Adding initial DataFrames Support

Signed-off-by: Jason Myers <jason@jasonamyers.com>

5 files changed, 56 insertions, 1 deletions

diff --git a/setup.py b/setup.py
index e46eb88..5691821 100755
--- a/setup.py
+++ b/setup.py
@@ -48,6 +48,7 @@ install = [
     'xlrd',
     'xlwt',
     'pyyaml',
+    'pandas',
 ]
 
 with open('tablib/core.py', 'r') as fd:
diff --git a/tablib/core.py b/tablib/core.py
index be648c2..13a4514 100644
--- a/tablib/core.py
+++ b/tablib/core.py
@@ -570,6 +570,18 @@ class Dataset(object):
         """
         pass
 
+    @property
+    def df():
+        """A DataFrame representation of the :class:`Dataset` object.
+
+        A dataset object can also be imported by setting the :class:`Dataset.df` attribute: ::
+
+            data = tablib.Dataset()
+            data.df = DataFrame(np.random.randn(6,4))
+
+        Import assumes (for now) that headers exist.
+        """
+        pass
 
     @property
     def json():
diff --git a/tablib/formats/__init__.py b/tablib/formats/__init__.py
index 5cca19f..94b5bc9 100644
--- a/tablib/formats/__init__.py
+++ b/tablib/formats/__init__.py
@@ -13,5 +13,6 @@ from . import _xlsx as xlsx
 from . import _ods as ods
 from . import _dbf as dbf
 from . import _latex as latex
+from . import _df as df
 
-available = (json, xls, yaml, csv, dbf, tsv, html, latex, xlsx, ods)
+available = (json, xls, yaml, csv, dbf, tsv, html, latex, xlsx, ods, df)
diff --git a/tablib/formats/_df.py b/tablib/formats/_df.py
new file mode 100644
index 0000000..5996ce9
--- /dev/null
+++ b/tablib/formats/_df.py
@@ -0,0 +1,40 @@
+""" Tablib - DataFrame Support.
+"""
+
+
+import sys
+
+
+if sys.version_info[0] > 2:
+    from io import BytesIO
+else:
+    from cStringIO import StringIO as BytesIO
+
+from pandas import DataFrame
+
+import tablib
+
+from tablib.compat import unicode
+
+title = 'df'
+extensions = ('df', )
+
+def detect(stream):
+    """Returns True if given stream is a DataFrame."""
+    try:
+        DataFrame(stream)
+        return True
+    except ValueError:
+        return False
+
+
+def export_set(dset, index=None):
+    """Returns DataFrame representation of DataBook."""
+    dataframe = DataFrame(dset.dict, columns=dset.headers)
+    return dataframe
+
+
+def import_set(dset, in_stream):
+    """Returns dataset from DataFrame."""
+    dset.wipe()
+    dset.dict = in_stream.to_dict(orient='records')
diff --git a/test_tablib.py b/test_tablib.py
index 03a46df..9bd5f12 100755
--- a/test_tablib.py
+++ b/test_tablib.py
@@ -383,6 +383,7 @@ class TablibTestCase(unittest.TestCase):
         data.ods
         data.html
         data.latex
+        data.df
 
     def test_datetime_append(self):
         """Passes in a single datetime and a single date and exports."""