fix install bugs

Signed-off-by: Sijie Shen <[email protected]>

scripts/gart

@@ -87,7 +87,7 @@ while [ : ]; do
         shift 2
         ;;
          --db-type)
-        db_type=$2
+        db_type=$(echo "$2" | awk '{print tolower($0)}')
         shift 2
         ;;
         --enable-bulkload)

scripts/install-mysql.sh

@@ -18,4 +18,9 @@ binlog-do-db=ldbc  # change the name to your database
 
 EOT
 
+# For some reason, the mysql user's home directory is not set correctly
+# see https://stackoverflow.com/questions/62987154/mysql-wont-start-error-su-warning-cannot-change-directory-to-nonexistent
+sudo usermod -d /var/lib/mysql/ mysql
+
+# Restart mysql to apply the changes
 sudo service mysql restart

scripts/update_kafka_config_file.py

@@ -2,11 +2,13 @@
 
 import argparse
 import os
-import etcd3
+import sys
+
 from urllib.parse import urlparse
-import yaml
 import shutil
-import sys
+
+import etcd3
+import yaml
 
 
 def get_parser():
@@ -28,6 +30,16 @@ def get_parser():
     return parser
 
 
+def get_etcd_client(etcd_endpoint):
+    if not etcd_endpoint.startswith(("http://", "https://")):
+        etcd_endpoint = "http://" + etcd_endpoint
+
+    parsed_url = urlparse(etcd_endpoint)
+    etcd_host = parsed_url.netloc.split(":")[0]
+    etcd_port = parsed_url.port
+    return etcd3.client(host=etcd_host, port=etcd_port)
+
+
 if __name__ == "__main__":
     arg_parser = get_parser()
     args = arg_parser.parse_args()
@@ -38,7 +50,7 @@ def get_parser():
         print("KAFKA_HOME is not set")
         sys.exit(1)
 
-    db_type = args.db_type
+    db_type = args.db_type.lower()
 
     kafka_server = args.kafka_endpoint
     kafka_port = kafka_server.split(":")[1]
@@ -60,10 +72,17 @@ def get_parser():
             kafka_home + "/config/connect-debezium-postgresql.properties.tmp"
         )
 
-    etcd_endpoint = args.etcd_endpoint
+    etcd_client = get_etcd_client(args.etcd_endpoint)
+
+    enable_bulkload = (
+        args.enable_bulkload == "1"
+        or args.enable_bulkload == 1
+        or args.enable_bulkload is True
+        or args.enable_bulkload.lower() == "true"
+    )
 
-    with open(kafka_config_file_name, "r") as file, open(
-        temp_file_name, "w"
+    with open(kafka_config_file_name, "r", encoding="UTF-8") as file, open(
+        temp_file_name, "w", encoding="UTF-8"
     ) as temp_file:
         for line in file:
             if line.startswith("database.hostname"):
@@ -79,14 +98,15 @@ def get_parser():
             elif line.startswith("database.include.list"):
                 temp_file.write(f"database.include.list={args.db_name}\n")
             elif line.startswith("table.include.list"):
-                if not etcd_endpoint.startswith(("http://", "https://")):
-                    etcd_endpoint = "http://" + etcd_endpoint
-                parsed_url = urlparse(etcd_endpoint)
-                etcd_host = parsed_url.netloc.split(":")[0]
-                etcd_port = parsed_url.port
-                etcd_client = etcd3.client(host=etcd_host, port=etcd_port)
                 rg_mapping_key = args.etcd_prefix + "gart_rg_mapping_yaml"
-                rg_mapping_str = etcd_client.get(rg_mapping_key)[0].decode("utf-8")
+
+                raw_value = etcd_client.get(rg_mapping_key)
+                if raw_value is None or raw_value[0] is None:
+                    print(f"ERROR: Key {rg_mapping_key} not found in etcd")
+                    exit(1)
+                else:
+                    rg_mapping_str = raw_value[0].decode("utf-8")
+
                 graph_schema = yaml.load(rg_mapping_str, Loader=yaml.SafeLoader)
 
                 # Extract the 'vertex_types' list from the dictionary
@@ -129,8 +149,7 @@ def get_parser():
                 # build a dict (table_name, src/dst_type_names)
                 edge_table_src_dst_type_mapping = {}
 
-                for idx in range(len(edge_table_names)):
-                    edge_table_name = edge_table_names[idx]
+                for idx, edge_table_name in enumerate(edge_table_names):
                     if edge_table_name not in both_vertex_edge_table_names:
                         continue
 
@@ -161,10 +180,9 @@ def get_parser():
                         both_vertex_edge_table_names
                     ):
                         break
-                    for idx in range(len(both_vertex_edge_table_names)):
+                    for idx, edge_table_name in enumerate(both_vertex_edge_table_names):
                         if both_vertex_edge_table_placed[idx] == 1:
                             continue
-                        edge_table_name = both_vertex_edge_table_names[idx]
                         src_dst_type_names = edge_table_src_dst_type_mapping[
                             edge_table_name
                         ]
@@ -189,13 +207,7 @@ def get_parser():
                 new_line = ",".join(new_line_list)
                 temp_file.write(f"table.include.list={new_line}\n")
             elif line.startswith("snapshot.mode"):
-                if (
-                    args.enable_bulkload == "1"
-                    or args.enable_bulkload == 1
-                    or args.enable_bulkload == True
-                    or args.enable_bulkload == "True"
-                    or args.enable_bulkload == "true"
-                ):
+                if enable_bulkload:
                     if db_type == "postgresql":
                         temp_file.write("snapshot.mode=always\n")
                     else: