Merge pull request #4228 from jreback/stata_bug

TST: little endian failure, GH4222

Author: jreback

pandas/io/stata.py

@@ -327,8 +327,16 @@ def _read_header(self):
             typlist = [ord(self.path_or_buf.read(1)) for i in range(self.nvar)]
         else:
             typlist = [self.OLD_TYPE_MAPPING[self._decode_bytes(self.path_or_buf.read(1))] for i in range(self.nvar)]
-        self.typlist = [self.TYPE_MAP[typ] for typ in typlist]
-        self.dtyplist = [self.DTYPE_MAP[typ] for typ in typlist]
+
+        try:
+            self.typlist = [self.TYPE_MAP[typ] for typ in typlist]
+        except:
+            raise ValueError("cannot convert stata types [{0}]".format(','.join(typlist)))
+        try:
+            self.dtyplist = [self.DTYPE_MAP[typ] for typ in typlist]
+        except:
+            raise ValueError("cannot convert stata dtypes [{0}]".format(','.join(typlist)))
+
         if self.format_version > 108:
             self.varlist = [self._null_terminate(self.path_or_buf.read(33)) for i in range(self.nvar)]
         else:

pandas/io/tests/test_stata.py

@@ -132,7 +132,7 @@ def test_read_dta4(self):
     def test_write_dta5(self):
         if not is_little_endian():
             raise nose.SkipTest("known failure of test_write_dta5 on non-little endian")
-        
+
         original = DataFrame([(np.nan, np.nan, np.nan, np.nan, np.nan)],
                              columns=['float_miss', 'double_miss', 'byte_miss', 'int_miss', 'long_miss'])
         original.index.name = 'index'
@@ -145,7 +145,7 @@ def test_write_dta5(self):
     def test_write_dta6(self):
         if not is_little_endian():
             raise nose.SkipTest("known failure of test_write_dta6 on non-little endian")
-        
+
         original = self.read_csv(self.csv3)
         original.index.name = 'index'
 
@@ -191,6 +191,9 @@ def test_read_dta9(self):
             )
 
     def test_read_dta10(self):
+        if not is_little_endian():
+            raise nose.SkipTest("known failure of test_write_dta10 on non-little endian")
+
         original = DataFrame(
             data=
             [