Merge pull request #4305 from cpcloud/add-urlerror-info

CLN: let URLError fall through in pandas.io.html._read

Author: cpcloud

doc/source/release.rst

@@ -128,6 +128,8 @@ pandas 0.12
     of the default datetime.min and datetime.max (respectively), thanks @SleepingPills
   - ``read_html`` now raises when no tables are found and BeautifulSoup==4.2.0
     is detected (:issue:`4214`)
+  - ``read_html`` now raises a ``URLError`` instead of catching and raising a
+    ``ValueError`` (:issue:`4303`, :issue:`4305`)
 
 **API Changes**
 

doc/source/v0.12.0.txt

@@ -348,6 +348,9 @@ Other Enhancements
   - ``read_html`` now raises when no tables are found and BeautifulSoup==4.2.0
     is detected (:issue:`4214`)
 
+  - ``read_html`` now raises a ``URLError`` instead of catching and raising a
+    ``ValueError`` (:issue:`4303`, :issue:`4305`)
+
 Experimental Features
 ~~~~~~~~~~~~~~~~~~~~~
 

pandas/io/html.py

@@ -113,11 +113,8 @@ def _read(io):
     raw_text : str
     """
     if _is_url(io):
-        try:
-            with urlopen(io) as url:
-                raw_text = url.read()
-        except urllib2.URLError:
-            raise ValueError('Invalid URL: "{0}"'.format(io))
+        with urlopen(io) as url:
+            raw_text = url.read()
     elif hasattr(io, 'read'):
         raw_text = io.read()
     elif os.path.isfile(io):

pandas/io/tests/test_html.py

@@ -4,6 +4,7 @@
 from unittest import TestCase
 import warnings
 from distutils.version import LooseVersion
+import urllib2
 
 import nose
 from nose.tools import assert_raises
@@ -24,7 +25,7 @@
 from pandas.util.testing import (assert_frame_equal, network,
                                  get_data_path)
 
-from pandas.util.testing import makeCustomDataframe as mkdf
+from pandas.util.testing import makeCustomDataframe as mkdf, rands
 
 
 def _have_module(module_name):
@@ -285,9 +286,15 @@ def test_file_like(self):
 
         assert_framelist_equal(df1, df2)
 
+    @network
     def test_bad_url_protocol(self):
-        self.assertRaises(ValueError, self.run_read_html, 'git://github.com',
-                          '.*Water.*')
+        self.assertRaises(urllib2.URLError, self.run_read_html,
+                          'git://github.com', '.*Water.*')
+
+    @network
+    def test_invalid_url(self):
+        self.assertRaises(urllib2.URLError, self.run_read_html,
+                          'http://www.a23950sdfa908sd.com')
 
     @slow
     def test_file_url(self):