BUG: add parse_dates check in cparser (pandas-dev#31251)

Author: sathyz

pandas/io/parsers.py

@@ -6,6 +6,7 @@
 import csv
 import datetime
 from io import BufferedIOBase, RawIOBase, StringIO, TextIOWrapper
+from itertools import chain
 import re
 import sys
 from textwrap import fill
@@ -1423,6 +1424,26 @@ def __init__(self, kwds):
         # keep references to file handles opened by the parser itself
         self.handles = []
 
+    def _confirm_parse_dates_presence(self, columns):
+        """
+        if user has provided names for parse_dates, check if those columns
+        are available.
+        """
+        if isinstance(self.parse_dates, list):
+            cols_needed = self.parse_dates
+        elif isinstance(self.parse_dates, dict):
+            cols_needed = chain(*self.parse_dates.values())
+        else:
+            cols_needed = []
+
+        missing_cols = ", ".join(
+            [col for col in cols_needed if isinstance(col, str) and col not in columns]
+        )
+        if missing_cols:
+            raise ValueError(
+                f"Missing column provided to 'parse_dates': '{missing_cols}'"
+            )
+
     def close(self):
         for f in self.handles:
             f.close()
@@ -1942,6 +1963,7 @@ def __init__(self, src, **kwds):
             if len(self.names) < len(usecols):
                 _validate_usecols_names(usecols, self.names)
 
+        self._confirm_parse_dates_presence(self.names)
         self._set_noconvert_columns()
 
         self.orig_names = self.names
@@ -2312,6 +2334,7 @@ def __init__(self, f, **kwds):
             if self.index_names is None:
                 self.index_names = index_names
 
+        self._confirm_parse_dates_presence(self.columns)
         if self.parse_dates:
             self._no_thousands_columns = self._set_no_thousands_columns()
         else:
@@ -3283,7 +3306,9 @@ def _isindex(colspec):
                 if isinstance(colspec, int) and colspec not in data_dict:
                     colspec = orig_names[colspec]
                 elif colspec not in orig_names:
-                    raise ValueError(f"'{colspec}' is not in list")
+                    raise ValueError(
+                        f"Missing column provided to 'parse_dates': '{colspec}'"
+                    )
                 if _isindex(colspec):
                     continue
                 data_dict[colspec] = converter(data_dict[colspec])

pandas/tests/io/parser/test_parse_dates.py

@@ -1518,12 +1518,13 @@ def test_hypothesis_delimited_date(date_format, dayfirst, delimiter, test_dateti
     assert result == expected
 
 
-def test_missing_column(all_parsers):
+@pytest.mark.parametrize("parse_dates", [["time", ], {"date": ["time", ]}])
+def test_missing_column(all_parsers, parse_dates):
+    """GH31251 column names provided in parse_dates could be missing."""
     parser = all_parsers
-    content = StringIO("time,val\n" "212.23, 32\n")
-    date_cols = ["time"]
-    msg = "'time' is not in list"
+    content = StringIO("time,val\n2020-01-31,32\n")
+    msg = "Missing column provided to 'parse_dates': 'time'"
     with pytest.raises(ValueError, match=msg):
         parser.read_csv(
-            content, sep=",", usecols=["val"], parse_dates=date_cols,
+            content, sep=",", usecols=["val", ], parse_dates=parse_dates,
         )