Re: Problem with swedish letters

> good catch. How about '-' (dash)? I'm not crazy about the idea of
> using a regular alphabetical character (X) for the replacement.

Dash should work OK as far as I know. I do agree that it is better
with a non-character replacement.

> Any hints? If you could provide a skeleton of the algorithm, or if
you
> can point me to an open source java library that implements this
> mapping, I can attempt to include the feature. Otherwise, it will
have
> to wait for later versions

I don't know anything about Java, but I was thinking of something like
an index vector with the same length as the latin-1 charset (i.e.
0..255). Pseude code below (with "indentation underscore" so that
Yahoo does not screw up my indentation...):

# Initialize vector somewhere in your code
# Fill vector with "dash"
for c = 0..255
___vector[c] = '-'
endfor

# Fill in "accepted characters" in vector
for c = ascii 46, 48..57, 65..90, 95, 97..122
___vector[c] = c
endfor

# Fill in look-alikes
c[192..198] = 'A'
c[199] = 'C'
c[200..203] = 'E'
c[204..207] = 'I'
c[208} = 'D'
c[209] = 'N'
c[210..214, 216] = 'O'
c[217..220] = 'U'
c[221] = 'Y'
c[223] = 'B'
c[224..230] = 'a'
c[231] = 'c'
c[232..235] = 'e'
c[236..239] = 'i'
c[240] = 'd'
c[241] = 'n'
c[242..246, 248] = 'o'
c[249..252] = 'u'
c[253, 255] = 'y'

# Then just use the vector when translation is needed
for each character c in a filename
___if c is not latin-1 then
______c = '-'
___else
______c = vector[c]
___endif
endfor

I guess that you have access to a Latin-1 table, otherwise you can
take a look at e.g. 
http://www.houseofmabel.com/programs/html3/docs/iso-1.html

/Gunnar




 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
     http://groups.yahoo.com/group/pixory-beta/

<*> To unsubscribe from this group, send an email to:
     [email protected]

<*> Your use of Yahoo! Groups is subject to:
     http://docs.yahoo.com/info/terms/