[Gretl-devel] Re: intended effect of 'set force_decpoint off' ?

On Thu, Aug 15, 2024 at 6:25 AM Sven Schreiber <sven.schreiber(a)fu-berlin.de&gt; wrote: > > Am 14.08.2024 um 23:49 schrieb Cottrell, Allin: > > I can confirm this (bad, unintended) behavior, but it's limited to the > > combination of Windows + you're operating in a locale that specifies > > the decimal comma by default, but choose to use the decimal dot. On > > Linux it works as intended: both "set force_decpoint on" and "set > > force_decpoint off" do nothing if the decimal dot is already selected, > > either because it's native to the locale or the user has registered a > > preference for the dot. Locale issues are difficult in Windows due to > > their non-standard legacy specification. However it ought to be > > possible to fix this (I hope). > > > Thanks, Allin, that's good to hear, because this special case (not using > the local separator) is probably not super-common. Ah, actually I was wrong about Linux. I tested by setting LANG in the environment and there wasn't a problem, but if I use gretl's built-in language selection mechanism I get the same wrong behavior that I saw on Windows. However, if the problem is not Windows-specific it should be easier to solve.