ENH: Raise ParserWarning when length of names does not match length of data

doc/source/whatsnew/v1.3.0.rst

@@ -41,6 +41,7 @@ Other enhancements
 
 - Added :meth:`MultiIndex.dtypes` (:issue:`37062`)
 - Improve error message when ``usecols`` and ``names`` do not match for :func:`read_csv` and ``engine="c"`` (:issue:`29042`)
+- :func:`read_csv` now raising ``ParserWarning`` if length of header or given names does not match length of data when usecols is not specified (:issue:`21768`)
 
 .. ---------------------------------------------------------------------------
 

pandas/_libs/parsers.pyx

@@ -729,7 +729,9 @@ cdef class TextReader:
                 field_count = max(field_count, len(self.names))
 
             passed_count = len(header[0])
-
+            print(self.allow_leading_cols)
+            print(passed_count)
+            print(field_count)
             if (self.has_usecols and self.allow_leading_cols and
                     not callable(self.usecols)):
                 nuse = len(self.usecols)
@@ -743,6 +745,11 @@ cdef class TextReader:
             # oh boy, #2442, #2981
             elif self.allow_leading_cols and passed_count < field_count:
                 self.leading_cols = field_count - passed_count
+            elif not self.allow_leading_cols and passed_count < field_count:
+                warnings.warn(
+                    "Length of header or names does not match length of data. This leads "
+                    "to a loss of data with index_col=False.", ParserWarning, stacklevel=6,
+                )
 
         return header, field_count, unnamed_cols
 

pandas/io/parsers.py

@@ -2505,6 +2505,14 @@ def _exclude_implicit_index(self, alldata):
         if self._col_indices is not None and len(names) != len(self._col_indices):
             names = [names[i] for i in sorted(self._col_indices)]
 
+        if not self.index_col and len(names) != len(alldata) and names:
+            warnings.warn(
+                "Length of header or names does not match length of data. This leads "
+                "to a loss of data with index_col=False.",
+                ParserWarning,
+                stacklevel=6,
+            )
+
         return {name: alldata[i + offset] for i, name in enumerate(names)}, names
 
     # legacy

pandas/tests/io/parser/test_common.py

@@ -15,7 +15,7 @@
 import pytest
 
 from pandas._libs.tslib import Timestamp
-from pandas.errors import DtypeWarning, EmptyDataError, ParserError
+from pandas.errors import DtypeWarning, EmptyDataError, ParserError, ParserWarning
 import pandas.util._test_decorators as td
 
 from pandas import DataFrame, Index, MultiIndex, Series, compat, concat, option_context
@@ -1062,14 +1062,17 @@ def test_skip_initial_space(all_parsers):
     tm.assert_frame_equal(result, expected)
 
 
+@pytest.mark.filterwarnings("ignore:Lenght of header:pandas.errors.ParserWarning")
 def test_trailing_delimiters(all_parsers):
     # see gh-2442
     data = """A,B,C
 1,2,3,
 4,5,6,
 7,8,9,"""
     parser = all_parsers
-    result = parser.read_csv(StringIO(data), index_col=False)
+
+    with tm.assert_produces_warning(ParserWarning):
+        result = parser.read_csv(StringIO(data), index_col=False)
 
     expected = DataFrame({"A": [1, 4, 7], "B": [2, 5, 8], "C": [3, 6, 9]})
     tm.assert_frame_equal(result, expected)
@@ -2178,7 +2181,8 @@ def test_no_header_two_extra_columns(all_parsers):
     ref = DataFrame([["foo", "bar", "baz"]], columns=column_names)
     stream = StringIO("foo,bar,baz,bam,blah")
     parser = all_parsers
-    df = parser.read_csv(stream, header=None, names=column_names, index_col=False)
+    with tm.assert_produces_warning(ParserWarning):
+        df = parser.read_csv(stream, header=None, names=column_names, index_col=False)
     tm.assert_frame_equal(df, ref)