[cups.general] Character set encoding names

[Tim Waugh]

I have a problem with printing a document that is encoded in SJIS.  When
I try to print it (using CUPS-1.2.2) like this:

  LC_ALL=ja_JP.sjis lpr file.sjis

there is no print output, but there is an error in error_log from
texttops:

[Job 518] envp[16]="CHARSET=windows-932"
....
[Job 518] Unable to open /usr/share/cups/charsets/windows-932: No such
file or directory

I've also tried with a text-to-PostScript filter based on the paps
project (using pango), and that tries to use value of $CHARSET with
iconv_open().  "windows-932" isn't accepted as a valid character set
encoding.

I looked up "Windows-932" on Wikipedia and discovered that IANA refers
to this as WINDOWS-31J -- and sure enough, that name is accepted by
iconv_open().

So I have two questions:

1. Are filters expected to be able to use $CHARSET as an encoding name
suitable to passing to iconv_open(), and if not how are they meant to
interpret them?

2. Should CUPS ship a windows-932 charset file, or is it a bug that
texttops looks for one?

Sample SJIS file attached.

Thanks,
Tim.
*/

-------------- next part --------------
A non-text attachment was scrubbed...
Name: testpage.sjis
Type: application/octet-stream
Size: 414 bytes
Desc: not available
URL: <https://lists.cups.org/pipermail/cups/attachments/20060816/84e27e4f/attachment-0001.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <https://lists.cups.org/pipermail/cups/attachments/20060816/84e27e4f/attachment.bin>