[MRG] fix tax argument parsing

src/sourmash/cli/tax/annotate.py

@@ -23,7 +23,7 @@ def subparser(subparsers):
                                       aliases=['annotate'],
                                       usage=usage)
     subparser.add_argument(
-        '-g', '--gather-csv', nargs='*', default = [],
+        '-g', '--gather-csv', nargs='*', default = [], action='extend',
         help='CSV output files from sourmash gather'
     )
     subparser.add_argument(
@@ -36,7 +36,7 @@ def subparser(subparsers):
     )
     subparser.add_argument(
         '-t', '--taxonomy-csv', '--taxonomy', metavar='FILE',
-        nargs="+", required=True,
+        nargs='*', required=True, action="extend",
         help='database lineages CSV'
     )
     subparser.add_argument(

src/sourmash/cli/tax/genome.py

@@ -41,7 +41,7 @@ def subparser(subparsers):
                                       aliases=['classify'],
                                       usage=usage)
     subparser.add_argument(
-        '-g', '--gather-csv', nargs='*', default = [],
+        '-g', '--gather-csv', action='extend', nargs='*', default = [],
         help='CSVs output by sourmash gather for this sample'
     )
     subparser.add_argument(
@@ -54,7 +54,7 @@ def subparser(subparsers):
     )
     subparser.add_argument(
         '-t', '--taxonomy-csv', '--taxonomy', metavar='FILE',
-        nargs='+', required=True,
+        nargs='*', required=True, action='extend',
         help='database lineages CSV'
     )
     subparser.add_argument(
@@ -82,7 +82,8 @@ def subparser(subparsers):
         help='fail quickly if taxonomy is not available for an identifier',
     )
     subparser.add_argument(
-        '-F', '--output-format', default=['csv_summary'], nargs='+', choices=["csv_summary", "krona", "human", "lineage_csv"],
+        '-F', '--output-format', default=[], nargs='*', action='extend',
+        choices=["csv_summary", "krona", "human", "lineage_csv"],
         help='choose output format(s)',
     )
     subparser.add_argument(
@@ -102,4 +103,8 @@ def main(args):
     if not args.rank:
         if any(x in ["krona"] for x in args.output_format):
             raise ValueError(f"Rank (--rank) is required for krona output format.")
+    if not args.output_format:
+        # change to "human" for 5.0
+        args.output_format = ["csv_summary"]
+
     return sourmash.tax.__main__.genome(args)

src/sourmash/cli/tax/grep.py

@@ -55,7 +55,7 @@ def subparser(subparsers):
     )
     subparser.add_argument(
         '-t', '--taxonomy-csv', '--taxonomy', metavar='FILE',
-        nargs="+", required=True,
+        nargs="+", required=True, action="extend",
         help='database lineages'
     )
     subparser.add_argument(

src/sourmash/cli/tax/metagenome.py

@@ -30,7 +30,7 @@ def subparser(subparsers):
                                       aliases=['summarize'],
                                       usage=usage)
     subparser.add_argument(
-        '-g', '--gather-csv', nargs='*', default = [],
+        '-g', '--gather-csv', action="extend", nargs='*', default = [],
         help='CSVs from sourmash gather'
     )
     subparser.add_argument(
@@ -51,7 +51,7 @@ def subparser(subparsers):
     )
     subparser.add_argument(
         '-t', '--taxonomy-csv', '--taxonomy', metavar='FILE',
-        nargs='+', required=True,
+        action="extend", nargs='+', required=True,
         help='database lineages CSV'
     )
     subparser.add_argument(
@@ -67,7 +67,8 @@ def subparser(subparsers):
         help='fail quickly if taxonomy is not available for an identifier',
     )
     subparser.add_argument(
-        '-F', '--output-format', default=['csv_summary'], nargs='+', choices=["human", "csv_summary", "krona", "lineage_summary", "kreport"],
+        '-F', '--output-format', default=[], nargs='*', action="extend",
+        choices=["human", "csv_summary", "krona", "lineage_summary",  "kreport"],
         help='choose output format(s)',
     )
     subparser.add_argument(
@@ -89,4 +90,7 @@ def main(args):
     if not args.rank:
         if any(x in ["krona", "lineage_summary"] for x in args.output_format):
             raise ValueError(f"Rank (--rank) is required for krona and lineage_summary output formats.")
+    if not args.output_format:
+        # change to "human" for 5.0
+        args.output_format = ["csv_summary"]
     return sourmash.tax.__main__.metagenome(args)

src/sourmash/cli/tax/prepare.py

@@ -25,7 +25,7 @@ def subparser(subparsers):
     )
     subparser.add_argument(
         '-t', '--taxonomy-csv', '--taxonomy', metavar='FILE',
-        nargs="+", required=True,
+        nargs="+", required=True, action="extend",
         help='database lineages'
     )
     subparser.add_argument(

src/sourmash/tax/__main__.py

@@ -364,6 +364,7 @@ def prepare(args):
     notify("loading taxonomies...")
     try:
         tax_assign = MultiLineageDB.load(args.taxonomy_csv,
+                                         force=args.force,
                        keep_full_identifiers=args.keep_full_identifiers,
                        keep_identifier_versions=args.keep_identifier_versions)
     except ValueError as exc:

src/sourmash/tax/tax_utils.py

@@ -1076,6 +1076,8 @@ def _save_csv(self, fp):
     @classmethod
     def load(cls, locations, **kwargs):
         "Load one or more taxonomies from the given location(s)"
+        force = kwargs.get('force', False)
+
         if isinstance(locations, str):
             raise TypeError("'locations' should be a list, not a string")
 
@@ -1098,12 +1100,14 @@ def load(cls, locations, **kwargs):
                     loaded = True
                 except (ValueError, csv.Error) as exc:
                     # for the last loader, just pass along ValueError...
-                    raise ValueError(f"cannot read taxonomy assignments from '{location}': {str(exc)}")
+                    if not force:
+                        raise ValueError(f"cannot read taxonomy assignments from '{location}': {str(exc)}")
 
             # nothing loaded, goodbye!
-            if not loaded:
+            if not loaded and not force:
                 raise ValueError(f"cannot read taxonomy assignments from '{location}'")
 
-            tax_assign.add(this_tax_assign)
+            if loaded:
+                tax_assign.add(this_tax_assign)
 
         return tax_assign