Merge pull request #2393 from kernc/from_url

io, OWFile: if URL without scheme, assume http by default

Author: astaric

Orange/data/io.py

@@ -923,6 +923,12 @@ def write(cls, filename, tree):
 
 
 class UrlReader(FileFormat):
+    def __init__(self, filename):
+        filename = filename.strip()
+        if not urlparse(filename).scheme:
+            filename = 'http://' + filename
+        super().__init__(filename)
+
     @staticmethod
     def urlopen(url):
         req = Request(

Orange/tests/test_table.py

@@ -1368,6 +1368,22 @@ def test_calling_new_with_keyword_argument_filename_calls_read_data(
 
 
 class CreateTableWithUrl(TableTests):
+    def test_url_no_scheme(self):
+
+        class SkipRest(Exception):
+            pass
+
+        mock_urlopen = Mock(side_effect=SkipRest())
+        url = 'www.foo.bar/xx.csv'
+
+        with patch('Orange.data.io.UrlReader.urlopen', mock_urlopen):
+            try:
+                Table.from_url(url)
+            except SkipRest:
+                pass
+
+        mock_urlopen.assert_called_once_with('http://' + url)
+
     class _MockUrlOpen(MagicMock):
         headers = {'content-disposition': 'attachment; filename="Something-FormResponses.tsv"; '
                                           'filename*=UTF-8''Something%20%28Responses%29.tsv'}

Orange/widgets/data/owfile.py

@@ -1,6 +1,7 @@
 import os
 import logging
 from warnings import catch_warnings
+from urllib.parse import urlparse
 
 import numpy as np
 from AnyQt.QtWidgets import \
@@ -240,6 +241,15 @@ def select_sheet(self):
         self.load_data()
 
     def _url_set(self):
+        url = self.url_combo.currentText()
+        pos = self.recent_urls.index(url)
+        url = url.strip()
+
+        if not urlparse(url).scheme:
+            url = 'http://' + url
+            self.url_combo.setItemText(pos, url)
+            self.recent_urls[pos] = url
+
         self.source = self.URL
         self.load_data()
 

Orange/widgets/data/tests/test_owfile.py

@@ -1,8 +1,7 @@
 # Test methods with long descriptive names can omit docstrings
 # pylint: disable=missing-docstring
 from os import path, remove
-from unittest.mock import Mock
-from unittest import TestCase
+from unittest.mock import Mock, patch
 import pickle
 import tempfile
 
@@ -40,7 +39,7 @@ class TestOWFile(WidgetTest):
     event_data = None
 
     def setUp(self):
-        self.widget = self.create_widget(OWFile)
+        self.widget = self.create_widget(OWFile)  # type: OWFile
 
     def test_dragEnterEvent_accepts_urls(self):
         event = self._drag_enter_event(QUrl.fromLocalFile(TITANIC_PATH))
@@ -274,3 +273,13 @@ def test_domain_editor_conversions(self):
             self.assertEqual(len(data2[0].metas[0]), 1)
             # discrete integer values should stay the same after conversion to continuous
             self.assertAlmostEqual(float(data1[0][2].value), data2[0][1])
+
+    def test_url_no_scheme(self):
+        mock_urlreader = Mock(side_effect=ValueError())
+        url = 'foo.bar/xxx.csv'
+
+        with patch('Orange.widgets.data.owfile.UrlReader', mock_urlreader):
+            self.widget.url_combo.insertItem(0, url)
+            self.widget.url_combo.activated.emit(0)
+
+        mock_urlreader.assert_called_once_with('http://' + url)