MAINT: Transpose the result, rather than working with a transposed view

.transpose does not specify that it must return a view, so subclasses (like np.ma.array) could otherwise break this. This exposes some more need for matrix special casing.
numpy · shoyer · Feb 12, 2017 · Jan 2, 2017 · Jan 2, 2017 · Jan 2, 2017
commit c8efc57ae6f3e927c2ded4b68c9c031fe03544ed
diff --git a/numpy/lib/shape_base.py b/numpy/lib/shape_base.py
@@ -103,39 +103,50 @@ def apply_along_axis(func1d, axis, arr, *args, **kwargs):
             % (axis, nd))
 
     # arr, with the iteration axis at the end
-    dims_in = list(range(nd))
-    inarr_view = transpose(arr, dims_in[:axis] + dims_in[axis+1:] + [axis])
+    in_dims = list(range(nd))
+    inarr_view = transpose(arr, in_dims[:axis] + in_dims[axis+1:] + [axis])
 
-    # indices
-    inds = ndindex(inarr_view.shape[:nd-1])
+    # compute indices for the iteration axes
+    inds = ndindex(inarr_view.shape[:-1])
 
     # invoke the function on the first item
     ind0 = next(inds)
-    res = func1d(inarr_view[ind0], *args, **kwargs)
-    res = asanyarray(res)
+    res = asanyarray(func1d(inarr_view[ind0], *args, **kwargs))
+
+    # build a buffer for storing evaluations of func1d.
+    # remove the requested axis, and add the new ones on the end.
+    # laid out so that each write is contiguous.
+    # for a tuple index inds, buff[inds] = func1d(inarr_view[inds])
+    buff = zeros(inarr_view.shape[:-1] + res.shape, res.dtype)
+
+    # permutation of axes such that out = buff.transpose(buff_permute)
+    buff_dims = list(range(buff.ndim))
+    buff_permute = (
+        buff_dims[0 : axis] +
+        buff_dims[buff.ndim-res.ndim : buff.ndim] +
+        buff_dims[axis : buff.ndim-res.ndim]
+    )
 
-    # insert as many axes as necessary to create the output
-    outshape = arr.shape[:axis] + res.shape + arr.shape[axis+1:]
-    outarr = zeros(outshape, res.dtype)
-    # matrices call reshape in __array_prepare__, and this is apparently ok!
+    # matrices have a nasty __array_prepare__ and __array_wrap__
     if not isinstance(res, matrix):
-        outarr = res.__array_prepare__(outarr)
-    if outshape != outarr.shape:
-        raise ValueError('__array_prepare__ should not change the shape of the resultant array')
-
-    # outarr, with inserted dimensions at the end
-    # this means that outarr_view[i] = func1d(inarr_view[i])
-    dims_out = list(range(outarr.ndim))
-    outarr_view = transpose(outarr, dims_out[:axis] + dims_out[axis+res.ndim:] + dims_out[axis:axis+res.ndim])
+        buff = res.__array_prepare__(buff)
 
-    # save the first result
-    outarr_view[ind0] = res
+    # save the first result, then compute and save all remaining results
+    buff[ind0] = res
     for ind in inds:
-        outarr_view[ind] = asanyarray(func1d(inarr_view[ind], *args, **kwargs))
+        buff[ind] = asanyarray(func1d(inarr_view[ind], *args, **kwargs))
 
-    outarr = res.__array_wrap__(outarr)
+    if not isinstance(res, matrix):
+        # wrap the array, to preserve subclasses
+        buff = res.__array_wrap__(buff)
 
-    return outarr
+        # finally, rotate the inserted axes back to where they belong
+        return transpose(buff, buff_permute)
+
+    else:
+        # matrices have to be transposed first, because they collapse dimensions!
+        out_arr = transpose(buff, buff_permute)
+        return res.__array_wrap__(out_arr)
 
 
 def apply_over_axes(func, a, axes):