UNIX for Dummies Questions & Answers

Could someone please tell me how to convert double byte charaters to single byte characters in HP-UX 11.11 without installing any applications?
e.g. I have a double byte string below and I want to convert it to a single byte string ABCdef123

ＡＢＣｄｅｆ１２３

I tried using the iconv command but I just get a message "iconv: can not initialize the conversion".

Code:

iconv -f eucJP -t engle < note.txt > note2.txt

Which encoding is engle ?

Code:

[/tmp]$ iconv -l | grep -i eucjp
EUCJP-MS//
EUCJP-OPEN//
EUCJP-WIN//
EUCJP//
[/tmp]$ iconv -l | grep -i engle
[/tmp]$

Perhaps that may be the missing link.

Hi Vino,

Thanks for your quick response.

I was just looking for a normal single byte english encoding and I found "engle" in /usr/lib/nls/iconv/config.iconv as the only one with the word "English" in the description so I tried that.

Could you tell me the correct encoding for normal single byte english?

BTW on my HP-UX 11.11 the "-l" option for iconv does not work.

Cheers
Steve

ISO8859-1 is typically regarded as the "English" encoding by most people.

However ....

My understanding is that the full-width version of the characters are of different characters from their half-width counterparts, so they are considered different and I am not inclined to believe that you can use iconv to convert between them, without looking up an external mapping table.

For example, with Chinese that I use, input method editors (some OS, e.g. Windows, IME is part of the OS) typically carry this table as part of the install, but I am not sure if you can find a version that allows you to extract this table and use that directly. Of course, as you said, some applications exist which knows how to convert but you mentioned you do not wish to use them. My feeling is that if you know both the half-width and full-width code assignments, you can hack this table yourself.