Here's the typetest output for the attachment...
[EMAIL PROTECTED] hylafax]# typetest fax.txt
match against (..., 439)
....
rule: offset 0 ascii = -- failed (unprintable char 0xc3)
no match
fax.txt: Can not determine file type
HylaFAX uses isprint() to determine whether or not the "ascii" (which
really should be "text") rule matches or not...
case ASCII:
{
u_int i;
for (i = 0; i < size; i++)
if (!isprint(cp[i]) && !isspace(cp[i])) {
if (verbose)
printf("failed (unprintable char %#x)\n", cp[i]);
return (false);
}
ok = true;
goto done;
}
... there is some indication in the isprint man page that the behavior
of isprint() may actually be locale-specific. I notice that your locale
settings are for en_US...
Locale: LANG=en_US, LC_CTYPE=en_US (charmap=ISO-8859-1)
So that, at least, will need to be changed. However the TypeRules.c++
code does not setlocale() at all, and so there's certainly more to do in
order to get isprint() to recognize 0xc3 as printable in your locale. I
tried some quick tests to see if I could get setlocal() to work for
LANG_C=de_DE and your text file, but I met with failure and the code
solution is not clear to me as I've not done this kind of stuff before.
For a short-term "quick fix" you could enable the rule in typerules that
reads like this:
# Support 7-bit text which would otherwise be seen as binary.
#
0 byte x ps %F/textfmt -B -f
Courier-Bold \
-p 11 -s %s >%o <%i
Enabling this rule basically matches all previously non-matched files as
formatable with textfmt. So you need to be careful, because this will
also match on non-text files such as executable or gzip files... and
then you end up with a very long fax of garbage.
Lee.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
