DEPR: pd.read_table

- pd.read_table is deprecated and replaced by pd.read_csv.

- add whatsnew note

- change tests to test for warning messages

- change DataFrame.from_csv to use pandas.read_csv instead of
  pandas.read_table

- Change pandas.read_clipboard to use pandas.read_csv instead
  of pandas.read_table

Author: dahlbaek

doc/source/whatsnew/v0.24.0.txt

@@ -319,6 +319,7 @@ Deprecations
 - :meth:`DataFrame.to_stata`, :meth:`read_stata`, :class:`StataReader` and :class:`StataWriter` have deprecated the ``encoding`` argument.  The encoding of a Stata dta file is determined by the file type and cannot be changed (:issue:`21244`).
 - :meth:`MultiIndex.to_hierarchical` is deprecated and will be removed in a future version  (:issue:`21613`)
 - :meth:`Series.ptp` is deprecated. Use ``numpy.ptp`` instead (:issue:`21614`)
+- :func:`pandas.read_table` is deprecated. Use ``pandas.read_csv`` instead (:issue:`21948`)
 -
 
 .. _whatsnew_0240.prior_deprecations:

pandas/core/frame.py

@@ -1592,8 +1592,8 @@ def from_csv(cls, path, header=0, sep=',', index_col=0, parse_dates=True,
                       "for from_csv when changing your function calls",
                       FutureWarning, stacklevel=2)
 
-        from pandas.io.parsers import read_table
-        return read_table(path, header=header, sep=sep,
+        from pandas.io.parsers import read_csv
+        return read_csv(path, header=header, sep=sep,
                           parse_dates=parse_dates, index_col=index_col,
                           encoding=encoding, tupleize_cols=tupleize_cols,
                           infer_datetime_format=infer_datetime_format)

pandas/io/clipboards.py

@@ -9,7 +9,7 @@
 
 def read_clipboard(sep=r'\s+', **kwargs):  # pragma: no cover
     r"""
-    Read text from clipboard and pass to read_table. See read_table for the
+    Read text from clipboard and pass to read_csv. See read_csv for the
     full argument list
 
     Parameters
@@ -31,7 +31,7 @@ def read_clipboard(sep=r'\s+', **kwargs):  # pragma: no cover
             'reading from clipboard only supports utf-8 encoding')
 
     from pandas.io.clipboard import clipboard_get
-    from pandas.io.parsers import read_table
+    from pandas.io.parsers import read_csv
     text = clipboard_get()
 
     # try to decode (if needed on PY3)
@@ -51,7 +51,7 @@ def read_clipboard(sep=r'\s+', **kwargs):  # pragma: no cover
     # that this came from excel and set 'sep' accordingly
     lines = text[:10000].split('\n')[:-1][:10]
 
-    # Need to remove leading white space, since read_table
+    # Need to remove leading white space, since read_csv
     # accepts:
     #    a  b
     # 0  1  2
@@ -80,7 +80,7 @@ def read_clipboard(sep=r'\s+', **kwargs):  # pragma: no cover
     if kwargs.get('engine') == 'python' and PY2:
         text = text.encode('utf-8')
 
-    return read_table(StringIO(text), sep=sep, **kwargs)
+    return read_csv(StringIO(text), sep=sep, **kwargs)
 
 
 def to_clipboard(obj, excel=True, sep=None, **kwargs):  # pragma: no cover

pandas/io/parsers.py

@@ -326,6 +326,10 @@
 """ % (_parser_params % (_sep_doc.format(default="','"), _engine_doc))
 
 _read_table_doc = """
+
+.. deprecated:: 0.24.0
+   Use :func:`pandas.read_csv` instead, passing `sep='\t'` if necessary.
+
 Read general delimited file into DataFrame
 
 %s
@@ -539,6 +543,10 @@ def _make_parser_function(name, sep=','):
 
     default_sep = sep
 
+    # prepare read_table deprecation
+    if name == "read_table":
+        sep = False
+
     def parser_f(filepath_or_buffer,
                  sep=sep,
                  delimiter=None,
@@ -606,11 +614,24 @@ def parser_f(filepath_or_buffer,
                  memory_map=False,
                  float_precision=None):
 
+        # deprecate read_table
+        if name == "read_table":
+            if sep is False and delimiter is None:
+                warnings.warn("read_table is deprecated, use read_csv "
+                              "with sep='\\t' instead.",
+                              FutureWarning, stacklevel=2)
+            else:
+                warnings.warn("read_table is deprecated, use read_csv "
+                              "instead.",
+                              FutureWarning, stacklevel=2)
+            if sep is False:
+                sep = '\t'
+
         # Alias sep -> delimiter.
         if delimiter is None:
             delimiter = sep
 
-        if delim_whitespace and delimiter is not default_sep:
+        if delim_whitespace and delimiter != default_sep:
             raise ValueError("Specified a delimiter with both sep and"
                              " delim_whitespace=True; you can only"
                              " specify one.")

pandas/tests/io/conftest.py

@@ -1,5 +1,6 @@
 import pytest
 from pandas.io.parsers import read_table
+import pandas.util.testing as tm
 
 
 @pytest.fixture
@@ -17,7 +18,10 @@ def jsonl_file(datapath):
 @pytest.fixture
 def salaries_table(datapath):
     """DataFrame with the salaries dataset"""
-    return read_table(datapath('io', 'parser', 'data', 'salaries.csv'))
+    with tm.assert_produces_warning(FutureWarning,
+                                    check_stacklevel=False):
+        df = read_table(datapath('io', 'parser', 'data', 'salaries.csv'))
+    return df
 
 
 @pytest.fixture

pandas/tests/io/formats/test_format.py

@@ -1225,8 +1225,10 @@ def test_to_string(self):
         lines = result.split('\n')
         header = lines[0].strip().split()
         joined = '\n'.join(re.sub(r'\s+', ' ', x).strip() for x in lines[1:])
-        recons = read_table(StringIO(joined), names=header,
-                            header=None, sep=' ')
+        with tm.assert_produces_warning(FutureWarning,
+                                        check_stacklevel=False):
+            recons = read_table(StringIO(joined), names=header,
+                                header=None, sep=' ')
         tm.assert_series_equal(recons['B'], biggie['B'])
         assert recons['A'].count() == biggie['A'].count()
         assert (np.abs(recons['A'].dropna() -

pandas/tests/io/parser/c_parser_only.py

@@ -34,7 +34,9 @@ def test_buffer_overflow(self, malf):
         # buffer overflows in tokenizer.c
         cperr = 'Buffer overflow caught - possible malformed input file.'
         with pytest.raises(pd.errors.ParserError) as excinfo:
-            self.read_table(StringIO(malf))
+            with tm.assert_produces_warning(FutureWarning,
+                                            check_stacklevel=False):
+                self.read_table(StringIO(malf))
         assert cperr in str(excinfo.value)
 
     def test_buffer_rd_bytes(self):

pandas/tests/io/parser/comment.py

@@ -24,8 +24,10 @@ def test_comment(self):
         df = self.read_csv(StringIO(data), comment='#')
         tm.assert_numpy_array_equal(df.values, expected)
 
-        df = self.read_table(StringIO(data), sep=',', comment='#',
-                             na_values=['NaN'])
+        with tm.assert_produces_warning(FutureWarning,
+                                        check_stacklevel=False):
+            df = self.read_table(StringIO(data), sep=',', comment='#',
+                                 na_values=['NaN'])
         tm.assert_numpy_array_equal(df.values, expected)
 
     def test_line_comment(self):