ENH: Python parser now accepts delim_whitespace=True

Title is self-explanatory.

Author: gfyoung <[email protected]>

Closes #12958 from gfyoung/delim-whitespace-python and squashes the following commits:

08da127 [gfyoung] ENH: Python parser now accepts delim_whitespace=True

Author: gfyoung

doc/source/io.rst

@@ -101,8 +101,9 @@ delim_whitespace : boolean, default False
   Specifies whether or not whitespace (e.g. ``' '`` or ``'\t'``)
   will be used as the delimiter. Equivalent to setting ``sep='\+s'``.
   If this option is set to True, nothing should be passed in for the
-  ``delimiter`` parameter. This parameter is currently supported for
-  the C parser only.
+  ``delimiter`` parameter.
+
+  .. versionadded:: 0.18.1 support for the Python parser.
 
 Column and Index Locations and Names
 ++++++++++++++++++++++++++++++++++++

doc/source/whatsnew/v0.18.1.txt

@@ -74,6 +74,7 @@ Partial string indexing now matches on ``DateTimeIndex`` when part of a ``MultiI
 Other Enhancements
 ^^^^^^^^^^^^^^^^^^
 
+- ``pd.read_csv()`` now supports ``delim_whitespace=True`` for the Python engine (:issue:`12958`)
 - ``pd.read_csv()`` now supports opening ZIP files that contains a single CSV, via extension inference or explict ``compression='zip'`` (:issue:`12175`)
 - ``pd.read_csv()`` now supports opening files using xz compression, via extension inference or explicit ``compression='xz'`` is specified; ``xz`` compressions is also supported by ``DataFrame.to_csv`` in the same way (:issue:`11852`)
 - ``pd.read_msgpack()`` now always gives writeable ndarrays even when compression is used (:issue:`12359`).

pandas/io/parsers.py

@@ -57,7 +57,10 @@
     Specifies whether or not whitespace (e.g. ``' '`` or ``'\t'``) will be
     used as the sep. Equivalent to setting ``sep='\+s'``. If this option
     is set to True, nothing should be passed in for the ``delimiter``
-    parameter. This parameter is currently supported for the C parser only.
+    parameter.
+
+    .. versionadded:: 0.18.1 support for the Python parser.
+
 header : int or list of ints, default 'infer'
     Row number(s) to use as the column names, and the start of the data.
     Default behavior is as if set to 0 if no ``names`` passed, otherwise
@@ -390,7 +393,20 @@ def _read(filepath_or_buffer, kwds):
 }
 
 _c_unsupported = set(['skip_footer'])
-_python_unsupported = set(_c_parser_defaults.keys())
+_python_unsupported = set([
+    'as_recarray',
+    'na_filter',
+    'compact_ints',
+    'use_unsigned',
+    'low_memory',
+    'memory_map',
+    'buffer_lines',
+    'error_bad_lines',
+    'warn_bad_lines',
+    'dtype',
+    'decimal',
+    'float_precision',
+])
 
 
 def _make_parser_function(name, sep=','):
@@ -647,8 +663,13 @@ def _get_options_with_defaults(self, engine):
                 value = kwds[argname]
 
                 if engine != 'c' and value != default:
-                    raise ValueError('The %r option is not supported with the'
-                                     ' %r engine' % (argname, engine))
+                    if ('python' in engine and
+                            argname not in _python_unsupported):
+                        pass
+                    else:
+                        raise ValueError(
+                            'The %r option is not supported with the'
+                            ' %r engine' % (argname, engine))
             else:
                 value = default
             options[argname] = value
@@ -691,6 +712,9 @@ def _clean_options(self, options, engine):
                                   " different from '\s+' are"\
                                   " interpreted as regex)"
                 engine = 'python'
+        elif delim_whitespace:
+            if 'python' in engine:
+                result['delimiter'] = '\s+'
 
         if fallback_reason and engine_specified:
             raise ValueError(fallback_reason)