More unicode fixes

pandas/core/format.py

@@ -298,7 +298,7 @@ def to_latex(self, force_unicode=False, column_format=None):
         if column_format is None:
             column_format = '|l|%s|' % '|'.join('c' for _ in strcols)
         else:
-            assert isinstance(column_format, str)
+            assert isinstance(column_format, basestring)
 
         self.buf.write('\\begin{tabular}{%s}\n' % column_format)
         self.buf.write('\\hline\n')

pandas/core/frame.py

@@ -1366,7 +1366,7 @@ def info(self, verbose=True, buf=None):
             counts = self.count()
             assert(len(cols) == len(counts))
             for col, count in counts.iteritems():
-                if not isinstance(col, (unicode, str)):
+                if not isinstance(col, basestring):
                     col = str(col)
                 lines.append(_put_str(col, space) +
                              '%d  non-null values' % count)

pandas/io/parsers.py

@@ -156,7 +156,7 @@ def _read(cls, filepath_or_buffer, kwds):
     if skipfooter is not None:
         kwds['skip_footer'] = skipfooter
 
-    if isinstance(filepath_or_buffer, str) and _is_url(filepath_or_buffer):
+    if isinstance(filepath_or_buffer, basestring) and _is_url(filepath_or_buffer):
         from urllib2 import urlopen
         filepath_or_buffer = urlopen(filepath_or_buffer)
         if py3compat.PY3:  # pragma: no cover

pandas/io/tests/test_parsers.py

@@ -108,7 +108,12 @@ def test_empty_string(self):
 
 
     def test_read_csv(self):
-        pass
+        if not py3compat.PY3:
+            fname=u"file://"+unicode(self.csv1)
+            try:
+                df1 = read_csv(fname, index_col=0, parse_dates=True)
+            except IOError:
+                assert(False), "read_csv should accept unicode objects as urls"
 
     def test_dialect(self):
         data = """\

pandas/tseries/converter.py

@@ -33,7 +33,7 @@ def _to_ordinalf(tm):
     return tot_sec
 
 def time2num(d):
-    if isinstance(d, str):
+    if isinstance(d, basestring):
         parsed = tools.to_datetime(d)
         if not isinstance(parsed, datetime):
             raise ValueError('Could not parse time %s' % d)
@@ -150,7 +150,7 @@ def try_parse(values):
             return dates.date2num(values)
         elif (com.is_integer(values) or com.is_float(values)):
             return values
-        elif isinstance(values, str):
+        elif isinstance(values, basestring):
             return try_parse(values)
         elif isinstance(values, (list, tuple, np.ndarray)):
             if not isinstance(values, np.ndarray):

pandas/tseries/tests/test_converter.py

@@ -0,0 +1,28 @@
+from datetime import datetime, time, timedelta
+import sys
+import os
+import unittest
+
+import nose
+
+import numpy as np
+
+import pandas.tseries.converter as converter
+
+def test_timtetonum_accepts_unicode():
+    assert(converter.time2num("00:01")==converter.time2num(u"00:01"))
+
+class TestDateTimeConverter(unittest.TestCase):
+
+    def setUp(self):
+        self.dtc = converter.DatetimeConverter()
+
+    def test_convert_accepts_unicode(self):
+        r1 = self.dtc.convert("12:22",None,None)
+        r2 = self.dtc.convert(u"12:22",None,None)
+        assert(r1==r2), "DatetimeConverter.convert should accept unicode"
+
+if __name__ == '__main__':
+    import nose
+    nose.runmodule(argv=[__file__,'-vvs','-x','--pdb', '--pdb-failure'],
+                   exit=False)

pandas/tseries/tools.py

@@ -36,7 +36,7 @@ def _infer(a, b):
 
 
 def _maybe_get_tz(tz):
-    if isinstance(tz, (str, unicode)):
+    if isinstance(tz, basestring):
         import pytz
         tz = pytz.timezone(tz)
     if com.is_integer(tz):