[Backport #14520] BUG: don't close user-provided file handles in C parser (GH14418) (#14520)

(cherry picked from commit eb7bd99)

Author: jorisvandenbossche

doc/source/whatsnew/v0.19.1.txt

@@ -38,6 +38,7 @@ Bug Fixes
 
 - Bug in ``pd.read_csv`` for the C engine in which quotation marks were improperly parsed in skipped rows (:issue:`14459`)
 - Bug in ``pd.read_csv`` for Python 2.x in which Unicode quote characters were no longer being respected (:issue:`14477`)
+- Fixed regression where user-provided file handles were closed in ``read_csv`` (c engine) (:issue:`14418`).
 - Bug in localizing an ambiguous timezone when a boolean is passed (:issue:`14402`)
 - Bug in ``TimedeltaIndex`` addition with a Datetime-like object where addition overflow in the negative direction was not being caught (:issue:`14068`, :issue:`14453`)
 

pandas/io/parsers.py

@@ -1456,6 +1456,8 @@ def __init__(self, src, **kwds):
     def close(self):
         for f in self.handles:
             f.close()
+
+        # close additional handles opened by C parser (for compression)
         try:
             self._reader.close()
         except:

pandas/io/tests/parser/common.py

@@ -1602,3 +1602,26 @@ def test_internal_eof_byte(self):
         expected = pd.DataFrame([["1\x1a", 2]], columns=['a', 'b'])
         result = self.read_csv(StringIO(data))
         tm.assert_frame_equal(result, expected)
+
+    def test_file_handles(self):
+        # GH 14418 - don't close user provided file handles
+
+        fh = StringIO('a,b\n1,2')
+        self.read_csv(fh)
+        self.assertFalse(fh.closed)
+
+        with open(self.csv1, 'r') as f:
+            self.read_csv(f)
+            self.assertFalse(f.closed)
+
+        # mmap not working with python engine
+        if self.engine != 'python':
+
+            import mmap
+            with open(self.csv1, 'r') as f:
+                m = mmap.mmap(f.fileno(), 0, access=mmap.ACCESS_READ)
+                self.read_csv(m)
+                # closed attribute new in python 3.2
+                if PY3:
+                    self.assertFalse(m.closed)
+                m.close()

pandas/parser.pyx

@@ -272,7 +272,7 @@ cdef class TextReader:
         parser_t *parser
         object file_handle, na_fvalues
         object true_values, false_values
-        object dsource
+        object handle
         bint na_filter, verbose, has_usecols, has_mi_columns
         int parser_start
         list clocks
@@ -554,9 +554,9 @@ cdef class TextReader:
     def close(self):
         # we need to properly close an open derived
         # filehandle here, e.g. and UTFRecoder
-        if self.dsource is not None:
+        if self.handle is not None:
             try:
-                self.dsource.close()
+                self.handle.close()
             except:
                 pass
 
@@ -641,6 +641,7 @@ cdef class TextReader:
             else:
                 raise ValueError('Unrecognized compression type: %s' %
                                  self.compression)
+            self.handle = source
 
         if isinstance(source, basestring):
             if not isinstance(source, bytes):
@@ -684,8 +685,6 @@ cdef class TextReader:
             raise IOError('Expected file path name or file-like object,'
                           ' got %s type' % type(source))
 
-        self.dsource = source
-
     cdef _get_header(self):
         # header is now a list of lists, so field_count should use header[0]