There is no thing as a “Hungarian florin”; some versions of *recode* have “Hungarian florinth” which is also non-existent as well. This must come from a confusion between the names “florin” and “forint” which are actual cognates. The ƒ sign (U+0192) has no meaning in Hungarian, the fillér’s symbol was always a regular, simple, plain “f” and it has also been abolished in the 90s. Yes I know about these variants and have included them in my Unicode proposal. 0x9E was probably used because Pt resembled Ft enough to keep it readable when CP437 was accidentally used – the whole premise of CWI–2. The choice of 0xA8 is harder to explain; „¿” has also no use in Hungarian but does not resemble Ft at all. CWI–2, like Polish Mazowia, had multiple conflicting variants, but since FreeDOS, the most prevalent current source of a CWI-family codepage, uses 0x9F for the forint sign, I regard that as authoritative. *Recode* also has multiple mistakes (nonexistent or misidentified characters) in its private use section, such as separate umlaut and dieresis accents which are the same in Unicode’s eye, and a “gamma function sign” which is actually just a plain old uppercase Greek gamma (or, at most, an italic one, which could be found in Mathematical Alphanumeric Symbols and so has a non-private use Unicode codepoint). Actually it doesn’t matter for me what place we give to the forint sign in a 8-bit national codepage; FreeDOS could move it to 0x9E or 0xA8 if one was really keen on following the misguided mapping tables of an obsolete tool; what matters is that there is at least one 8-bit codepage in current use which includes a forint sign, and for which a Unicode mapping could be eventually a possibility.
If you’re interested about the whole forint sign issue, read my Unicode proposal and its follow-up: https://www.unicode.org/L2/L2023/23060r-forint-sign.pdf https://www.unicode.org/L2/L2023/23131r-forint-sign-follow-up.pdf Vacek On Sun, Jul 2, 2023 at 11:51 PM H. Peter Anvin <h...@zytor.com> wrote: > On 6/12/23 14:03, Vacek Nules wrote: > > Hi everyone, > > > > I'm Vacek Nules, a software developer from Hungary and the author of > > Unicode proposal L2/23-060, which concerns the addition of a "Ft" > > character representing the Hungarian forint to Unicode, in a similar > > vein to CP437's peseta sign. > > Although the initial opinion was supporting, acceptance of this proposal > > has been stalled by some dissenting members of the Unicode Technical > > Committee despite this character having been supported on many different > > (mostly historic and Hungarian) computer systems, including FreeDOS, > > which has a Ft sign at 0x9F in code page 57781 (Hungarian), which is > > currently unmappable to Unicode because of this character missing. > > I'd like to ask the community's opinion and possible endorsements to get > > the UTC to accept the symbol, which is also needed for porting other DOS > > software from Hungary (some still in productive non-hobby use, e.g. > > point-of-sale systems). > > > > Thanks, > > Vacek > > > > I notice that while the CWI-2 Hungarian code set 0x9F appears to be > U+0192 as in CP437, I also see that it has been listed as "Hungarian > Florin" (= fillét??) with a private encoding (U+E01F) in at least one > transcoding program; Ft has been seen at 0x9E or 0xA8. > > This private encoding conflicts with (U)CSUR, and private encodings are, > well, private to start out with. > > It would seem to me that unless the "Hungarian Florin" can be unified > with U+0192 (you ought to know far better than me), *both* of these > characters ought to be encoded based on historical precedent. > > DOS has a very critical requirement to be able to do two-way conversion > between its code pages and UCS-2 (not even UTF-16) because of the way > long filenames work in VFAT. > > -hpa >
_______________________________________________ Freedos-devel mailing list Freedos-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-devel
