Hi folk.  The Kermit Project at Columbia University (a Unicode
Consortium member) is happy to announce a Unicode-aware FTP client
for UNIX (potentially all varieties: Linux, AIX, Solaris, etc etc),
available now for testing:

  http://www.columbia.edu/kermit/ftpclient.html

In fact, it's a new release (7.1) of C-Kermit:

  http://www.columbia.edu/kermit/ckermit.html

The Unicode awareness consists of being able to convert the
character-sets of text files as part of the transfer process.
Any text file can be converted between any pair of character
sets that C-Kermit knows about.  These include:

 ascii               dec-multinational   koi8r
 british             dg-international    koi8u
 bulgaria-pc         dutch               latin1-iso
 canadian-french     elot927-greek       latin2-iso
 cp1250              elot928-greek       latin9-iso
 cp1251-cyrillic     euc-jp              macintosh-latin
 cp1252              finnish             mazovia-pc
 cp437               french              next-multinational
 cp850               german              norwegian
 cp852               greek-iso           portuguese
 cp855-cyrillic      hebrew-7            shift-jis-kanji
 cp858               hebrew-iso          short-koi
 cp862-hebrew        hp-roman8           spanish
 cp866-cyrillic      hungarian           swedish
 cp869-greek         iso2022jp-kanji     swiss
 cyrillic-iso        italian             ucs2
 danish              jis7-kanji          utf8
 dec-kanji           koi8

It can also convert a local text file from any of these sets
to any other one (with, obviously, the expected amount of
loss when going from a bigger set to a smaller one).  The
national-looking names (like french, german, dutch, etc)
are the ISO 646 7-bit sets.

The character-set features work independently of the server;
therefore no particular server is required, nor are any special
advanced FTP features.

The same FTP client will be available for Windows 9x/ME/NT/2000
after the testing phase is over.

Reports, comments, suggestions welcome.  (But don't bother
suggesting UTF-16 instead of UCS-2 -- I know that already.  One
thing at a time!)

- Frank

Reply via email to