docs: adds comment to fix using ndarray and fixes indenting

Author: hexgnu

pandas/_libs/hashtable_class_helper.pxi.in

@@ -258,44 +258,44 @@ dtypes = [('Float64', 'float64', 'val != val', True),
 def get_dispatch(dtypes):
   for (name, dtype, null_condition, float_group) in dtypes:
     unique_template = """\
-      cdef:
-         Py_ssize_t i, n = len(values)
-         int ret = 0
-         {dtype}_t val
-         khiter_t k
-         bint seen_na = 0
-         {name}Vector uniques = {name}Vector()
-         {name}VectorData *ud
-
-      ud = uniques.data
-
-      with nogil:
-          for i in range(n):
-              val = values[i]
-              IF {float_group}:
-                if val == val:
-                    k = kh_get_{dtype}(self.table, val)
-                    if k == self.table.n_buckets:
-                        kh_put_{dtype}(self.table, val, &ret)
-                        if needs_resize(ud):
-                            with gil:
-                                uniques.resize()
-                        append_data_{dtype}(ud, val)
-                elif not seen_na:
-                    seen_na = 1
-                    if needs_resize(ud):
-                        with gil:
-                            uniques.resize()
-                    append_data_{dtype}(ud, NAN)
-              ELSE:
-                k = kh_get_{dtype}(self.table, val)
-                if k == self.table.n_buckets:
-                    kh_put_{dtype}(self.table, val, &ret)
-                    if needs_resize(ud):
-                        with gil:
-                            uniques.resize()
-                    append_data_{dtype}(ud, val)
-      return uniques.to_array()
+        cdef:
+           Py_ssize_t i, n = len(values)
+           int ret = 0
+           {dtype}_t val
+           khiter_t k
+           bint seen_na = 0
+           {name}Vector uniques = {name}Vector()
+           {name}VectorData *ud
+
+        ud = uniques.data
+
+        with nogil:
+            for i in range(n):
+                val = values[i]
+                IF {float_group}:
+                  if val == val:
+                      k = kh_get_{dtype}(self.table, val)
+                      if k == self.table.n_buckets:
+                          kh_put_{dtype}(self.table, val, &ret)
+                          if needs_resize(ud):
+                              with gil:
+                                  uniques.resize()
+                          append_data_{dtype}(ud, val)
+                  elif not seen_na:
+                      seen_na = 1
+                      if needs_resize(ud):
+                          with gil:
+                              uniques.resize()
+                      append_data_{dtype}(ud, NAN)
+                ELSE:
+                  k = kh_get_{dtype}(self.table, val)
+                  if k == self.table.n_buckets:
+                      kh_put_{dtype}(self.table, val, &ret)
+                      if needs_resize(ud):
+                          with gil:
+                              uniques.resize()
+                      append_data_{dtype}(ud, val)
+        return uniques.to_array()
       """
 
     unique_template = unique_template.format(name=name, dtype=dtype, null_condition=null_condition, float_group=float_group)
@@ -497,9 +497,14 @@ cdef class {{name}}HashTable(HashTable):
 
     @cython.boundscheck(False)
     def unique(self, ndarray[{{dtype}}_t, ndim=1] values):
-      if values.flags.writeable:
-        return self.unique_memview(values)
-
+        if values.flags.writeable:
+          # If the value is writeable (mutable) then use memview
+          return self.unique_memview(values)
+
+        # We cannot use the memoryview version on readonly-buffers due to
+        # a limitation of Cython's typed memoryviews. Instead we can use
+        # the slightly slower Cython ndarray type directly.
+        # see https://github.com/cython/cython/issues/1605
 {{unique_template}}
 
     @cython.boundscheck(False)