Adding anonymizeSnippets in reports config (#569)

* Adding anonymizeSnippets in reports config

* Updated config processing logic, it is now using pydantic models and validation.

* Updated docs

docs/gh_pages/docs/config.md

@@ -34,12 +34,13 @@ Notes:
   ![Pebblo Reports](../static/img/report-comparision.png)
 
 - `cacheDir`: Sets the directory where pebblo stores metadata, generated reports, and other temporary files. Default value is `~/.pebblo`.
+- `anonymizeSnippets`: Flag to anonymize snippets in report and Pebblo local UI. Possible values are 'True' and 'False'. When its value is 'True', snippets in reports and Pebblo local UI will be shown as anonymized and vice versa.
 - `outputDir`: Deprecated. Use `cacheDir` instead.
 
 ### Classifier
 
 - `mode`: Specifies mode for classify API. Possible values are `all`, `entity` or `topic`. Default value is `all`. When its value is `all`, both entities and topics will get classified, if value is `entity`, only entities will get classified and vice-versa. It is used for classification in /classify and /loader/doc APIs.
-- `anonymizeSnippets`: Flag to anonymize snippets in report. Possible values are 'True' and 'False'. When its value is 'True', snippets in reports will be shown as anonymized and vice versa.
+- `anonymizeSnippets` is deprecated, use 'anonymizeSnippets' in reports instead.
 
 ### Storage
 This is beta feature introduced in 0.1.18.

pebblo/app/config/config.py

@@ -1,114 +1,60 @@
-import pathlib
+import sys
 from contextvars import ContextVar
-from typing import Optional, Tuple, Union
+from typing import Optional, Tuple
 
 import yaml
-from pydantic import Field
-from pydantic_settings import BaseSettings
-
-from pebblo.app.config.config_validation import validate_config, validate_input
-from pebblo.app.enums.common import ClassificationMode, DBStorageTypes, StorageTypes
-
-# Default config value
-dir_path = pathlib.Path().absolute()
-
-
-# Port BaseModel
-class PortConfig(BaseSettings):
-    host: str = Field(default="localhost")
-    port: int = Field(default=8000)
-
-
-# Report BaseModel
-class ReportConfig(BaseSettings):
-    format: str = Field(default="pdf")
-    renderer: str = Field(default="xhtml2pdf")
-    cacheDir: str = Field(default=str(dir_path))
-
-
-# Logging Defaults
-DEFAULT_LOGGER_NAME = "pebblo"
-DEFAULT_LOG_MAX_FILE_SIZE = 8 * 1024 * 1024
-DEFAULT_LOG_BACKUP_COUNT = 3
-DEFAULT_LOG_LEVEL = "INFO"
-DEFAULT_LOG_FILE_PATH = "/tmp/logs"
-DEFAULT_LOG_FILE = f"{DEFAULT_LOG_FILE_PATH}/{DEFAULT_LOGGER_NAME}.log"
-
-
-# Logging BaseModel
-class LoggingConfig(BaseSettings):
-    level: str = Field(default=DEFAULT_LOG_LEVEL)
-    file: str = Field(default=DEFAULT_LOG_FILE)
-    maxFileSize: int = Field(default=DEFAULT_LOG_MAX_FILE_SIZE)
-    backupCount: int = Field(default=DEFAULT_LOG_BACKUP_COUNT)
-
-
-class ClassifierConfig(BaseSettings):
-    mode: str = Field(default=ClassificationMode.ALL.value)
-    anonymizeSnippets: bool = Field(default=True)
-
-
-class StorageConfig(BaseSettings):
-    type: str = Field(default=StorageTypes.FILE.value)
-    db: Union[str, None] = Field(default=DBStorageTypes.SQLITE.value)
-    location: Union[str, None] = Field(default=str(dir_path))
-    name: Union[str, None] = Field(default=str("pebblo"))
-    # This is default value for current version(0.1.18), it needs to be changed in next version to db.
-
-
-# ConfigFile BaseModel
-class Config(BaseSettings):
-    daemon: PortConfig
-    reports: ReportConfig
-    logging: LoggingConfig
-    classifier: ClassifierConfig
-    storage: StorageConfig
 
+from pebblo.app.config.models import (
+    ClassifierConfig,
+    Config,
+    DaemonConfig,
+    LoggingConfig,
+    ReportConfig,
+    StorageConfig,
+)
+from pebblo.app.enums.common import ClassificationMode
 
 var_server_config: ContextVar[Config] = ContextVar("server_config", default=None)
 var_server_config_dict: ContextVar[dict] = ContextVar("server_config_dict", default={})
 
 
+def get_default_config_values():
+    # set default config value
+    conf_obj = Config(
+        daemon=DaemonConfig(host="localhost", port=8000),
+        reports=ReportConfig(format="pdf", renderer="xhtml2pdf", cacheDir="~/.pebblo"),
+        logging=LoggingConfig(),
+        classifier=ClassifierConfig(
+            mode=ClassificationMode.ALL.value, anonymizeSnippets=False
+        ),
+        storage=StorageConfig(type="file", db=None),
+        # for now, a default storage type is FILE, but in the next release DB will be the default storage type.
+    )
+    return conf_obj.dict(), conf_obj
+
+
 def load_config(path: Optional[str]) -> Tuple[dict, Config]:
     try:
-        # If Path does not exist in command, set default config value
-        conf_obj = Config(
-            daemon=PortConfig(host="localhost", port=8000),
-            reports=ReportConfig(
-                format="pdf", renderer="xhtml2pdf", cacheDir="~/.pebblo"
-            ),
-            logging=LoggingConfig(),
-            classifier=ClassifierConfig(
-                mode=ClassificationMode.ALL.value, anonymizeSnippets=False
-            ),
-            storage=StorageConfig(type="file", db=None),
-            # for now, a default storage type is FILE, but in the next release DB will be the default storage type.
-        )
         if not path:
-            # Setting Default config details
-            return conf_obj.dict(), conf_obj
+            # If Path does not exist in command, set default config value
+            get_default_config_values()
 
         # If Path exist, set config value
         try:
             with open(path, "r") as output:
                 cred_yaml = yaml.safe_load(output)
-                cred_yaml = validate_input(cred_yaml)
-
-                # Replace missing fields with default values
-                for key in conf_obj.dict().keys():
-                    if key not in cred_yaml:
-                        cred_yaml[key] = conf_obj.dict()[key]
                 parsed_config = Config.parse_obj(cred_yaml)
                 config_dict = parsed_config.dict()
                 config_dict["logging"]["level"] = (
                     config_dict.get("logging").get("level").upper()
                 )
-                validate_config(config_dict)
                 return config_dict, parsed_config
         except IOError as err:
             print(f"no config file found at {path}. Error : {err}")
-            return conf_obj.dict(), conf_obj
+            return get_default_config_values()
 
     except Exception as err:
         print(f"Error while loading config details, err: {err}")
+        print("Exiting due to validation error...")
+        sys.exit()
         return {}, {}