html_url,issue_url,id,node_id,user,created_at,updated_at,author_association,body,reactions,performed_via_github_app,issue
https://github.com/pydata/xarray/issues/416#issuecomment-107806428,https://api.github.com/repos/pydata/xarray/issues/416,107806428,MDEyOklzc3VlQ29tbWVudDEwNzgwNjQyOA==,4365718,2015-06-02T04:56:35Z,2015-06-02T04:56:35Z,NONE,"Thanks for posting, Stephan!  I'm interested in assisting with a possible enhancement!  Here's a quick wrapper I wrote to deal with this issue in my code:

``` python
def wrap_open_dataset(fname, verbose=False):
    """"""
    Forces switch to native endianness when dataset is read in so
    that endian-dependent functions (e.g., nanmin from bottleneck)
    will always work.
    Note that ec2 and my mac are little endian, but only ec2 was
    broken with a
    ""ValueError: Big-endian buffer not supported on little-endian
    compiler.""
    """"""
    ds = xray.open_dataset(fname)
    all_vars = ds.coords.keys() + ds.data_vars.keys()
    dilist = []
    for kk in all_vars:
        di = ds[kk].dtype
        dnew = di.newbyteorder('N')
        ds[kk].values = ds[kk].values.astype(dnew, casting='equiv', copy=False)
        df = ds[kk].dtype
        if verbose is True:
            dilist.append(di)
    if verbose is True:
        for kk, di in zip(all_vars, dilist):
            df = ds[kk].dtype
            print 'converted', di, 'to', df
    return ds
```
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,83700033