Hi, Dňa 5. októbra 2024 10:33:27 UTC používateľ Carsten Schiefner via mailop <mailop@mailop.org> napísal:
>would a specifically registered Umlaut-Domain be helpful. One member already send me existing domain, with which i was able to see how these libs works with IDNA, it has not PTR record, but for A record: + dnspython requires extra effort to get it works with IDNA2008 and U-labels (by default it encodes "ß" as "ss") + aiodns by default works with IDNA2008, thus no extra care is needed + both are able to use underscored labels (eg. _dmarc), in contrast of idna library, which rejects that domain name with invalid char (BTW, that domain returned the same IP for both forms, with ss and with xn--) Due lack of PTR record i cannot see in which form the name is returned, but i hope/guess, that nowadays it will not differ from what mentioned Cyrillic domains returns (the IDNA2008 A-label). In other words, i got what i needed, thus thanks a lot for your effort, but do not worry anymore, >E.g., a fully operational one, even with Anycast and/or DNSSEC, but without >any function, so it could be played with? For me i only need to see how names are processed in libraries. I do not collect how DNS providers works with that (i believe, that properly) nor other functionality. To be honest, i have own script to show some info about IP address, including PTR name and FCrDNS. Some time ago i implement IDNA decoding, but in some cases it fails with valid names (then FCrDNS fails too), thus i start to think/study/test how to implement it properly. It is far from finished, but now i have better idea about what i did wrong. But, when i think about it now, IMO fully operational testing domain can be usefull for many of users/devs, something as example.com is, but with names where people can test IDNA processing, especially domain with ß, in case of .de. Thus, if you can, try suggest to create it :-) regards -- Slavko https://www.slavino.sk/ _______________________________________________ mailop mailing list mailop@mailop.org https://list.mailop.org/listinfo/mailop
