Re: Re^4: Cleaning the Log

Replies are listed 'Best First'.
Re: Re: Re^4: Cleaning the Log by BrowserUk (Patriarch) on Sep 03, 2002 at 11:43 UTC
Also the simple degenerate case of the user typing '{BS}', but then, so would any solution that didn't involve modifying the keystroke recorder! Well It's better than the Abottoire, but Yorkshire!	[reply] [d/l]
Re: Re: Re: Re^4: Cleaning the Log by demerphq (Chancellor) on Sep 03, 2002 at 14:03 UTC
Actually neither my version nor yours had this error. The key issue here is that you cant just replace .{BS} with // as this allows other char sequences to "become" a {BS} element. The process of converting {BS} to a \b char and _then_ converting the .\b's resolves this issue. But you are correct that ultimately there is no way to recover what was actually _entered_ into the computer without knowing the input rules that applied at the time of entry. Basically IMO this entire task is an interesting game but ultimately a futile one from the POV of actually solving the problem. Yves / DeMerphq --- Software Engineering is Programming when you can't. -- E. W. Dijkstra (RIP)	[reply]
Re^8: Cleaning the Log by Aristotle (Chancellor) on Sep 03, 2002 at 14:11 UTC
Actually, you just brought me to the conclusion that this all was an excercise in futility because - duh, in hindsight - what happens if the user types the literal string `{BS}`? Whoops. Makeshifts last the longest.	[reply]
Re^6: Cleaning the Log by Aristotle (Chancellor) on Sep 03, 2002 at 12:54 UTC
`perl -pe's/{BS}/\b/g;1while s/(^\|.)\010//' file` Makeshifts last the longest.	[reply] [d/l]


"be consistent"
	PerlMonks