make sure batch_size is not negative

src/python/sourmash_plugin_directsketch/__init__.py

@@ -4,6 +4,7 @@
 from sourmash.logging import notify
 from sourmash.plugins import CommandLinePlugin
 import importlib.metadata
+import argparse
 
 from . import sourmash_plugin_directsketch
 
@@ -32,6 +33,12 @@ def set_thread_pool(user_cores):
     actual_tokio_cores = sourmash_plugin_directsketch.set_tokio_thread_pool(num_threads)
     return actual_tokio_cores
 
+def non_negative_int(value):
+    ivalue = int(value)
+    if ivalue < 0:
+        raise argparse.ArgumentTypeError(f"Batch size cannot be negative (input value: {value})")
+    return ivalue
+
 class Download_and_Sketch_Assemblies(CommandLinePlugin):
     command = 'gbsketch'
     description = 'download and sketch GenBank assembly datasets'
@@ -43,7 +50,7 @@ def __init__(self, p):
                        help='output zip file for the signatures')
         p.add_argument('-f', '--fastas',
                        help='Write fastas here', default = '.')
-        p.add_argument('--batch-size', type=int, default = 0,
+        p.add_argument('--batch-size', type=non_negative_int, default = 0,
                        help='Write smaller zipfiles, each containing sigs associated with this number of accessions. \
                             This allows gbsketch to recover after unexpected failures, rather than needing to \
                             restart sketching from scratch. Default: write all sigs to single zipfile.')
@@ -118,7 +125,7 @@ def __init__(self, p):
         p.add_argument('input_csv', help="a txt file or csv file containing accessions in the first column")
         p.add_argument('-o', '--output', default=None,
                        help='output zip file for the signatures')
-        p.add_argument('--batch-size', type=int, default = 0,
+        p.add_argument('--batch-size', type=non_negative_int, default = 0,
                        help='Write smaller zipfiles, each containing sigs associated with this number of accessions. \
                             This allows urlsketch to recover after unexpected failures, rather than needing to \
                             restart sketching from scratch. Default: write all sigs to single zipfile.')

tests/test_gbsketch.py

@@ -686,3 +686,17 @@ def test_gbsketch_simple_batch_restart(runtmp, capfd):
 
     # Assert that all expected signatures are found (ignoring order)
     assert all_siginfo == expected_siginfo
+
+
+def test_gbsketch_negative_batch_size(runtmp):
+    # negative int provided for batch size
+    acc_csv = runtmp.output('acc.csv')
+    output = runtmp.output('simple.zip')
+    failed = runtmp.output('failed.csv')
+
+    with pytest.raises(utils.SourmashCommandFailed):
+        runtmp.sourmash('scripts', 'gbsketch', acc_csv,
+                    '--failed', failed, '-r', '1', '--batch-size', '-2',
+                    '--param-str', "dna,k=31,scaled=1000")
+
+    assert "Batch size cannot be negative (input value: -2)" in runtmp.last_result.err

tests/test_urlsketch.py

@@ -582,3 +582,17 @@ def test_urlsketch_simple_batch_restart(runtmp, capfd):
 
     # Verify that the loaded signatures match the expected signatures, order-independent
     assert all_siginfo == expected_siginfo, f"Loaded sigs: {all_siginfo}, expected: {expected_siginfo}"
+
+
+def test_urlsketch_negative_batch_size(runtmp):
+    # negative int provided for batch size
+    acc_csv = runtmp.output('acc1.csv')
+    output = runtmp.output('simple.zip')
+    failed = runtmp.output('failed.csv')
+
+    with pytest.raises(utils.SourmashCommandFailed):
+        runtmp.sourmash('scripts', 'urlsketch', acc_csv,
+                    '--failed', failed, '-r', '1', '--batch-size', '-2',
+                    '--param-str', "dna,k=31,scaled=1000")
+
+    assert "Batch size cannot be negative (input value: -2)" in runtmp.last_result.err