[MRG] add optional reporting for duplicated names in sketch fromfile (#…

…2580)

When `sketch fromfile` encounters duplicated names, it exits with an
error. This PR introduces an option, `--report-duplicated`, to report
those names for faster/easier debugging.

We could instead do this by default, if we want.

src/sourmash/cli/sketch/fromfile.py

@@ -71,6 +71,10 @@ def subparser(subparsers):
         '--output-manifest-matching',
         help='output a manifest file of already-existing signatures'
     )
+    file_args.add_argument(
+        '--report-duplicated', action='store_true',
+        help='report duplicated names'
+    )
 
 
 def main(args):

src/sourmash/command_sketch.py

@@ -424,6 +424,7 @@ def fromfile(args):
     skipped_sigs = 0
     n_missing_name = 0
     n_duplicate_name = 0
+    duplicate_names = set()
 
     for csvfile in args.csvs:
         with sourmash_args.FileInputCSV(csvfile) as r:
@@ -439,11 +440,14 @@ def fromfile(args):
 
                 if name in all_names:
                     n_duplicate_name += 1
+                    duplicate_names.add(name)
                 else:
                     all_names[name] = (genome, proteome)
 
     fail_exit = False
     if n_duplicate_name:
+        if args.report_duplicated:
+            notify("duplicated:\n" + '\n'.join(sorted(duplicate_names)))
         error(f"** ERROR: {n_duplicate_name} entries have duplicate 'name' records. Exiting!")
         fail_exit = True
 

tests/test_sourmash_sketch.py

@@ -1796,6 +1796,28 @@ def test_fromfile_dna_and_protein_dup_name(runtmp):
 
     print(out)
     print(err)
+    assert "GCA_903797575 Salmonella enterica" not in err
+    assert "ERROR: 1 entries have duplicate 'name' records. Exiting!" in err
+
+
+def test_fromfile_dna_and_protein_dup_name_report(runtmp):
+    # duplicate names
+    test_inp = utils.get_test_data('sketch_fromfile')
+    shutil.copytree(test_inp, runtmp.output('sketch_fromfile'))
+
+    with pytest.raises(SourmashCommandFailed):
+        runtmp.sourmash('sketch', 'fromfile',
+                        'sketch_fromfile/salmonella.csv',
+                        'sketch_fromfile/salmonella.csv',
+                        '--report-duplicated',
+                        '-o', 'out.zip', '-p', 'dna', '-p', 'protein')
+
+    out = runtmp.last_result.out
+    err = runtmp.last_result.err
+
+    print(out)
+    print(err)
+    assert "GCA_903797575 Salmonella enterica" in err
     assert "ERROR: 1 entries have duplicate 'name' records. Exiting!" in err