File no_float128.patch of Package python-GridDataFormats

From 069422d54344034b8f31bb19eee8a097b582c0e1 Mon Sep 17 00:00:00 2001
From: Dominik 'Rathann' Mierzejewski <dominik@greysector.net>
Date: Fri, 19 Jan 2018 14:56:52 +0100
Subject: [PATCH] use longdouble instead of float128 in
 test_write_dx_ValueError

float128 doesn't exist on 32-bit arches and longdouble maps to float96
there. See also https://github.com/numpy/numpy/issues/5272 .
This fixes issue #44.
---
 gridData/tests/test_dx.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gridData/tests/test_dx.py b/gridData/tests/test_dx.py
index 9c7714b..36c1e81 100644
--- a/gridData/tests/test_dx.py
+++ b/gridData/tests/test_dx.py
@@ -85,5 +85,5 @@ def test_write_dx_byte(nptype="uint8", dxtype="byte"):
 
 
 @raises(ValueError)
-def test_write_dx_ValueError(nptype="float128", dxtype="unknown"):
+def test_write_dx_ValueError(nptype="longdouble", dxtype="unknown"):
     return _test_write_dx(nptype=nptype, dxtype=dxtype)