Encodings on RB 5

[Robert Brenstein]

>Hi All,
>
>Just trying to finalise my windows version so it talks to the Mac. Behind
>the scenes I have an ID code to identify each record. This code can use any
>of the 256 characters except for obvious ones such as return, ", etc.
>
>On my Mac I may have a code for a record such as: 00TÈX@
Kw
>
>When I look at the files on the PC that I have copied over from the Mac, the
>same record's identity code is now: 00TÈX at Y´Kw
>
>This means that the PC changes the codings as soon as the database is
>dragged over.
>
>If I want to change a record on the PC, it won't find the appropriate record
>on the Macintosh because it can't find the record.
>
>I presume this has something to do with encodings. I have encodings working
>fine within machines - MacRoman on the Mac and Windows Latin 1 for Windows
>and these are going fine. But I can't tell how to make the right codings go
>from the Windows version to Macintosh across TCP/IP.
>
>If someone could help me on this I would really appreciate it.
>
>Regards
>
>Cindy
>

Hey, Ruslan, another issue where having explicit encoding as field 
attribute would avoid the problem :)

Cindy, your problem is with using high-ascii in your id strings. WHen 
valentina fetches these on another platform, it auto-converts them 
between Mac and Window character sets. This messes you up. You should 
restrict your codes to standard ascii set or encode them with base64 
or similar to preserve the special characters. Hopefully, one day, 
Valentina allows to specify both language and character encoding for 
textual fields, and then you would be able to work as you programmed 
it now.

Robert Brenstein