Merge pull request #3702 from cpcloud/read-and-friends-path-or-buf

ENH: allow to_html and to_latex to take a path for their first argument

Author: jreback

RELEASE.rst

@@ -185,6 +185,8 @@ pandas 0.11.1
      into today's date
   - ``DataFrame.from_records`` did not accept empty recarrays (GH3682_)
   - ``DataFrame.to_csv`` will succeed with the deprecated option ``nanRep``, @tdsmith
+  - ``DataFrame.to_html`` and ``DataFrame.to_latex`` now accept a path for
+    their first argument (GH3702_)
 
 .. _GH3164: https://github.com/pydata/pandas/issues/3164
 .. _GH2786: https://github.com/pydata/pandas/issues/2786
@@ -261,6 +263,7 @@ pandas 0.11.1
 .. _GH3676: https://github.com/pydata/pandas/issues/3676
 .. _GH3675: https://github.com/pydata/pandas/issues/3675
 .. _GH3682: https://github.com/pydata/pandas/issues/3682
+.. _GH3702: https://github.com/pydata/pandas/issues/3702
 
 pandas 0.11.0
 =============

doc/source/v0.11.1.txt

@@ -273,6 +273,9 @@ Bug Fixes
 
   - ``DataFrame.from_records`` did not accept empty recarrays (GH3682_)
 
+  - ``DataFrame.to_html`` and ``DataFrame.to_latex`` now accept a path for
+    their first argument (GH3702_)
+
 See the `full release notes
 <https://github.com/pydata/pandas/blob/master/RELEASE.rst>`__ or issue tracker
 on GitHub for a complete list.
@@ -314,3 +317,4 @@ on GitHub for a complete list.
 .. _GH3675: https://github.com/pydata/pandas/issues/3675
 .. _GH3682: https://github.com/pydata/pandas/issues/3682
 .. _GH3679: https://github.com/pydata/pandas/issues/3679
+.. _GH3702: https://github.com/pydata/pandas/issues/3702

pandas/core/format.py

@@ -364,21 +364,31 @@ def get_col_type(dtype):
             raise AssertionError(('column_format must be str or unicode, not %s'
                                   % type(column_format)))
 
-        self.buf.write('\\begin{tabular}{%s}\n' % column_format)
-        self.buf.write('\\toprule\n')
-
-        nlevels = frame.index.nlevels
-        for i, row in enumerate(izip(*strcols)):
-            if i == nlevels:
-                self.buf.write('\\midrule\n')  # End of header
-            crow = [(x.replace('_', '\\_')
-                     .replace('%', '\\%')
-                     .replace('&', '\\&') if x else '{}') for x in row]
-            self.buf.write(' & '.join(crow))
-            self.buf.write(' \\\\\n')
-
-        self.buf.write('\\bottomrule\n')
-        self.buf.write('\\end{tabular}\n')
+        def write(buf, frame, column_format, strcols):
+            buf.write('\\begin{tabular}{%s}\n' % column_format)
+            buf.write('\\toprule\n')
+
+            nlevels = frame.index.nlevels
+            for i, row in enumerate(izip(*strcols)):
+                if i == nlevels:
+                    buf.write('\\midrule\n')  # End of header
+                crow = [(x.replace('_', '\\_')
+                         .replace('%', '\\%')
+                         .replace('&', '\\&') if x else '{}') for x in row]
+                buf.write(' & '.join(crow))
+                buf.write(' \\\\\n')
+
+            buf.write('\\bottomrule\n')
+            buf.write('\\end{tabular}\n')
+
+        if hasattr(self.buf, 'write'):
+            write(self.buf, frame, column_format, strcols)
+        elif isinstance(self.buf, basestring):
+            with open(self.buf, 'w') as f:
+                write(f, frame, column_format, strcols)
+        else:
+            raise TypeError('buf is not a file name and it has no write '
+                            'method')
 
     def _format_col(self, i):
         formatter = self._get_formatter(i)
@@ -392,7 +402,14 @@ def to_html(self, classes=None):
         Render a DataFrame to a html table.
         """
         html_renderer = HTMLFormatter(self, classes=classes)
-        html_renderer.write_result(self.buf)
+        if hasattr(self.buf, 'write'):
+            html_renderer.write_result(self.buf)
+        elif isinstance(self.buf, basestring):
+            with open(self.buf, 'w') as f:
+                html_renderer.write_result(f)
+        else:
+            raise TypeError('buf is not a file name and it has no write '
+                            ' method')
 
     def _get_formatted_column_labels(self):
         from pandas.core.index import _sparsify
@@ -574,7 +591,6 @@ def write_result(self, buf):
             indent = self._write_body(indent)
 
         self.write('</table>', indent)
-
         _put_lines(buf, self.elements)
 
     def _write_header(self, indent):

pandas/core/series.py

@@ -1222,7 +1222,11 @@ def to_string(self, buf=None, na_rep='NaN', float_format=None,
         if buf is None:
             return the_repr
         else:
-            print >> buf, the_repr
+            try:
+                buf.write(the_repr)
+            except AttributeError:
+                with open(buf, 'w') as f:
+                    f.write(the_repr)
 
     def _get_repr(self, name=False, print_header=False, length=True, dtype=True,
                   na_rep='NaN', float_format=None):

pandas/tests/test_format.py

@@ -1216,6 +1216,26 @@ def test_to_html(self):
         frame = DataFrame(index=np.arange(200))
         frame.to_html()
 
+    def test_to_html_filename(self):
+        biggie = DataFrame({'A': randn(200),
+                            'B': tm.makeStringIndex(200)},
+                           index=range(200))
+
+        biggie['A'][:20] = nan
+        biggie['B'][:20] = nan
+        with tm.ensure_clean('test.html') as path:
+            biggie.to_html(path)
+            with open(path, 'r') as f:
+                s = biggie.to_html()
+                s2 = f.read()
+                self.assertEqual(s, s2)
+
+        frame = DataFrame(index=np.arange(200))
+        with tm.ensure_clean('test.html') as path:
+            frame.to_html(path)
+            with open(path, 'r') as f:
+                self.assertEqual(frame.to_html(), f.read())
+
     def test_to_html_with_no_bold(self):
         x = DataFrame({'x': randn(5)})
         ashtml = x.to_html(bold_rows=False)
@@ -1474,6 +1494,13 @@ def test_dict_entries(self):
         self.assertTrue("'a': 1" in val)
         self.assertTrue("'b': 2" in val)
 
+    def test_to_latex_filename(self):
+        with tm.ensure_clean('test.tex') as path:
+            self.frame.to_latex(path)
+
+            with open(path, 'r') as f:
+                self.assertEqual(self.frame.to_latex(), f.read())
+
     def test_to_latex(self):
         # it works!
         self.frame.to_latex()