John Darrington <[EMAIL PROTECTED]> writes: > We could assume that files we read or write are in the current > locale. > > But what happens if that assumption is false? I don't think PSPP > should then claim that the file is "corrupt", and refuse to load it. > If someone gives me a Swedish system file, but I don't have a Swedish > locale on my machine, I'd still like to be able to use that file (I'd > accept that the strings wouldn't display properly for me).
[...] > In view of this rather less than ideal design of SPSS system files, I > think the only thing we can do is to forget validation of variable > names when reading system files (or at least make the validation > rather sloppy). How about if we disallow names that begin with the "forbidden" characters @ and #, but forgo other forms of validation? Allowing @ and # could throw a wrench into other code, but other "invalid" characters will just make it difficult or impossible to refer to those variables. -- "Then, I came to my senses, and slunk away, hoping no one overheard my thinking." --Steve McAndrewSmith in the Monastery _______________________________________________ pspp-dev mailing list [email protected] http://lists.gnu.org/mailman/listinfo/pspp-dev
