Forums Feature Suggestions SJIS/JIS switch

I couldn't find a thread on it in the search system, so I guess I'm either the first to suggest it, or people suggested it in non-obvious language

I would like to recommend implementing the current IRC-Messages-"SJIS/JIS" conversion switch as a per-channel (or per window) flag rather than a program-wide setting. Half of the channels I frequent use JIS and the other half SJIS, which makes for very nasty headaches after a while, since there's no way to hotswap the option like a simple "if ($window($active).jencoding == JIS) { /encoding SJIS }"...

I can do this for english to japanese (which is a simple font swap alias), but not for JIS to and from SJIS japanese, which is a serious problem most of the time. Even with mirc's multiserver system, I find I have to use two mirc clients just so I have one that's always set to JIS and one that's always set to SJIS.

It's just very inconvenient =)

- Mike "Pomax" Kamermans

What encoding is JIS exactly? I know I switch between SJIS and ISO-2022-JP, but I can find no single encoding named just "JIS" on my system. Though I can find around three named "JIS-something".

Here's my take on the situation, and a big apology to those board members who have heard me say it a zillion times before.

A lot of users (Me included) use other alphabets in addition to the roman alphabet, kana and kanji. Therefore, rather than just implementing a SJIS/JIS switch, it would be more practical to make a major new release of mIRC with support for the Unicode-based UTF-8 encoding. This encoding aims to support all language that are in any extended use.

Given that most people would probably download this new release within a few days, they would then be able to use the UTF-8 encoding, and a SJIS/JIS swtich would be unnecessary. This solution would be MUCH more practical for non-Japanese users wanting to chat in Japanese, and for Japanese users wanting to chat in other languages than Japanese.

And before you ask, this is apparently to be implemented in the future. Just thought I'd let you know.

JIS is another name for ISO-2022-JP. JIS/ISO-2022-JP is the "real" Japanese encoding, Shift-JIS is Microsoft's bastard version of it. Or that's how I've understood it. And I still have no idea which one is preferred in IRC - with AppLocale it "just works". But AppLocale is a lousy workaround...

Indeed, I would prefer going straight to Unicode/UTF-8 as well. Make it selectable on a per-window basis and we won't have problems during the transition period, when a lot of people still use country-specific encodings in for example European channels.

And seriously, implementing Unicode at a basic level is not as hard as people on this board have presented it to be. I don't think any of mIRC's core would need changing if the strings were kept as they are (UTF-8 encoded), and simply accept Unicode input (converting it to UTF-8 and continuing like normally) and decode the incoming text from UTF-8 (to UCS-2, which Windows uses) and render it as Unicode. Windows already supports Unicode just fine. Difficulty is no excuse in this case, and I know what I'm talking about. :-)

I realise true unicode would be a lot easier, but until that's really added (and let's be honest, todo list items are not fixed sequencially) moving at least one unicode-related problem to the background would stop a lot of people I know from bitching about it (me included =).

porting a program wide switch to a window-relevant switch is a var addition to whatever struct or class that is used to control window settings, and moving the interpretation function into the window class from the probably lowerlevel incoming prvmsg it is now... that's debug included hard to imagine being more than an hour's work.

I'd gladly pay $30 for mirc if it had at least enough control to use for Japanese, for me. Of course that wouldn't help people in israel, thailand, or china, but then again those people aren't worse off either. it's just a temporary quickfix for a segement of users until the real issue (true unicode and codepage selection) is finally tackled by the mirc devteam.

Mike "Pomax" Kamermans

Well, you could always replace the "JIS conversion" with a pulldown menu offering you conversion from whatever you're inputting to whatever you want it as. E.g. UTF-8 or JIS.

I'm not a script programmer, I'm a user. I have no intention of intercepting every message to and from all servers, comparing them to known channels and usernames that should use a particular encoding, and then running a codepage conversion function on them, in mircscript.

Since mirc already has the required conversion system coded into it - just at a different location than would be functionally useful - adding a userfriendly SJIS/JIS variable to mirc windows would have preference over scripting something, as it would be less prone to bugs on the side of whoever tries to make the scripts as suggested, and doesn't require users to load all kinds of scripts from non-official sources, but can just select the option from the window options under the window icon like we can now for the window buffer, font, logging option, etc.

Please bear in mind that I am not recommending this feature because I'm the only one in need of it. (If I was, I might actually have already done what you suggested, though I don't quite like the idea of intecepting data before mirc is supposed to handle it). Rather, I'm asking this for at least fifty people I know, and a few thousand others at the least that can't be bothered to actually write a message on the official mirc board, but yak on about it in random channels that are in or about Japanese, when the issue of Japanese support (and not so much unicode in general but specifically the SJIS/JIS conversion setup as it exists currently) in mIRC is raised.

Err... I meant actually hardcoding it into mIRC.

There's no reason to have just SJIS/JIS conversion when you could have <anything> to <anything>-conversion.