[Gretl-devel] Re: problem with sprintf in ternary statement
On Fri, 27 Dec 2019, Sven Schreiber wrote:
Am 27.12.2019 um 21:38 schrieb Allin Cottrell:
> On Mon, 23 Dec 2019, Sven Schreiber wrote:
>> # short form with anonymous sprintf
>> string snull = 1 ? sprintf("\"hey\"") : "null" #
error
>> print snull
>> </hansl>
>
> Hmm, it turns out we were special-casing the interpretation of
> backslash+double-quote in the context of the ternary query operator. On
> disabling that special treatment your example works fine. Since I can't
> think of any good reason to maintain the special case I've now disabled
> it in git.
>
> I haven't yet run all my regression tests to verify that this change
> doesn't break any old formulations but I'll do so before long.
Thanks - any way to find out when that special case was introduced?
Good thought, but offhand I don't know.
The only free-floating mental association I have when thinking of
backslash plus quotes which isn't meant to be an escaped quote is in a
Windows path.
Me too. So I tried testing some ternary examples with Windows-style
paths and didn't run into a problem, so my current thought is that
the exception was not really warranted. But I'll go ahead and run
all my tests and see what happens.
Allin