Heikki Linnakangas <hlinnaka@iki.fi> writes:
> On 05/20/2017 01:48 PM, Tom Lane wrote:
>> Attached is a proposed patch. I'm vacillating on whether to
>> back-patch this --- it will fix a reported bug, but it seems
>> at least somewhat conceivable that it will also break cases
>> that were working acceptably before. Thoughts?
> +1 for back-patching. If I understand correctly, it would change the
> behavior when you pass a string with non-ASCII leading or trailing
> whitespace characters to those functions. I suppose that can happen, but
> it's only pure luck if the current code happens to work in that case.
Well, it'd work properly for e.g. no-break space in LATINn. But that
seems like a very narrow use-case, and probably not enough to justify
the misbehavior you might get in multi-byte character sets.
A compromise possibly worth considering is to apply isspace() only in
single-byte encodings. I think that's more complication than it's
worth, but others might think differently.
regards, tom lane