Re: CVS commit: src/usr.bin/mail

[christos%zoulas.com@localhost (Christos Zoulas)]

On Mar 8, 12:31pm, sdaoden%gmail.com@localhost (Steffen "Daode" Nurpmeso) wrote:
-- Subject: Re: CVS commit: src/usr.bin/mail

| christos%zoulas.com@localhost (Christos Zoulas) wrote:
|  |On Mar 7,  7:15pm, sdaoden%gmail.com@localhost (Steffen "Daode" Nurpmeso) 
wrote:
|  |-- Subject: Re: CVS commit: src/usr.bin/mail
|  |
|  || Not at all;  so it's a plain improvement of the codebase.
|  || How about caching the result, just as S-nail does; i think with
|  || your change you fork a shell subprocess for each and every
|  || getfold() now?
|  |
|  |Yes, that sucks. But how did it work before? I guess I will have to
|  |revert the patch and figure it out.
| 
| I'll look into it and send a patch if you like -- i like the idea
| as such myself, and in fact i (still) often do 'set folder=~ ..'
| and then realize the special treatment of this variable, and this
| hurts since unfortunately there is neither history nor
| tab-completion support...

Sure, please go for it.

| I still don't know what do you mean by that; doesn't Mail(1) use
| iconv(3) etc. already?  I.e., if CHARSET_SUPPORT is set.
| The only other solution i can think of regarding multibyte safety
| is a forward search, which should address even those multibyte
| encodings that use clearly identifiable sequence starts.

I was talking about providing error messages in multiple languages,
handing multi-byte stuff in mailrc etc.

christos