Problem with some characters

demond · Post by **demond** » Sun Dec 19, 2004 9:22 pm

awyeah wrote:Taken from: eggdrop1.6.xx/doc/known-problems.txt

Things that are broken, but aren't getting fixed anytime soon:

* High-bit characters are being filtered from channel names. This is a
fault of the Tcl interpreter, and not Eggdrop. The Tcl interpreter
filters the characters when it reads a file for interpreting. Update
your Tcl to version 8.1 or higher.

* Version 8.1 of Tcl doesn't support unicode characters, for example, è.
If those characters are handled in a script as text, you run into errors.
Eggdrop can't handle these errors at the moment.

this should not have anything to do with Tcl - if that guy really doesn't use scripts for processing channel chatter, as he says; note that the docs say "from channel names", not channel chatter, channel messages or anything like that - for a good reason - the reason being that eggdrop doesn't process channel chatter at all, whereas it processes channel names (i.e. channel names are being submitted to scripts, at least to the config file, which is, after all, a Tcl script)

awyeah · Post by **awyeah** » Sun Dec 19, 2004 9:38 pm

But I do know eggdrops have problems with special character's. Well generally all ASCII [0-255] work. However with tcl scripts I've noticed you have to use their hex formats in html \x80, \xA2, \xFF and not themselves plainly.

As for non ASCII characters such as arabic, turkish, thai, chinese, japanese and etc the eggdrop could definately have problems, even the ircd.

demond · Post by **demond** » Mon Dec 20, 2004 12:33 am

awyeah wrote:But I do know eggdrops have problems with special character's. Well generally all ASCII [0-255] work. However with tcl scripts I've noticed you have to use their hex formats in html \x80, \xA2, \xFF and not themselves plainly.

this is irrelevant to that guy's problem, he doesn't use Tcl scripts and talks about channel messages, not channel names (did you read my last post at all?)

As for non ASCII characters such as arabic, turkish, thai, chinese, japanese and etc the eggdrop could definately have problems, even the ircd.

Tcl (8.0 and higher) internally uses Unicode for character representation, so there's no info loss in that regard, only sloppy script writers

tnd · Post by **tnd** » Mon Dec 20, 2004 10:05 am

If you need any information from me or do you want me to run some script for testing please let me know

I am really appreciated your help

De Kus · Post by **De Kus** » Tue Dec 21, 2004 11:34 am

I just played today a bit with Japanese input, and tried if I can write something Japanese in mIRC... (after I managed to find a suitable font I manged to get it working ^^).

eggdrop channel log wrote:[15:55] <StarZ|De_Kus> $B$h(B

local mIRC log wrote:[15:55:37] <@StarZ|De_Kus> よ

login wrote:Your host is irc.hes.de.euirc.net, running version euIRCd1.1.5

This looks kinda strange to me... (remember, local log is not precessed by the ircd). Local it is right, 1 Japanse character to 2 unicode characters, but the bot seems to get something more, but I dont know its mircs send mode fault ^^. If you want I can try on QuakeNet, too ^^.

PS: I asked one person in the channel how many chars he sees when I write 1 Japanse char and he saw 8(omfg!). Seems to be the fault of the ircd?!

QuakeNet wrote:[16:48:17] <StarZ|De_Kus> よ　なに　明日美
[16:48:23] <+Klinke> ????
...
[17:02:52] <StarZ|De_Kus> nur interessehalber... wieviele zeichen hab ihr da oben eigentlich gesehen?
[17:02:58] <+Klinke> よ　なに　明日美

after asking him how many chars he had seen he copied from my line and I could read it again.

[17:08] [StarZ|De_Kus!DeKus@De_Kus.users.quakenet.org] $B$h!!$J$K!!L@F|H~(B

this is the same line again in the eggdrop log... looks wired again.

PPS: to thoese who can "read" what there is written in Japanase... it was not supposed to make any sense, please don't ask about it

. Probably I should have pressed space in the first 2 words, too. I dont know if they would be used in Hiragana.. you see... I am just playing around

.