On 2017-11-24 18:14, Dr. Werner Fink wrote:
Hi,
there is a drop request against fortune[1]. I'd like to vote not to drop fortune even it it does not use iconv but recode.
Werner
I have no idea what the recode vs iconv issues are. apropos recode throws tons of hits, so I'm baffled. It seems a set of tools to convert from one charset to another. Is this going to disappear, too? apropos iconv throws several hits, but fewer. It appears to do a similar thing. fortune.c is a single source file of 41000 bytes, 1722 lines. Not big. It includes recode.h. I was a C programmer, but I haven't touched it in more than a decade, and not in Linux. I don't know, but I think I can't help much with switching from recode to iconv. I see in the repos a "misfortune" package, described as a "Fortune-mod clone". Can we use that as a drop-in replacement to "fortune", or is it sufficiently similar. Can it use the same fortune listings? I see it in "https://github.com/mokus0/misfortune", I think. README.md This is a fortune-mod clone. In addition to the features generally expected of a fortune program, this can be used as a Haskell library (import Data.Fortune) and also supports UTF-8 fortune files, configurable search paths, automatic merging of fortune databases with the same name (so you can have a local fortunes folder that just adds to existing fortune databases), filtering fortunes by line lengths, and a "print fortune matching regex" mode (instead of just "print all fortunes matching regex" mode). Could we use that one instead, or does it has also issues of its own? I see it is in TW. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)