Merge pull request #626 from VesnaT/fix_import_doc

[FIX] Import Documents: Handle loading folder with no readable files

orangecontrib/text/import_documents.py

@@ -44,6 +44,10 @@
 log = logging.getLogger(__name__)
 
 
+class NoDocumentsException(Exception):
+    pass
+
+
 class Reader(metaclass=Registry):
     def __init__(self, path, replace_white_space=False):
         self.path = path
@@ -168,6 +172,9 @@ def run(self):
         n_paths = len(paths)
         batch = []
 
+        if n_paths == 0:
+            raise NoDocumentsException()
+
         for path in paths:
             if len(batch) == 1 and self._report_progress is not None:
                 self._report_progress(

orangecontrib/text/widgets/owimportdocuments.py

@@ -34,7 +34,8 @@
 from Orange.widgets.widget import Output
 
 from orangecontrib.text.corpus import Corpus
-from orangecontrib.text.import_documents import ImportDocuments
+from orangecontrib.text.import_documents import ImportDocuments, \
+    NoDocumentsException
 
 try:
     from orangecanvas.preview.previewbrowser import TextLabel
@@ -526,13 +527,15 @@ def __onRunFinished(self):
         task = self.__pendingTask
         self.__pendingTask = None
 
+        corpus, errors = None, []
         try:
             corpus, errors = task.future.result()
+        except NoDocumentsException:
+            state = State.Error
+            self.error("Folder contains no readable files.")
         except Exception:
             sys.excepthook(*sys.exc_info())
             state = State.Error
-            corpus = None
-            errors = []
             self.error(traceback.format_exc())
         else:
             state = State.Done
@@ -544,7 +547,8 @@ def __onRunFinished(self):
                 if corpus.domain.class_var else 0
 
         self.corpus = corpus
-        self.corpus.name = "Documents"
+        if self.corpus:
+            self.corpus.name = "Documents"
         self.skipped_documents = errors
 
         if len(errors):

orangecontrib/text/widgets/tests/test_owimportdocuments.py

@@ -1,5 +1,6 @@
 import os
 import unittest
+from unittest.mock import patch, Mock
 
 from Orange.widgets.tests.base import WidgetTest
 from orangecontrib.text.widgets.owimportdocuments import OWImportDocuments
@@ -81,6 +82,16 @@ def test_info_box(self):
             "No document set selected", self.widget.info_area.text()
         )
 
+    @patch("orangecontrib.text.import_documents.ImportDocuments.scan",
+           Mock(return_value=[]))
+    def test_load_empty_folder(self):
+        widget = self.create_widget(OWImportDocuments)
+        path = os.path.join(os.path.dirname(__file__), "data/documents")
+        widget.setCurrentPath(path)
+        widget.reload()
+        self.wait_until_finished(widget=widget)
+        self.assertIsNone(self.get_output(widget.Outputs.data))
+
 
 if __name__ == "__main__":
     unittest.main()