On 1/8/2008 1:58 AM, Colin Watson wrote: > After writing this, it occurred to me to look for newer upstream > releases. less-418 actually does address similar bugs, and Kenshi's > sample file works correctly with less-418. However, it does this by > disabling the search for CSI in UTF-8 mode, and it still steps back > through the string byte by byte, although it does now have a check to > avoid stepping off the start. The former seems odd (why shouldn't CSI be > allowed to work in UTF-8 mode? U+009B still means CSI according to the > Unicode database), and if you fix that then you still have to fix the > invalid sequence removal code. Stepping back character by character > seems more elegant anyway. > > Mark, would you consider the attached less-418.csi.diff?
Hi Colin, Thanks for sending this report and patch. I think your solution is more complete than mine, since it does handle CSI chars in UTF-8 files. I will incorporate your change into the next release of less. --Mark -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
