Make docker compose more seamless (#7693)

- Add docker compose files to https://github.com/cBioPortal/cbioportal-docker-compose, initial work by @nr23730 & the Hyve - thanks!  I figured it might be cleaner to keep the docker compose out of this repo, since it is using containers from various other repos (e.g. session-service)
- Update docs to recommend to use that docker compose repo to set up a local instance
- Update entrypoint for web-and-data image to include a check for database liveness and migrate the db automatically

.dockerignore

@@ -4,6 +4,8 @@
 .DS_Store
 .github
 target
-docker
 docs
 */target/*
+docker
+# don't ignore entrypoint scripts
+!docker/web-and-data/docker-entrypoint.sh

README.md

@@ -13,7 +13,7 @@ See [https://docs.cbioportal.org](https://docs.cbioportal.org)
 See [LICENSE](./LICENSE)
 
 ## 💻 Run Backend
-cBioPortal consists of several components, please read the [Architecture docs](https://docs.cbioportal.org/2.1-deployment/architecture-overview) to figure out what repo would be relevant to edit. If you e.g. only want to make frontend changes, one can directly edit [the frontend repo](https://github.com/cbioportal/cbioportal-frontend) instead. Read the instructions in that repo for more info on how to do frontend development. This repo only contains the backend part. Before editing the backend, it's good to read the [backend code organization](docs/Backend-Code-Organization.md). For development of the backend repo one should first set up a database. Please follow the [Docker deployment documentation](https://docs.cbioportal.org/2.1.1-deploy-with-docker-recommended/docker) to do so. In step 6 instead of using the `latest` image, you can create your own image and use that:
+cBioPortal consists of several components, please read the [Architecture docs](https://docs.cbioportal.org/2.1-deployment/architecture-overview) to figure out what repo would be relevant to edit. If you e.g. only want to make frontend changes, one can directly edit [the frontend repo](https://github.com/cbioportal/cbioportal-frontend) instead. Read the instructions in that repo for more info on how to do frontend development. This repo only contains the backend part. Before editing the backend, it's good to read the [backend code organization](docs/Backend-Code-Organization.md). For development of the backend repo one should first set up a database. Please follow the [Docker deployment documentation](https://docs.cbioportal.org/2.1.1-deploy-with-docker-recommended/docker) to do so. Change the [docker-compose](https://github.com/cBioPortal/cbioportal-docker-compose/blob/5da068f0eb9b4f42db52ab5e91321b26a1826d7a/docker-compose.yml#L6) file to use your image instead:
 
 ```
 docker build -t cbioportal/cbioportal:my-dev-cbioportal-image -f docker/web-and-data/Dockerfile .

docker/web-and-data/Dockerfile

@@ -74,3 +74,7 @@ ENV PORTAL_HOME=/cbioportal
 
 RUN mkdir -p $PORTAL_WEB_HOME
 RUN unzip /app.war -d  $PORTAL_WEB_HOME
+
+# add entrypoint
+COPY --from=build /cbioportal/docker/web-and-data/docker-entrypoint.sh /usr/local/bin/
+ENTRYPOINT ["docker-entrypoint.sh"]

docker/web-and-data/docker-entrypoint.sh

@@ -0,0 +1,86 @@
+#!/usr/bin/env bash
+set -eo pipefail
+shopt -s nullglob
+
+BAKED_IN_WAR_CONFIG_FILE=/cbioportal-webapp/WEB-INF/classes/portal.properties
+CUSTOM_PROPERTIES_FILE=/cbioportal/portal.properties
+
+# check to see if this file is being run or sourced from another script
+_is_sourced() {
+    # https://unix.stackexchange.com/a/215279
+    [ "${#FUNCNAME[@]}" -ge 2 ] \
+            && [ "${FUNCNAME[0]}" = '_is_sourced' ] \
+                && [ "${FUNCNAME[1]}" = 'source' ]
+}
+
+parse_db_params_from_command_line() {
+    echo $@ | sed 's/-D/\n-D/g' | grep -- '-Ddb' | sed 's/-D//g' | grep db.
+}
+
+parse_db_params_from_config_and_command_line() {
+    if [[ -f $CUSTOM_PROPERTIES_FILE ]]; then
+        PROPERTIES_FILE=$CUSTOM_PROPERTIES_FILE
+    else
+        PROPERTIES_FILE=$BAKED_IN_WAR_CONFIG_FILE
+    fi
+
+    for param in db.host db.user db.portal_db_name db.password; do
+        if $(parse_db_params_from_command_line $@ | grep -q $param); then
+            parse_db_params_from_command_line $@ | grep "^$param"
+        else
+            grep -v '^#' $PROPERTIES_FILE | grep "^$param"
+        fi
+    done
+}
+
+check_db_connection() {
+    eval "$(parse_db_params_from_config_and_command_line $@ | sed 's/^db\./db_/g')"
+    POTENTIAL_DB_PARAMS=$@
+
+    while ! mysqladmin ping -s -h$(echo ${db_host} | cut -d: -f1) -u${db_user} -p${db_password};
+    do
+        sleep 5s;
+        if [ -n "$SHOW_DEBUG_INFO" ] && [ "$SHOW_DEBUG_INFO" != "false" ]; then
+            echo mysqladmin ping -s -h$(echo ${db_host} | cut -d: -f1) -u${db_user} -p${db_password}
+        fi
+        echo "Database not available yet (first time can take a few minutes to load seed database)... Attempting reconnect..."
+    done
+    echo "Database connection success"
+}
+
+migrate_db() {
+    echo "Migrating database if necessary..."
+    POTENTIAL_DB_PARAMS=$@
+
+    if [[ -f $CUSTOM_PROPERTIES_FILE ]]; then
+        python3 /cbioportal/core/src/main/scripts/migrate_db.py -y -p $CUSTOM_PROPERTIES_FILE -s /cbioportal/db-scripts/src/main/resources/migration.sql
+    else
+        python3 /cbioportal/core/src/main/scripts/migrate_db.py -y -p <(parse_db_params_from_config_and_command_line $POTENTIAL_DB_PARAMS) -s /cbioportal/db-scripts/src/main/resources/migration.sql
+    fi
+}
+
+_main() {
+    # when running the webapp, check db and do migration first
+    # check if command is something like "java -jar webapp-runner.jar"
+    if [[ "$@" == *java* ]] && [[ "$@" == *-jar* ]] && [[ "$@" == *webapp-runner.jar* ]]; then
+        # Parse database config. Use command line parameters (e.g. -Ddb.host) if
+        # available, otherwise use portal.properties
+        if [ -n "$SHOW_DEBUG_INFO" ] && [ "$SHOW_DEBUG_INFO" != "false" ]; then
+            echo "Using database config:"
+            parse_db_params_from_config_and_command_line $@
+        fi
+
+        check_db_connection $@
+        migrate_db $@
+
+        if [ -n "$SHOW_DEBUG_INFO" ] && [ "$SHOW_DEBUG_INFO" != "false" ]; then
+            echo Running: "$@"
+        fi
+    fi
+    exec "$@"
+}
+
+# If we are sourced from elsewhere, don't perform any further actions
+if ! _is_sourced; then
+    _main "$@"
+fi