Merge pull request #962 from jrdnbradford/validate-config-yaml

Validate tljh specific config

.github/workflows/unit-test.yaml

@@ -53,12 +53,13 @@ jobs:
         with:
           python-version: "${{ matrix.python_version }}"
 
-      - name: Install venv, git and setup venv
+      - name: Install venv, git, pip and setup venv
         run: |
           export DEBIAN_FRONTEND=noninteractive
           apt-get update
           apt-get install --yes \
               python3-venv \
+              python3-pip \
               bzip2 \
               git
 

integration-tests/Dockerfile

@@ -1,5 +1,5 @@
 # Systemd inside a Docker container, for CI only
-ARG BASE_IMAGE=ubuntu:20.04
+ARG BASE_IMAGE=ubuntu:22.04
 FROM $BASE_IMAGE
 
 # DEBIAN_FRONTEND is set to avoid being asked for input and hang during build:

tljh/config.py

@@ -154,6 +154,26 @@ def remove_item_from_config(config, property_path, value):
     return config_copy
 
 
+def validate_config(config, validate):
+    """
+    Validate changes to the config with tljh-config against the schema
+    """
+    import jsonschema
+
+    from .config_schema import config_schema
+
+    try:
+        jsonschema.validate(instance=config, schema=config_schema)
+    except jsonschema.exceptions.ValidationError as e:
+        if validate:
+            print(
+                f"Config validation error: {e.message}.\n"
+                "You can still apply this change without validation by re-running your command with the --no-validate flag.\n"
+                "If you think this validation error is incorrect, please report it to https://github.com/jupyterhub/the-littlest-jupyterhub/issues."
+            )
+            exit()
+
+
 def show_config(config_path):
     """
     Pretty print config from given config_path
@@ -167,30 +187,29 @@ def show_config(config_path):
     yaml.dump(config, sys.stdout)
 
 
-def set_config_value(config_path, key_path, value):
+def set_config_value(config_path, key_path, value, validate=True):
     """
     Set key at key_path in config_path to value
     """
     # FIXME: Have a file lock here
-    # FIXME: Validate schema here
     try:
         with open(config_path) as f:
             config = yaml.load(f)
     except FileNotFoundError:
         config = {}
-
     config = set_item_in_config(config, key_path, value)
 
+    validate_config(config, validate)
+
     with open(config_path, "w") as f:
         yaml.dump(config, f)
 
 
-def unset_config_value(config_path, key_path):
+def unset_config_value(config_path, key_path, validate=True):
     """
     Unset key at key_path in config_path
     """
     # FIXME: Have a file lock here
-    # FIXME: Validate schema here
     try:
         with open(config_path) as f:
             config = yaml.load(f)
@@ -198,17 +217,17 @@ def unset_config_value(config_path, key_path):
         config = {}
 
     config = unset_item_from_config(config, key_path)
+    validate_config(config, validate)
 
     with open(config_path, "w") as f:
         yaml.dump(config, f)
 
 
-def add_config_value(config_path, key_path, value):
+def add_config_value(config_path, key_path, value, validate=True):
     """
     Add value to list at key_path
     """
     # FIXME: Have a file lock here
-    # FIXME: Validate schema here
     try:
         with open(config_path) as f:
             config = yaml.load(f)
@@ -216,17 +235,17 @@ def add_config_value(config_path, key_path, value):
         config = {}
 
     config = add_item_to_config(config, key_path, value)
+    validate_config(config, validate)
 
     with open(config_path, "w") as f:
         yaml.dump(config, f)
 
 
-def remove_config_value(config_path, key_path, value):
+def remove_config_value(config_path, key_path, value, validate=True):
     """
     Remove value from list at key_path
     """
     # FIXME: Have a file lock here
-    # FIXME: Validate schema here
     try:
         with open(config_path) as f:
             config = yaml.load(f)
@@ -234,6 +253,7 @@ def remove_config_value(config_path, key_path, value):
         config = {}
 
     config = remove_item_from_config(config, key_path, value)
+    validate_config(config, validate)
 
     with open(config_path, "w") as f:
         yaml.dump(config, f)
@@ -336,6 +356,18 @@ def main(argv=None):
     argparser.add_argument(
         "--config-path", default=CONFIG_FILE, help="Path to TLJH config.yaml file"
     )
+
+    argparser.add_argument(
+        "--validate", action="store_true", help="Validate the TLJH config"
+    )
+    argparser.add_argument(
+        "--no-validate",
+        dest="validate",
+        action="store_false",
+        help="Do not validate the TLJH config",
+    )
+    argparser.set_defaults(validate=True)
+
     subparsers = argparser.add_subparsers(dest="action")
 
     show_parser = subparsers.add_parser("show", help="Show current configuration")
@@ -383,13 +415,19 @@ def main(argv=None):
     if args.action == "show":
         show_config(args.config_path)
     elif args.action == "set":
-        set_config_value(args.config_path, args.key_path, parse_value(args.value))
+        set_config_value(
+            args.config_path, args.key_path, parse_value(args.value), args.validate
+        )
     elif args.action == "unset":
-        unset_config_value(args.config_path, args.key_path)
+        unset_config_value(args.config_path, args.key_path, args.validate)
     elif args.action == "add-item":
-        add_config_value(args.config_path, args.key_path, parse_value(args.value))
+        add_config_value(
+            args.config_path, args.key_path, parse_value(args.value), args.validate
+        )
     elif args.action == "remove-item":
-        remove_config_value(args.config_path, args.key_path, parse_value(args.value))
+        remove_config_value(
+            args.config_path, args.key_path, parse_value(args.value), args.validate
+        )
     elif args.action == "reload":
         reload_component(args.component)
     else:

tljh/config_schema.py

@@ -0,0 +1,117 @@
+"""
+The schema against which the TLJH config file can be validated.
+
+Validation occurs when changing values with tljh-config.
+"""
+
+config_schema = {
+    "$schema": "http://json-schema.org/draft-07/schema#",
+    "title": "Littlest JupyterHub YAML config file",
+    "definitions": {
+        "BaseURL": {
+            "type": "string",
+        },
+        "Users": {
+            "type": "object",
+            "additionalProperties": False,
+            "properties": {
+                "extra_user_groups": {"type": "object", "items": {"type": "string"}},
+                "allowed": {"type": "array", "items": {"type": "string"}},
+                "banned": {"type": "array", "items": {"type": "string"}},
+                "admin": {"type": "array", "items": {"type": "string"}},
+            },
+        },
+        "Services": {
+            "type": "object",
+            "properties": {
+                "cull": {
+                    "type": "object",
+                    "additionalProperties": False,
+                    "properties": {
+                        "enabled": {"type": "boolean"},
+                        "timeout": {"type": "integer"},
+                        "every": {"type": "integer"},
+                        "concurrency": {"type": "integer"},
+                        "users": {"type": "boolean"},
+                        "max_age": {"type": "integer"},
+                        "remove_named_servers": {"type": "boolean"},
+                    },
+                }
+            },
+        },
+        "HTTP": {
+            "type": "object",
+            "additionalProperties": False,
+            "properties": {
+                "address": {"type": "string", "format": "ipv4"},
+                "port": {"type": "integer"},
+            },
+        },
+        "HTTPS": {
+            "type": "object",
+            "additionalProperties": False,
+            "properties": {
+                "enabled": {"type": "boolean"},
+                "address": {"type": "string", "format": "ipv4"},
+                "port": {"type": "integer"},
+                "tls": {"$ref": "#/definitions/TLS"},
+                "letsencrypt": {"$ref": "#/definitions/LetsEncrypt"},
+            },
+        },
+        "LetsEncrypt": {
+            "type": "object",
+            "additionalProperties": False,
+            "properties": {
+                "email": {"type": "string", "format": "email"},
+                "domains": {
+                    "type": "array",
+                    "items": {"type": "string", "format": "hostname"},
+                },
+                "staging": {"type": "boolean"},
+            },
+        },
+        "TLS": {
+            "type": "object",
+            "additionalProperties": False,
+            "properties": {"key": {"type": "string"}, "cert": {"type": "string"}},
+        },
+        "Limits": {
+            "description": "User CPU and memory limits.",
+            "type": "object",
+            "additionalProperties": False,
+            "properties": {"memory": {"type": "string"}, "cpu": {"type": "integer"}},
+        },
+        "UserEnvironment": {
+            "type": "object",
+            "additionalProperties": False,
+            "properties": {
+                "default_app": {
+                    "type": "string",
+                    "enum": ["jupyterlab", "classic"],
+                    "default": "jupyterlab",
+                }
+            },
+        },
+        "TraefikAPI": {
+            "type": "object",
+            "additionalProperties": False,
+            "properties": {
+                "ip": {"type": "string", "format": "ipv4"},
+                "port": {"type": "integer"},
+                "username": {"type": "string"},
+                "password": {"type": "string"},
+            },
+        },
+    },
+    "properties": {
+        "additionalProperties": False,
+        "base_url": {"$ref": "#/definitions/BaseURL"},
+        "user_environment": {"$ref": "#/definitions/UserEnvironment"},
+        "users": {"$ref": "#/definitions/Users"},
+        "limits": {"$ref": "#/definitions/Limits"},
+        "https": {"$ref": "#/definitions/HTTPS"},
+        "http": {"$ref": "#/definitions/HTTP"},
+        "traefik_api": {"$ref": "#/definitions/TraefikAPI"},
+        "services": {"$ref": "#/definitions/Services"},
+    },
+}