From b9b8186e68fe5c9be425ef301c0f7d709adcb285 Mon Sep 17 00:00:00 2001
From: Brandon Thomas <echelon@gmail.com>
Date: Fri, 24 Jun 2022 22:00:48 -0400
Subject: [PATCH] Utility to delete old tts_inference_jobs records

---
 .gitignore                         |  4 ++
 tools/delete_tts_inference_jobs.py | 61 ++++++++++++++++++++++++++++++
 2 files changed, 65 insertions(+)
 create mode 100755 tools/delete_tts_inference_jobs.py

diff --git a/.gitignore b/.gitignore
index d761e8f511..3cd297751e 100644
--- a/.gitignore
+++ b/.gitignore
@@ -7,6 +7,8 @@
 .idea/
 __pycache__/
 _ignored_schema.rs
+db.secrets
+db_secrets.toml
 models/Wav2Lip/python/
 models/tacotron2/python/
 production_database_secret_connect.sh
@@ -15,4 +17,6 @@ production_database_voice_clone_report_download.sh
 secrets.toml
 secrets.yaml
 target/
+tools/.secrets
+tools/db_secrets.toml
 twitch_secrets.toml
diff --git a/tools/delete_tts_inference_jobs.py b/tools/delete_tts_inference_jobs.py
new file mode 100755
index 0000000000..b17a112f16
--- /dev/null
+++ b/tools/delete_tts_inference_jobs.py
@@ -0,0 +1,61 @@
+#!/usr/bin/env python
+
+# FIXME: Needs packages 'toml' and 'mysql-connector-python'. Put in requirements.txt.
+
+import time
+import toml
+import mysql.connector
+
+DB_SECRETS_FILE = 'db_secrets.toml'
+
+# TODO: Share db-related setup when more scripts are added.
+
+def read_database_secrets(secrets_file = DB_SECRETS_FILE):
+    secrets = None
+    with open(secrets_file, "r") as f:
+        secrets = toml.load(f)
+
+    required_keys = ['database', 'hostname', 'port', 'password', 'username']
+    for key in required_keys:
+        if key not in secrets:
+            raise Exception(f"Required key '#{key}' not present")
+
+    return secrets
+
+def connect_to_mysql():
+    secrets = read_database_secrets()
+    connection = mysql.connector.connect(
+        host = secrets['hostname'],
+        port = secrets['port'],
+        user = secrets['username'],
+        password = secrets['password'],
+        database = secrets['database'],
+    )
+    return connection
+
+connection = connect_to_mysql()
+cursor = connection.cursor()
+
+SAFE_ID = 76000000 # Reasonably completed on 2022-06-24
+LIMIT = 100000
+
+total_deleted = 0
+
+keep_deleting = True
+
+while keep_deleting:
+    query = f"DELETE FROM tts_inference_jobs WHERE id < {SAFE_ID} ORDER BY ID desc LIMIT {LIMIT}"
+
+    cursor.execute(query)
+
+    #result = cursor.fetchall()
+    count = cursor.rowcount
+    total_deleted += count
+
+    print(f'Deleted {count}; total deleted = {total_deleted}')
+
+    if count < LIMIT - 1:
+        keep_deleting = False
+
+    time.sleep(1)
+