Print contents of array in __repr__

heat/core/printing.py

@@ -210,7 +210,8 @@ def __repr__(dndarray) -> str:
     Returns a printable representation of the passed DNDarray.
     Unlike the __str__ method, which prints a representation targeted at users, this method targets developers by showing key internal parameters of the DNDarray.
     """
-    return f"<DNDarray(MPI-rank: {dndarray.comm.rank}, Shape: {dndarray.shape}, Split: {dndarray.split}, Local Shape: {dndarray.lshape}, Device: {dndarray.device}, Dtype: {dndarray.dtype.__name__})>"
+    tensor_string = torch._tensor_str._tensor_str(dndarray.larray, __INDENT + 1)
+    return f"DNDarray(MPI-rank: {dndarray.comm.rank}, Shape: {dndarray.shape}, Split: {dndarray.split}, Local Shape: {dndarray.lshape}, Device: {dndarray.device}, Dtype: {dndarray.dtype.__name__}\n{' ' * __INDENT} {tensor_string})"
 
 
 def _torch_data(dndarray, summarize) -> DNDarray:

heat/core/tests/test_printing.py

@@ -431,14 +431,36 @@ def test_split_2_above_threshold(self):
             self.assertEqual(comparison, __str)
 
     def test___repr__(self):
-        a = ht.array([1, 2, 3, 4])
+        a = ht.array([1, 2, 3, 4], split=0)
         r = a.__repr__()
+        expect_meta = f"DNDarray(MPI-rank: {a.comm.rank}, Shape: {a.shape}, Split: {a.split}, Local Shape: {a.lshape}, Device: {a.device}, Dtype: {a.dtype.__name__}"
+        self.assertEqual(r[:r.index('\n')], expect_meta)
+
+        if ht.comm.size == 1:
+            loc_data_str  = '1, 2, 3, 4'
+        elif ht.comm.size == 2:
+            loc_data_str = f'{ht.comm.rank*2+1}, {ht.comm.rank*2+2}'
+        elif ht.comm.size == 3:
+            if ht.comm.rank == 0:
+                loc_data_str = '1, 2'
+            else:
+                loc_data_str = f'{ht.comm.rank + 2}'
+        else:
+            if ht.comm.rank < 4:
+                loc_data_str = f'{ht.comm.rank + 1}'
+            else:
+                loc_data_str = ''
+
+        expect = f'{expect_meta}\n         [{loc_data_str}])'
+
         self.assertEqual(
             r,
-            f"<DNDarray(MPI-rank: {a.comm.rank}, Shape: {a.shape}, Split: {a.split}, Local Shape: {a.lshape}, Device: {a.device}, Dtype: {a.dtype.__name__})>",
+            expect,
         )
 
 
+
+
 class TestPrintingGPU(TestCase):
     def test_print_GPU(self):
         # this test case also includes GPU now, checking the output is not done; only test whether the routine itself works...