Hello all,

thanks for the great software MIRC Been using it for more than 6 years now.

However I feel uneasy reporting a bug on such a great IRC client, but it's been here over a year and it is still lurking in the last 6.35 version (i reckon it was introduced at the same time when support for so-much-wanted UTF8 was added).

Settings for the bug to be reproduced:
UTF8 selected, display&encode (it is a must for me as I use several languages in mirc). Default encoding is Western.

Short summary:
It all works unless I use tab for autocompletion :P
With the settings as above, that 100% reliably b0rks encoding in the text which has been already entered.

Steps:

• enter some non-english text in the input box
  
• switch input locale to english, enter beginning of some nick present in the current channel
  
• press tab and the nick is autocompleted indd, effectively b0rking already entered text in the input box

Again: that happens only if "Western" is used as default (that is what I have to use as I use two other languages beside English on IRC).

(and btw the reason why i used "nicks in English" is rather simple. Autocomplete never used to work and still doesnt' work whatsoever with non-english nicks, at least in UTF8 :P )