Fix json loader - it should go in 'validators' not 'utils'

Fixes #214

docs/scripts/scripts.py

@@ -68,7 +68,7 @@ def generate_module_docs():
             config_yaml = {}
             for key, value in manifest['configs'].items():
                 type = value.get('type', 'string')
-                if type == 'auto_archiver.utils.json_loader':
+                if type == 'json_loader':
                     value['type'] = 'json'
                 elif type == 'str':
                     type = "string"

src/auto_archiver/core/config.py

@@ -208,7 +208,7 @@ def read_yaml(yaml_filename: str) -> CommentedMap:
         pass
 
     if not config:
-        config = EMPTY_CONFIG
+        config = deepcopy(EMPTY_CONFIG)
 
     return config
 

src/auto_archiver/core/validators.py

@@ -1,6 +1,7 @@
 # used as validators for config values. Should raise an exception if the value is invalid.
 from pathlib import Path
 import argparse
+import json
 
 def example_validator(value):
     if "example" not in value:
@@ -17,3 +18,6 @@ def valid_file(value):
     if not Path(value).is_file():
         raise argparse.ArgumentTypeError(f"File '{value}' does not exist.")
     return value
+
+def json_loader(cli_val):
+    return json.loads(cli_val)

src/auto_archiver/modules/gsheet_feeder/__manifest__.py

@@ -35,7 +35,7 @@
                 "replaywebpage": "replaywebpage",
             },
             "help": "names of columns in the google sheet (stringified JSON object)",
-            "type": "auto_archiver.utils.json_loader",
+            "type": "json_loader",
         },
         "allow_worksheets": {
             "default": set(),

src/auto_archiver/modules/gsheet_feeder/gsheet_feeder.py

@@ -24,9 +24,8 @@ class GsheetsFeeder(Feeder):
     def setup(self) -> None:
         self.gsheets_client = gspread.service_account(filename=self.service_account)
         # TODO mv to validators
-        assert self.sheet or self.sheet_id, (
-            "You need to define either a 'sheet' name or a 'sheet_id' in your manifest."
-        )
+        if not (self.sheet or self.sheet_id):
+            raise ValueError("You need to define either a 'sheet' name or a 'sheet_id' in your manifest.")
 
     def open_sheet(self):
         if self.sheet:

src/auto_archiver/modules/telethon_extractor/__manifest__.py

@@ -18,7 +18,7 @@
             "channel_invites": {
                 "default": {},
                 "help": "(JSON string) private channel invite links (format: t.me/joinchat/HASH OR t.me/+HASH) and (optional but important to avoid hanging for minutes on startup) channel id (format: CHANNEL_ID taken from a post url like https://t.me/c/CHANNEL_ID/1), the telegram account will join any new channels on setup",
-                "type": "auto_archiver.utils.json_loader",
+                "type": "json_loader",
             }
         },
     "description": """

src/auto_archiver/utils/misc.py

@@ -59,10 +59,6 @@ def random_str(length: int = 32) -> str:
     return str(uuid.uuid4()).replace("-", "")[:length]
 
 
-def json_loader(cli_val):
-    return json.loads(cli_val)
-
-
 def calculate_file_hash(filename: str, hash_algo = hashlib.sha256, chunksize: int = 16000000) -> str:
     hash = hash_algo()
     with open(filename, "rb") as f: