Re: Rot13 and letters with accents
"Doug Ewell" <[email protected]>
| Newsgroups | gmane.text.unicode.devel |
|---|---|
| Message-ID | <001801c83837$23385430$6601a8c0@DGBP7M81> |
William J Poser <wjposer at ldc dot upenn dot edu> wrote: > Clearly the rot13 algorithm should be applied only to > text that has been normalized to composed form. Actually I was going to go the other way: the text should first be normalized to *decomposed* form (NFD). For maximum obfuscation, you want as many Basic Latin letters as possible in the source text. Precomposed letters-with-diacritic don't get transformed at all, which defeats the purpose. -- Qbht Rjryy * Shyyregba, Pnyvsbeavn, HFN * ESP 4645 * HGA #14 uggc://ubzr.ebnqehaare.pbz/~qrjryy uggc://jjj1.vrgs.bet/ugzy.punegref/ygeh-punegre.ugzy uggc://jjj.nyirfgenaq.ab/znvyzna/yvfgvasb/vrgs-ynathntrf ˆ