Filter entities by name when extracting data from Arkindex

Closes #134 (closed)

dan/datasets/extract/__init__.py

@@ -40,6 +40,15 @@ def validate_probability(proba):
     return proba
 
 
+def validate_char(char):
+    if len(char) != 1:
+        raise argparse.ArgumentTypeError(
+            f"`{char}` (of length {len(char)}) is not a valid character. Must be a string of length 1."
+        )
+
+    return char
+
+
 def add_extract_parser(subcommands) -> None:
     parser = subcommands.add_parser(
         "extract",
@@ -87,14 +96,9 @@ def add_extract_parser(subcommands) -> None:
         action="store_true",
         help="Extract text with their entities.",
     )
-    parser.add_argument(
-        "--allow-unknown-entities",
-        action="store_true",
-        help="Ignore entities that do not appear in the list of tokens.",
-    )
     parser.add_argument(
         "--entity-separators",
-        type=str,
+        type=validate_char,
         nargs="+",
         help="Removes all text that does not appear in an entity or in the list of given characters. Do not give any arguments for keeping the whole text.",
         required=False,