Wikipedia - add language to corpus

biolab · Feb 28, 2023 · d0103c9 · d0103c9
1 parent ef1dd29
commit d0103c9
Show file tree

Hide file tree

Showing 4 changed files with 17 additions and 5 deletions.
diff --git a/orangecontrib/text/tests/test_wikipedia.py b/orangecontrib/text/tests/test_wikipedia.py
@@ -59,13 +59,20 @@ def test_search(self, _, search_mock):
         self.assertListEqual(
             [["Article 1"], ["Article 2"]], result[:, "Title"].metas.tolist()
         )
+        self.assertEqual("en", result.language)
 
         self.assertEqual(on_progress.call_count, 2)
         progress = 0
         for arg in on_progress.call_args_list:
             self.assertGreater(arg[0][0], progress)
             progress = arg[0][0]
 
+        # if searched in it language Corpus's language should be it
+        result = api.search(
+            "it", ["Clinton"], articles_per_query=2, on_progress=on_progress
+        )
+        self.assertEqual("it", result.language)
+
     @patch(
         "orangecontrib.text.wikipedia_api.wikipedia.search",
         side_effect=[ARTICLES[:3], [ARTICLES[4]]],

diff --git a/orangecontrib/text/widgets/owwikipedia.py b/orangecontrib/text/widgets/owwikipedia.py
@@ -61,7 +61,7 @@ def __init__(self, *args, **kwargs):
 
         # Language
         row += 1
-        languages = tuple(sorted(set(LANG2ISO.items()) - {(None, None)}))
+        languages = sorted(set(LANG2ISO.items()) - {(None, None)})
         language_edit = ComboBox(self, 'language', languages)
         layout.addWidget(QLabel('Language:'), row, 0, 1, self.label_width)
         layout.addWidget(language_edit, row, self.label_width, 1, self.widgets_width)

diff --git a/orangecontrib/text/widgets/utils/widgets.py b/orangecontrib/text/widgets/utils/widgets.py
@@ -18,7 +18,6 @@
 
 class ListEdit(QTextEdit):
     PLACEHOLDER_COLOR = QColor(128, 128, 128)
-    USER_TEXT_COLOR = QColor(0, 0, 0)
 
     def __init__(self, master=None, attr=None, placeholder_text=None,
                  fixed_height=None, *args):
@@ -55,7 +54,6 @@ def focusInEvent(self, event):
         if self.toPlainText() == '':
             self.clear()
             self.setFontItalic(False)
-            self.setTextColor(self.USER_TEXT_COLOR)
 
     def focusOutEvent(self, event):
         self.set_placeholder()

diff --git a/orangecontrib/text/wikipedia_api.py b/orangecontrib/text/wikipedia_api.py
@@ -69,8 +69,15 @@ def search(self, lang, queries, articles_per_query=10, should_break=None, on_pro
             if callable(should_break) and should_break():
                 break
 
-        return Corpus.from_documents(results, 'Wikipedia', self.attributes,
-                                     self.class_vars, self.metas, title_indices=[-1])
+        return Corpus.from_documents(
+            results,
+            "Wikipedia",
+            self.attributes,
+            self.class_vars,
+            self.metas,
+            title_indices=[-1],
+            language=lang,
+        )
 
     def _get(self, article, query, should_break, recursive=True):
         try: