fix counterintuitive behavior of has_n_cols for csv

lib/galaxy/tool_util/verify/__init__.py

@@ -99,7 +99,9 @@ def get_filename(filename: str) -> str:
     assertions = attributes.get("assert_list", None)
     if assertions is not None:
         try:
-            verify_assertions(output_content, attributes["assert_list"], attributes.get("decompress", False))
+            # Auto-detect separator based on file type
+            sep = "," if attributes.get("ftype") == "csv" else "\t"
+            verify_assertions(output_content, attributes["assert_list"], attributes.get("decompress", False), sep=sep)
         except AssertionError as err:
             errmsg = f"{item_label} different than expected\n"
             errmsg += unicodify(err)

lib/galaxy/tool_util/verify/asserts/__init__.py

@@ -8,6 +8,7 @@
 from typing import (
     Callable,
     Dict,
+    Optional,
     Tuple,
 )
 
@@ -41,7 +42,9 @@
 assertion_functions: Dict[str, Callable] = {k: v[1] for (k, v) in assertion_module_and_functions.items()}
 
 
-def verify_assertions(data: bytes, assertion_description_list: list, decompress: bool = False):
+def verify_assertions(
+    data: bytes, assertion_description_list: list, decompress: bool = False, sep: Optional[str] = None
+):
     """This function takes a list of assertions and a string to check
     these assertions against."""
     if decompress:
@@ -51,10 +54,10 @@ def verify_assertions(data: bytes, assertion_description_list: list, decompress:
             with get_fileobj(tmpfh.name, mode="rb", compressed_formats=None) as fh:
                 data = fh.read()
     for assertion_description in assertion_description_list:
-        verify_assertion(data, assertion_description)
+        verify_assertion(data, assertion_description, sep=sep)
 
 
-def verify_assertion(data: bytes, assertion_description):
+def verify_assertion(data: bytes, assertion_description, sep: Optional[str] = None):
     tag = assertion_description["tag"]
     assert_function_name = "assert_" + tag
     assert_function = assertion_functions.get(assert_function_name)
@@ -103,5 +106,9 @@ def verify_assertion(data: bytes, assertion_description):
     if "children" in assert_function_args:
         args["children"] = assertion_description["children"]
 
+    # Only set sep if the assertion accepts it and it's not already specified in XML
+    if "sep" in assert_function_args and sep is not None and "sep" not in assertion_description["attributes"]:
+        args["sep"] = sep
+
     # TODO: Verify all needed function arguments are specified.
     assert_function(**args)

lib/galaxy/tool_util/verify/asserts/tabular.py

@@ -13,7 +13,7 @@
 )
 from ._util import _assert_number
 
-Sep = Annotated[str, AssertionParameter("Separator defining columns, default: tab")]
+Sep = Annotated[str, AssertionParameter("Separator defining columns, default: tab (or comma for csv)")]
 Comment = Annotated[
     str,
     AssertionParameter(
@@ -53,9 +53,10 @@ def assert_has_n_columns(
     Number of columns can optionally also be specified with ``delta``. Alternatively the
     range of expected occurences can be specified by ``min`` and/or ``max``.
 
-    Optionally a column separator (``sep``, default is ``\t``) `and comment character(s)
+    Optionally a column separator (``sep``) and comment character(s)
     can be specified (``comment``, default is empty string). The first non-comment
-    line is used for determining the number of columns.
+    line is used for determining the number of columns. The default separator is
+    tab for most tabular data types, but comma for csv files.
     """
     first_line = get_first_line(output, comment)
     n_columns = len(first_line.split(sep))

lib/galaxy/tool_util/xsd/galaxy.xsd

@@ -2606,9 +2606,10 @@ For instance, ``<has_n_columns n="3"/>``. The assertion tests only the first lin
 Number of columns can optionally also be specified with ``delta``. Alternatively the
 range of expected occurences can be specified by ``min`` and/or ``max``.
 
-Optionally a column separator (``sep``, default is ``       ``) `and comment character(s)
+Optionally a column separator (``sep``) and comment character(s)
 can be specified (``comment``, default is empty string). The first non-comment
-line is used for determining the number of columns.
+line is used for determining the number of columns. The default separator is
+tab for most tabular data types, but comma for csv files.
 
 $attribute_list::5]]></xs:documentation>
         </xs:annotation>
@@ -2635,7 +2636,7 @@ $attribute_list::5]]></xs:documentation>
           </xs:attribute>
           <xs:attribute name="sep" type="xs:string" use="optional">
             <xs:annotation>
-              <xs:documentation xml:lang="en"><![CDATA[Separator defining columns, default: tab]]></xs:documentation>
+              <xs:documentation xml:lang="en"><![CDATA[Separator defining columns, default: tab (or comma for csv)]]></xs:documentation>
             </xs:annotation>
           </xs:attribute>
           <xs:attribute name="comment" type="xs:string" use="optional">

test/unit/tool_util/test_verify_function.py

@@ -93,3 +93,79 @@ def test_sim_size_failure_still_updates(tmp_path):
     assert assertion_error
     assert (tmp_path / filename).exists()
     assert (tmp_path / filename).open("rb").read() == b"expected"
+
+
+def test_csv_ftype_auto_sep():
+    """test that ftype='csv' automatically sets separator for has_n_columns assertion"""
+    item_label = "csv test"
+    output_content = b"col1,col2,col3\n"
+    attributes = {
+        "ftype": "csv",
+        "assert_list": [
+            {
+                "tag": "has_n_columns",
+                "attributes": {"n": "3"},
+                "children": [],
+            }
+        ],
+    }
+
+    # This should pass because ftype="csv" triggers sep="," auto-detection
+    verify(
+        item_label,
+        output_content,
+        attributes=attributes,
+        filename=None,
+        get_filecontent=t_data_downloader_for(output_content),
+    )
+
+
+def test_tabular_ftype_auto_sep():
+    """test that ftype='tabular' uses tab separator for has_n_columns assertion"""
+    item_label = "tabular test"
+    output_content = b"col1\tcol2\tcol3\n"
+    attributes = {
+        "ftype": "tabular",
+        "assert_list": [
+            {
+                "tag": "has_n_columns",
+                "attributes": {"n": "3"},
+                "children": [],
+            }
+        ],
+    }
+
+    # This should pass because ftype="tabular" triggers sep="\t" (default)
+    verify(
+        item_label,
+        output_content,
+        attributes=attributes,
+        filename=None,
+        get_filecontent=t_data_downloader_for(output_content),
+    )
+
+
+def test_csv_ftype_explicit_sep_override():
+    """test that explicit sep in assertion overrides ftype='csv' auto-detection"""
+    item_label = "csv with explicit sep test"
+    # Tab-separated data (not comma-separated!)
+    output_content = b"col1\tcol2\tcol3\n"
+    attributes = {
+        "ftype": "csv",  # ftype is csv but data is actually tab-separated
+        "assert_list": [
+            {
+                "tag": "has_n_columns",
+                "attributes": {"n": "3", "sep": "\t"},  # Explicit sep overrides
+                "children": [],
+            }
+        ],
+    }
+
+    # This should pass because explicit sep="\t" overrides the ftype="csv" auto-detection
+    verify(
+        item_label,
+        output_content,
+        attributes=attributes,
+        filename=None,
+        get_filecontent=t_data_downloader_for(output_content),
+    )