONFIG_PATH
+VOLUME $DATA_PATH
+
+EXPOSE 8080
+
+ENTRYPOINT ["/usr/local/searxng/entrypoint.sh"]

container/entrypoint.sh

@@ -127,4 +127,4 @@ volume_handler "$DATA_PATH"
 # Check for files
 config_handler "$SEARXNG_SETTINGS_PATH" "/usr/local/searxng/searx/settings.yml"
 
-exec /usr/local/searxng/venv/bin/granian searx.webapp:app
+exec /usr/local/searxng/.venv/bin/granian searx.webapp:app