Forums Bug Reports Unable to enter certain channels

I noticed this behavior some time back but it still persists in the recent MIRC version 7.41. This wasn't an issue in really old version of mirc like 5.x.

I can usually join a channel name in my local language without any problems except when I have certain character in the channel name.

Specifically, I can't join a channel that contains "베" character.

So the channel name can be "Te베st", "Test베", "베Test" or something, and it just won't work.

I just get a message from irc saying "No such channel"

Any other characters seem to work fine (or at least ones I've tried).

Any idea why this might be happening?


Edit: Seems like the character isn't showing up correctly on this forum

Character in question in on the link below
http://www.unicodemap.org/details/0xBCA0/index.html


Edit2:
I just did few more testing and it seems like any character with A0 at the end of the header will bug out the same way. So same issue is observed with many other characters such as 0xACA0 0xAEA0 0xAFA0 just to name a few.


Edit3:
It actually seems like any UTF-8 hexadecimal headers ending with A0 is problematic. Actually, this problem isn't unique to specific language. It seems to apply for all other language/characters

See the UTF-8 dexadecimal codes from the link below
http://lcweb2.loc.gov/diglib/codetables/9.3.html

Some example characters with problems:
E385A0 EAB0A0 EAB2A0

Khaled removed support for language-specific code pages, favoring UTF-8 instead. You will no longer be able to join those channels with mIRC through conventional means.

You may be able to join by typing /raw -n JOIN #channel_with_funny_characters

Thanks for the reply but Korean language is supported in UTF-8 and I'm not trying to use any funky characters.

This is just my local language characters, and it only bugs out characters with certain headers as explained above.

If the person who created the channel used UTF-8 characters in the channel name, then mIRC will support it.

If the person who created the channel used code page 949 (Unified Hangeul Code) or EUC-KR (Wansung) encoded characters in the channel name, then mIRC will not be able to /join that channel as typed. mIRC converts the displayed characters into UTF-8 and sends non-corresponding bytes in its JOIN request to the server -- which isn't expecting UTF-8.

It is suggested you ask the channel operators to close their channel and create a new channel using UTF-8 encoded characters instead.


YEAY 2000th POST!

I am well aware of that issue, as I have been using old version of mirc for long time. That was the part of the reason in moving to newer mirc so I don't have to deal with all that encoding issues.

This problem has NOTHING to do with euc-kr or utf-8 encoding issues.

This is really odd bug where I cannot make/join a channel with specific characters.

To clarify any misunderstanding, for Example:
If character "N" was the bugged character, then I cannot make or join channels like "#SomeName" because it has character N inside the channel name.

edit: btw, congrats on your 2k

Have you attempted the /raw -n solution?

Others are telling me to link you to these posts.

https://forums.mirc.com/ubbthreads.php/topics/223676/mIRC_v7.1_languages_and_codepages

https://forums.mirc.com/ubbthreads.php/topics/224231/Unicode_Support

Yes, I already tried /raw but still same problem persists.

Those links aren't helpful for my problems. See my edit3 above.

This problem isn't unique to my language, any language or characters with A0 hex header at the end will suffer from the same problem.

The A0 hex character, or $chr(160) is often represented as a non-breaking space. Are you attempting to join channels that have a visually apparent space in them?

See the UTF-8 hex codes, which is 0x20 not 0xA0 for space.

But just to answer your question, I don't have spaces in channel names.

edit: I see that you mean no-break-space (0xc2a0). I don't even know what that is but like I said before, I'm not trying anything funky here.

And indeed, mIRC converts any instance 0xa0 into 0xc2a0 that you send to the server, no matter how it's displayed on your screen.

All characters with values 0x80 through 0xff will get 0xc2 added in front of it.

Thanks for tracking down on the problem.
I hope this bug gets fixed on the next release.

This is by design, and is NOT a bug. mIRC sends data to and receives data from a server as binary data(since that's all computers actually understand). Before mIRC 7.x it used the local machine's active CODE PAGE to decide how to convert the received data into text or how to convert the text to binary in order to send it to the server. With mIRC 7.x code page support was intentionally removed in favor of utf8*. Due to this any character who's ANSI value would be above 127 is re-encoded by a 2 (or more) byte representation.


*: The move makes sense as most operating systems are trying to move away from code pages in favor of utf/unicode anyways.

Further Reading:
Utf8/Unicode Explained
Why code pages were removed from mIRC

I don't know why you say this is not a bug. Certainly this isn't intended behavior restricting certain set of characters. Or rather unintended behavior that happens on certain combination of data.

It is a design flaw.

Just to clear up further confusions, I can send/read these characters perfectly fine between other clients. Meaning there is no issue during that exchange of data. It just happens when I'm trying to join or create a channel.

Please read the two posts both me and Raccoon have posted. This is not a design flaw, this is quite intended.

The reason you are able to read the characters correctly between two mIRC(v7.x)'s is b/c the server doesn't alter the underlying binary data. mirc #1 sends data to server -> server looks only at the part it needs then sends data to mirc #2 -> mirc reads unaltered data from mirc #1. Try your test again with one of the mIRC's being 7.x and the other is 6.x.

And here is a bug report detailing your exact problem, and the developer of mIRC(Khaled) stating that this is intended.

You are confusing yourself, as I stated multiple times above, this has nothing to do with reading characters between two mirc clients.

I have zero problems talking to other clients who use mirc 5.x. I am connected through UTF8 server and those others are connected from other server that supports EUC-KR charcode set. THIS IS NOT THE PROBLEM I AM DESCRIBING.

The link you posted has nothing to do with the issue above.
This is not a compatibility issue between clients.

It's a bug where any character with UTF-8 hexadecimal header that has A0 at the end that prevents me from joining or making a channel.

I believe this is actually a problem with the UTF8 server you are connected to. As stated earlier, character 160 is often used to feign a space. As a result, IRC servers have started to forbid this byte in channel names (regardless of whether it's part of a UTF8 character!), and the error you are getting is most likely the result of that check. You might want to talk to whoever runs the UTF8 server in order to disable the check.

Edit: just to make sure, this theory is easy to test, as the issue would be entirely mIRC unrelated: see if you get the same error when trying to join the same channel from another UTF8 IRC client.

You are right! It never occurred to me because I never used any other clients other than mirc.

I just tried few other irc clients and couple different web clients and found out they suffer from the same issue.