From 089f3cd8653479275787c60ade121cf07e15e474 Mon Sep 17 00:00:00 2001
From: Sebastian Urchs <surchs@users.noreply.github.com>
Date: Thu, 16 May 2024 12:05:38 -0400
Subject: [PATCH] [FIX] Enable graph setup to run on stack restart (#57)

---
 docker-compose.yml                        |  8 +++---
 docs/neurobagel_environment_variables.tsv |  2 --
 scripts/setup.sh                          | 33 ++++++++++++++++-------
 3 files changed, 29 insertions(+), 14 deletions(-)

diff --git a/docker-compose.yml b/docker-compose.yml
index 5ea4f1d..7071220 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -32,7 +32,7 @@ services:
       - "local_node_query"
       - "full_stack"
     volumes:
-      - "${NB_GRAPH_ROOT_HOST:-~/graphdb-home}:${NB_GRAPH_ROOT_CONT:-/opt/graphdb/home}"
+      - "graphdb_home:/opt/graphdb/home"
       - "./scripts:/usr/src/neurobagel/scripts"
       - ".env:/usr/src/neurobagel/.env"
       - "./vocab:/usr/src/neurobagel/vocab"
@@ -40,7 +40,6 @@ services:
     ports:
       - "${NB_GRAPH_PORT_HOST:-7200}:${NB_GRAPH_PORT:-7200}"
     environment:
-      NB_GRAPH_ROOT_CONT: ${NB_GRAPH_ROOT_CONT:-/opt/graphdb/home}
       NB_GRAPH_USERNAME: ${NB_GRAPH_USERNAME}
       NB_GRAPH_ADDRESS: ${NB_GRAPH_ADDRESS:-graph}
       NB_GRAPH_PORT: ${NB_GRAPH_PORT:-7200}
@@ -89,4 +88,7 @@ secrets:
   db_admin_password:
     environment: "NB_GRAPH_ADMIN_PASSWORD"
   db_user_password:
-    environment: "NB_GRAPH_PASSWORD"
\ No newline at end of file
+    environment: "NB_GRAPH_PASSWORD"
+
+volumes:
+  graphdb_home:
\ No newline at end of file
diff --git a/docs/neurobagel_environment_variables.tsv b/docs/neurobagel_environment_variables.tsv
index 201db03..3f957c6 100644
--- a/docs/neurobagel_environment_variables.tsv
+++ b/docs/neurobagel_environment_variables.tsv
@@ -15,8 +15,6 @@ Environment variable	Set manually in .env?	Description	Default value if not set
 `NB_FAPI_PORT`	No	Port number on which to run the Neurobagel federation API _in the API container_	`8000`	Docker
 `NB_GRAPH_IMG`	No	Graph server Docker image	`ontotext/graphdb:10.3.1`	Docker
 `NB_GRAPH_ADDRESS`	No	IP address for the graph database (or container name, if graph is hosted locally)	`206.12.99.17 (graph)` **	Docker, Python
-`NB_GRAPH_ROOT_HOST`	No	Path to directory on the _host machine_ to store graph database files and data (directory does not have to exist beforehand).	`~/graphdb-home`	Docker
-`NB_GRAPH_ROOT_CONT`	No	Path to directory for graph databases in the _graph server container_	`/opt/graphdb/home` *	Docker
 `NB_GRAPH_PORT_HOST`	No	Port number on the _host machine_ to map the graph server container port to	`7200`	Docker
 `NB_GRAPH_PORT`	No	Port number used by the _graph server container_	`7200` *	Docker, Python
 `NB_QUERY_TAG`	No	Docker image tag for the query tool	`latest`	Docker
diff --git a/scripts/setup.sh b/scripts/setup.sh
index 47d66cf..18cca8f 100755
--- a/scripts/setup.sh
+++ b/scripts/setup.sh
@@ -1,31 +1,46 @@
 #!/bin/bash
 
-/opt/graphdb/dist/bin/graphdb -Dgraphdb.home=${NB_GRAPH_ROOT_CONT} &
+/opt/graphdb/dist/bin/graphdb -Dgraphdb.home=/opt/graphdb/home &
 GRAPHDB_PID=$!
 
+# TODO revisit/test this also once we document how users can change (in addition to the data files being uploaded) the variables to set up a non-tester database after a first-time deployment
+export NB_GRAPH_ADMIN_PASSWORD=$(cat /run/secrets/db_admin_password)
+export NB_GRAPH_PASSWORD=$(cat /run/secrets/db_user_password)
+
 # Waiting for GraphDB to start
-while ! curl --silent "localhost:${NB_GRAPH_PORT}/rest/repositories" | grep '\[\]'; do
+while ! curl --silent "localhost:${NB_GRAPH_PORT}/rest/repositories" -u "${NB_GRAPH_USERNAME}:${NB_GRAPH_PASSWORD}" | grep '\['; do
     :
 done
 
+# We need to figure out if this is the first time the setup has been run
+repo_response=$(curl --silent "localhost:${NB_GRAPH_PORT}/rest/repositories" -u "${NB_GRAPH_USERNAME}:${NB_GRAPH_PASSWORD}")
+if [ "${repo_response}" = "[]" ]; then
+    export FIRST_TIME_SETUP="on"
+else
+    export FIRST_TIME_SETUP="off"
+fi
+
+echo "First time setup: ${FIRST_TIME_SETUP}"
+
 # TODO: Do we also want to use this elsewhere in the script or stick to ./<some_path>?
 SCRIPT_DIR=$(dirname "$0")
 mkdir -p ${SCRIPT_DIR}/logs
 
-export NB_GRAPH_ADMIN_PASSWORD=$(cat /run/secrets/db_admin_password)
-export NB_GRAPH_PASSWORD=$(cat /run/secrets/db_user_password)
-
 # Logic for main setup
 main() {
     echo "Setting up a Neurobagel graph backend..."
     echo -e "(The GraphDB server is being accessed inside the GraphDB container at http://localhost:${NB_GRAPH_PORT}.)\n"
 
-    echo "Setting up GraphDB server..."
-    ./graphdb_setup.sh --env-file-path /usr/src/neurobagel/.env "${NB_GRAPH_ADMIN_PASSWORD}"
-    echo "Finished server setup."
+    if [ "${FIRST_TIME_SETUP}" = "on" ]; then
+        echo "Setting up GraphDB server..."
+        ./graphdb_setup.sh --env-file-path /usr/src/neurobagel/.env "${NB_GRAPH_ADMIN_PASSWORD}"
+        echo "Finished server setup."export FIRST_TIME_SETUP="on"
+    else
+        echo "GraphDB server already set up, skipping setup."
+    fi
 
     echo "Adding datasets to the database..."
-    ./add_data_to_graph.sh /data localhost:${NB_GRAPH_PORT} ${NB_GRAPH_DB} "${NB_GRAPH_USERNAME}" "${NB_GRAPH_PASSWORD}"
+    ./add_data_to_graph.sh /data localhost:${NB_GRAPH_PORT} ${NB_GRAPH_DB} "${NB_GRAPH_USERNAME}" "${NB_GRAPH_PASSWORD}" --clear-data
     echo "Finished adding datasets to databases."
 
     echo "Adding Neurobagel vocabulary to the database"