llm_server/routes/helpers/client.py

@@ -2,8 +2,11 @@ from llm_server.cluster.cluster_config import cluster_config
 from llm_server.custom_redis import redis
 
 
-def format_sillytavern_err(msg: str, backend_url: str = 'none', error_type: str = 'info'):
+def format_sillytavern_err(msg: str, backend_url: str = None, error_type: str = 'info'):
-    cluster_backend_hash = cluster_config.get_backend(backend_url)['hash']
+    if backend_url:
+        cluster_backend_hash = cluster_config.get_backend(backend_url)['hash']
+    else:
+        cluster_backend_hash = 'none'
     http_host = redis.get('http_host', dtype=str)
     return f"""```
 === MESSAGE FROM LLM MIDDLEWARE AT {http_host} ===

llm_server/routes/v1/__init__.py

@@ -3,15 +3,16 @@ from flask import Blueprint, jsonify
 from llm_server.custom_redis import redis
 from llm_server.routes.helpers.client import format_sillytavern_err
 
-bp = Blueprint('old_v1', __name__)
+old_v1_bp = Blueprint('v1', __name__)
 
 
-@bp.route('/<first>')
+@old_v1_bp.route('/', defaults={'path': ''}, methods=['GET', 'POST'])
-@bp.route('/<first>/<path:rest>')
+@old_v1_bp.route('/<path:path>', methods=['GET', 'POST'])
-def fallback(first=None, rest=None):
+def fallback(path):
     base_client_api = redis.get('base_client_api', dtype=str)
-    error_msg = f'The /v1/ endpoint has been depreciated. Please see [{base_client_api}]({base_client_api}) for the new endpoint.'
+    error_msg = f'The /v1/ endpoint has been depreciated. Please visit {base_client_api} for more information.'
     response_msg = format_sillytavern_err(error_msg, error_type='API')
     return jsonify({
-        'results': [{'text': response_msg}]
+        'results': [{'text': response_msg}],
+        'result': base_client_api
     }), 200  # return 200 so we don't trigger an error message in the client's ST

llm_server/routes/v2/__init__.py

@@ -3,7 +3,7 @@ from flask import Blueprint
 from ..request_handler import before_request
 from ..server_error import handle_server_error
 
-bp = Blueprint('v1', __name__)
+bp = Blueprint('v2', __name__)
 
 
 @bp.before_request

server.py

@@ -1,3 +1,5 @@
+from llm_server.routes.v1 import old_v1_bp
+
 try:
     import gevent.monkey
 
@@ -70,6 +72,7 @@ from llm_server.routes.v2.generate_stats import generate_stats
 app = Flask(__name__)
 init_socketio(app)
 app.register_blueprint(bp, url_prefix='/api/v2/')
+app.register_blueprint(old_v1_bp, url_prefix='/api/v1/')
 app.register_blueprint(openai_bp, url_prefix='/api/openai/v1/')
 flask_cache.init_app(app)
 flask_cache.clear()