Merge pull request #6252 from markotoplak/instance-str

RowInstance: use existing _x, _y, _metas for printout

Orange/data/table.py

@@ -160,21 +160,20 @@ def __setitem__(self, key, value):
                 self._metas[-1 - key] = value
 
     def _str(self, limit):
-        def sp_values(matrix, variables):
-            if not sp.issparse(matrix):
-                if matrix.ndim == 1:
-                    matrix = matrix[:, np.newaxis]
-                return Instance.str_values(matrix[row], variables, limit)
+        def sp_values(row, variables, sparsity=None):
+            if sparsity is None:
+                return Instance.str_values(row, variables, limit)
 
+            # row is sparse
             row_entries, idx = [], 0
             while idx < len(variables):
                 # Make sure to stop printing variables if we limit the output
                 if limit and len(row_entries) >= 5:
                     break
 
                 var = variables[idx]
-                if var.is_discrete or matrix[row, idx]:
-                    row_entries.append("%s=%s" % (var.name, var.str_val(matrix[row, idx])))
+                if var.is_discrete or row[idx]:
+                    row_entries.append("%s=%s" % (var.name, var.str_val(row[idx])))
 
                 idx += 1
 
@@ -185,15 +184,13 @@ def sp_values(matrix, variables):
 
             return s
 
-        table = self.table
-        domain = table.domain
-        row = self.row_index
-        s = "[" + sp_values(table.X, domain.attributes)
+        domain = self._domain
+        s = "[" + sp_values(self._x, domain.attributes, self.sparse_x)
         if domain.class_vars:
-            s += " | " + sp_values(table.Y, domain.class_vars)
+            s += " | " + sp_values(self._y, domain.class_vars, self.sparse_y)
         s += "]"
-        if self._domain.metas:
-            s += " {" + sp_values(table.metas, domain.metas) + "}"
+        if domain.metas:
+            s += " {" + sp_values(self._metas, domain.metas, self.sparse_metas) + "}"
         return s
 
     def __str__(self):

Orange/tests/test_table.py

@@ -1182,7 +1182,10 @@ def test_repr_sparse_with_one_row(self):
         table = data.Table("iris")[:1]
         with table.unlocked_reference():
             table.X = sp.csr_matrix(table.X)
-        repr(table)     # make sure repr does not crash
+        r = repr(table)     # make sure repr does not crash
+        self.assertEqual(r.replace("\n", ""),
+                         "[[sepal length=5.1, sepal width=3.5, "
+                         "petal length=1.4, petal width=0.2 | Iris-setosa]]")
 
     def test_inf(self):
         a = np.array([[2, 0, 0, 0],
@@ -1192,6 +1195,34 @@ def test_inf(self):
             tab = data.Table.from_numpy(None, a)
         self.assertEqual(tab.get_nan_frequency_attribute(), 3/12)
 
+    def test_str(self):
+        iris = Table("iris")
+        # instance
+        self.assertEqual("[5.1, 3.5, 1.4, 0.2 | Iris-setosa]", str(iris[0]))
+        # table
+        table_str = str(iris)
+        lines = table_str.split('\n')
+        self.assertEqual(150, len(lines))
+        self.assertEqual("[[5.1, 3.5, 1.4, 0.2 | Iris-setosa],", lines[0])
+        self.assertEqual(" [5.9, 3.0, 5.1, 1.8 | Iris-virginica]]", lines[-1])
+
+    def test_str_sparse(self):
+        iris = Table("iris")
+        with iris.unlocked_reference():
+            iris.X = sp.csr_matrix(iris.X)
+        # instance
+        s0 = "[sepal length=5.1, sepal width=3.5, " \
+             "petal length=1.4, petal width=0.2 | Iris-setosa]"
+        self.assertEqual(s0, str(iris[0]))
+        # table
+        table_str = str(iris)
+        lines = table_str.split('\n')
+        self.assertEqual(150, len(lines))
+        self.assertEqual("[" + s0 + ",", lines[0])
+        slast = "[sepal length=5.9, sepal width=3.0, " \
+                "petal length=5.1, petal width=1.8 | Iris-virginica]"
+        self.assertEqual(" " + slast + "]", lines[-1])
+
 
 def column_sizes(table):
     return (len(table.domain.attributes),