On 06/07/2019 11:04, Mark Rotteveel wrote: > On 6-7-2019 15:14, Adriano dos Santos Fernandes wrote: >> On 06/07/2019 06:16, Mark Rotteveel wrote: >> >>> >>> 1. The minimal thing that can and should be done to address the problem >>> that kicked off this email thread, is to ensure that `decodeTimestampTz` >>> has a fallback if ICU is not available to render named zones properly >>> (that is: render the time as UTC or +00:00 which is 1) possible with the >>> current information available and 2) an acceptable fallback). >>> >> >> The fallback is there already: caller can detect problem and do what he >> wants. >> >> Automatically change something that may be re-entered as input and saved >> to the DB in a different way is very problematic and should not be done. > > I don't find that acceptable. Graceful degradation in > `decodeTimestampTz` improves the user experience for normal users of > ISQL over having to find obscure reasons why ISQL without a ICU install > won't work. >
I would accept the thing if it purposely formats the string with a invalid and distinguishable string, for example: GMT* Adriano Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
