🚧 fix action='extend' behavior with defaults

augur/argparse_.py

@@ -18,6 +18,12 @@
 SKIP_AUTO_DEFAULT_IN_HELP = "%(default).0s"
 
 
+# We use nargs='+' with action='extend' for a more intuitive user experience,
+# but that prevents any default from being overridden. Allow overriding defaults
+# by deferring defaults to a later point in the code.
+DEFER_LIST_DEFAULT = []
+
+
 def add_default_command(parser):
     """
     Sets the default command to run when none is provided.

augur/filter/__init__.py

@@ -1,9 +1,10 @@
 """
 Filter and subsample a sequence set.
 """
+from augur.argparse_ import DEFER_LIST_DEFAULT
 from augur.dates import numeric_date_type, SUPPORTED_DATE_HELP_TEXT
 from augur.filter.io import ACCEPTED_TYPES, column_type_pair
-from augur.io.metadata import DEFAULT_DELIMITERS, DEFAULT_ID_COLUMNS, METADATA_DATE_COLUMN
+from augur.io.metadata import METADATA_DATE_COLUMN
 from augur.types import EmptyOutputReportingMethod
 from . import constants
 
@@ -19,8 +20,8 @@ def register_arguments(parser):
     input_group.add_argument('--sequences', '-s', help="sequences in FASTA or VCF format")
     input_group.add_argument('--sequence-index', help="sequence composition report generated by augur index. If not provided, an index will be created on the fly.")
     input_group.add_argument('--metadata-chunk-size', type=int, default=100000, help="maximum number of metadata records to read into memory at a time. Increasing this number can speed up filtering at the cost of more memory used.")
-    input_group.add_argument('--metadata-id-columns', default=DEFAULT_ID_COLUMNS, nargs="+", action="extend", help="names of possible metadata columns containing identifier information, ordered by priority. Only one ID column will be inferred.")
-    input_group.add_argument('--metadata-delimiters', default=DEFAULT_DELIMITERS, nargs="+", action="extend", help="delimiters to accept when reading a metadata file. Only one delimiter will be inferred.")
+    input_group.add_argument('--metadata-id-columns', default=DEFER_LIST_DEFAULT, nargs="+", action="extend", help="names of possible metadata columns containing identifier information, ordered by priority. Only one ID column will be inferred.")
+    input_group.add_argument('--metadata-delimiters', default=DEFER_LIST_DEFAULT, nargs="+", action="extend", help="delimiters to accept when reading a metadata file. Only one delimiter will be inferred.")
 
     metadata_filter_group = parser.add_argument_group("metadata filters", "filters to apply to metadata")
     metadata_filter_group.add_argument(
@@ -102,7 +103,7 @@ def run(args):
     '''
     from .validate_arguments import validate_arguments
     # Validate arguments before attempting any I/O.
-    validate_arguments(args)
+    args = validate_arguments(args)
 
     from ._run import run as _run
     return _run(args)

augur/filter/validate_arguments.py

@@ -1,4 +1,6 @@
+from augur.argparse_ import DEFER_LIST_DEFAULT
 from augur.errors import AugurError
+from augur.io.metadata import DEFAULT_DELIMITERS, DEFAULT_ID_COLUMNS
 from augur.io.vcf import is_vcf as filename_is_vcf
 
 
@@ -43,3 +45,14 @@ def validate_arguments(args):
     # If user requested grouping, confirm that other required inputs are provided, too.
     if args.group_by and not any((args.sequences_per_group, args.subsample_max_sequences)):
         raise AugurError("You must specify a number of sequences per group or maximum sequences to subsample.")
+
+
+    # Set defaults
+
+    if args.metadata_delimiters is DEFER_LIST_DEFAULT:
+        args.metadata_delimiters = DEFAULT_DELIMITERS
+
+    if args.metadata_id_columns is DEFER_LIST_DEFAULT:
+        args.metadata_id_columns = DEFAULT_ID_COLUMNS
+
+    return args