OWFile: Warning when setting a discrete variable with more than 100 values

Author: PrimozGodec

Orange/widgets/data/owfile.py

@@ -123,6 +123,8 @@ class Warning(widget.OWWidget.Warning):
         file_too_big = widget.Msg("The file is too large to load automatically."
                                   " Press Reload to load.")
         load_warning = widget.Msg("Read warning:\n{}")
+        performance_warning = widget.Msg(
+            "Categorical variables with >100 values may decrease performance.")
 
     class Error(widget.OWWidget.Error):
         file_not_found = widget.Msg("File not found.")

Orange/widgets/utils/domaineditor.py

@@ -199,6 +199,7 @@ def __init__(self, widget):
         widget.contextAboutToBeOpened.connect(lambda args: self.set_domain(args[0]))
         widget.contextOpened.connect(lambda: self.model().set_variables(self.variables))
         widget.contextClosed.connect(lambda: self.model().set_variables([]))
+        self.widget = widget
 
         self.setModel(VarTableModel(self.variables, self))
 
@@ -268,6 +269,7 @@ def get_domain(self, domain, data):
         variables = self.model().variables
         places = [[], [], []]  # attributes, class_vars, metas
         cols = [[], [], []]  # Xcols, Ycols, Mcols
+        self.widget.Warning.performance_warning.clear()
 
         def numbers_are_round(var, col_data):
             if type(var) == ContinuousVariable:
@@ -304,6 +306,11 @@ def numbers_are_round(var, col_data):
                 var = orig_var
             elif tpe == DiscreteVariable:
                 values = list(str(i) for i in unique(col_data) if not self._is_missing(i))
+                # since some widgets performs poor when number of discrete
+                # variables is high, it will warn when user who set a variable
+                # with more than 100 unique values as discrete
+                if len(values) > 100:
+                    self.widget.Warning.performance_warning()
                 round_numbers = numbers_are_round(orig_var, col_data)
                 col_data = [np.nan if self._is_missing(x) else values.index(str(x))
                             for x in self._iter_vals(col_data)]