Re: file recovery and format
> > Oh, well, that's one for the collective memory . . . and looking > > at the man page for strings doesn't provide any hint that there > > is a way to change what it considers a character. I suppose > > I should have written a script to get all 7 bit ascii characters . . . > Try the command "tr" for example: > tr -d :cntrl: > or something similar. Yes, but it's too late :( I saved the output of strings to work with, and went back to using hte computer . . . hawk
Re: file recovery and format
"Garst R. Reese" <[EMAIL PROTECTED]> writes: | > "Stupid is as stupid does" | Well, I work with a lot of handicapped people (aren't we all?) and have | become very aware of how a small deficit in hand-eye coordination or | visual acuity can lead to disaster. Unix was not designed with these | people in mind. It's general philosophy is "I'm the infallible | programmer, do what I tell you, no questions asked." | Rick, the stupid lawyer com bungling scientist, LyX contributor fell | prey. | Garst My underlying point is that an command or any nonliving object for that matter can not be stupid. User are stupid sysadmins are stupid. And the stupidity is not aloway that they use "rm" wrong, but that they has not created an alternative delete command (or whatever). Lgb
Re: file recovery and format
garst gabbed, > > | Neither idea addresses the simple fact that rm is plain stupid. lars lamented, > > "Stupid is as stupid does" > > Lgb > Well, I work with a lot of handicapped people (aren't we all?) and have > become very aware of how a small deficit in hand-eye coordination or > visual acuity can lead to disaster. Unix was not designed with these > people in mind. It's general philosophy is "I'm the infallible > programmer, do what I tell you, no questions asked." > Rick, the stupid lawyer com bungling scientist, LyX contributor fell > prey. shh, my students might hear :) The really odd thing about the whole adventure is that it came from an *extra* space, where I tend to have a problem with many keyboards getting a space in the first place. re: expansion I'm aware of how * is expanded, but that's not what got me (that would have been annoying at the time, but wouldn't have killed me). Is ~ expanded by the shell or the application? re: what I got back Andre noted that he recovers files the same way (yikes, I open this doesn't happen to him *that* often :) Though I"ve recovered enough thwacked partitions to have gotten good hat it [hint: do not *ever* run fdisk from drdos; it has a tendency to shift the whole table . . .). Anyway, I attached a zip drive, and used strings to get everything, for a total of about 40M on a 260M partition. There were certainly other lyx files around. Why didn't I find anything? As I'm going through what I recovered, I've noticed a couple of things: 1) The paper wasn't nearly as long as I thought :) This was one of those that fermented for months, and then came out in a long stream. I have more work to do than I thought. 2) The line breaks seem consistant with where lyx would place them. I get separate lines where math, emphasis, etc. would land. 3) And as I write this message, I found the problem. There isn't a single backslash in the 40 megs that strings produced . . . Oh, well, that's one for the collective memory . . . and looking at the man page for strings doesn't provide any hint that there is a way to change what it considers a character. I suppose I should have written a script to get all 7 bit ascii characters . . . rick, the stupid lawyer who was at least smart enough to get out of active practice :)
Re: file recovery and format
"Lars Gullik Bjønnes" wrote: > > "Garst R. Reese" <[EMAIL PROTECTED]> writes: > > | John Weiss wrote: > | > > | > On Mon, Jan 22, 2001 at 04:46:03PM -0500, Lior Silberman wrote: > | > > > | > > You should alias rm to be 'rm -i'. This is possible on almost all shells. > | > > | > Wherever I can use GNU-fileutils, I alias rm to 'rm -v'. This > | > prevents one from having to confirm every single delete, while giving > | > you feedback on what's going on and simultaneously slowing down the > | > delete enought that you can abort a bad one. > | > > | Neither idea addresses the simple fact that rm is plain stupid. > > "Stupid is as stupid does" > > Lgb Well, I work with a lot of handicapped people (aren't we all?) and have become very aware of how a small deficit in hand-eye coordination or visual acuity can lead to disaster. Unix was not designed with these people in mind. It's general philosophy is "I'm the infallible programmer, do what I tell you, no questions asked." Rick, the stupid lawyer com bungling scientist, LyX contributor fell prey. Garst
Re: file recovery and format
On Tue, Jan 23, 2001 at 10:20:08AM +0100, Lars Gullik Bjønnes wrote: > Lior Silberman <[EMAIL PROTECTED]> writes: > > | > Under unix-like OS, the * is expanded by the shell, not by the rm command, > | > so the rm command never sees the *. You'd have to hack the shell, not the > | > rm command, and the question becomes, how do you do that consistently? Do > | > you specialcase for a command rm with one of the arguments being * ? > | > | You should alias rm to be 'rm -i'. This is possible on almost all shells. > > And this can be equally dangerous. If you assume that "rm" really is > "rm -i" and to a "rm *" to select the files that should be deleted... > (and this has happend in realy life one of the reasons I really abhore > "rm -i") > > Lgb Truly, heed well Lars's words above. You just need to be verrry verrry careful with random "rm"s (and do backups). The only real "solution" is to implement a "trash can" concept, where "rm" really does not remove, but moves files to a temporary holding area for later removal. -- Kayvan A. Sylvan | Proud husband of | Father to my kids: Sylvan Associates, Inc.| Laura Isabella Sylvan | Katherine Yelena http://www.successlinks.com/kayvan | Reach your goals now! | Robin Gregory
Re: file recovery and format
"Garst R. Reese" <[EMAIL PROTECTED]> writes: | John Weiss wrote: | > | > On Mon, Jan 22, 2001 at 04:46:03PM -0500, Lior Silberman wrote: | > > | > > You should alias rm to be 'rm -i'. This is possible on almost all shells. | > | > Wherever I can use GNU-fileutils, I alias rm to 'rm -v'. This | > prevents one from having to confirm every single delete, while giving | > you feedback on what's going on and simultaneously slowing down the | > delete enought that you can abort a bad one. | > | Neither idea addresses the simple fact that rm is plain stupid. "Stupid is as stupid does" Lgb
Re: file recovery and format
Lior Silberman <[EMAIL PROTECTED]> writes: | > Under unix-like OS, the * is expanded by the shell, not by the rm command, | > so the rm command never sees the *. You'd have to hack the shell, not the | > rm command, and the question becomes, how do you do that consistently? Do | > you specialcase for a command rm with one of the arguments being * ? | | You should alias rm to be 'rm -i'. This is possible on almost all shells. And this can be equally dangerous. If you assume that "rm" really is "rm -i" and to a "rm *" to select the files that should be deleted... (and this has happend in realy life one of the reasons I really abhore "rm -i") Lgb
Re: file recovery and format
John Weiss wrote: > > On Mon, Jan 22, 2001 at 04:46:03PM -0500, Lior Silberman wrote: > > > > You should alias rm to be 'rm -i'. This is possible on almost all shells. > > Wherever I can use GNU-fileutils, I alias rm to 'rm -v'. This > prevents one from having to confirm every single delete, while giving > you feedback on what's going on and simultaneously slowing down the > delete enought that you can abort a bad one. > Neither idea addresses the simple fact that rm is plain stupid. Garst
Re: file recovery and format
On Mon, Jan 22, 2001 at 04:46:03PM -0500, Lior Silberman wrote: > > You should alias rm to be 'rm -i'. This is possible on almost all shells. Wherever I can use GNU-fileutils, I alias rm to 'rm -v'. This prevents one from having to confirm every single delete, while giving you feedback on what's going on and simultaneously slowing down the delete enought that you can abort a bad one. -- John Weiss "Not through coersion. Not by force. But by compassion. By affection. And, a small fish." -His Holiness, the 14th Dalai Lama
Re: file recovery and format
On Mon, 22 Jan 2001, Eric J. Schwertfeger wrote: > On Mon, 22 Jan 2001, Garst R. Reese wrote: > > > "Richard E. Hawkins" wrote: > > > > > > Several months ago, in one of those tragic typing accidents, I > > > inserted a space into "rm -r *~" on my laptop. I ceased using it > > Can't help you, but this is an obvious stupidity in rm. > > rm -r [asterisk] ~ makes no sense. I think rm should be modified to > > strip blanks after an *. > > Under unix-like OS, the * is expanded by the shell, not by the rm command, > so the rm command never sees the *. You'd have to hack the shell, not the > rm command, and the question becomes, how do you do that consistently? Do > you specialcase for a command rm with one of the arguments being * ? > > You should alias rm to be 'rm -i'. This is possible on almost all shells. Lior.
Re: file recovery and format
On Mon, 22 Jan 2001, Garst R. Reese wrote: > "Richard E. Hawkins" wrote: > > > > Several months ago, in one of those tragic typing accidents, I > > inserted a space into "rm -r *~" on my laptop. I ceased using it > Can't help you, but this is an obvious stupidity in rm. > rm -r [asterisk] ~ makes no sense. I think rm should be modified to > strip blanks after an *. Under unix-like OS, the * is expanded by the shell, not by the rm command, so the rm command never sees the *. You'd have to hack the shell, not the rm command, and the question becomes, how do you do that consistently? Do you specialcase for a command rm with one of the arguments being * ?
Re: file recovery and format
"Richard E. Hawkins" wrote: > > Several months ago, in one of those tragic typing accidents, I > inserted a space into "rm -r *~" on my laptop. I ceased using it Can't help you, but this is an obvious stupidity in rm. rm -r [asterisk] ~ makes no sense. I think rm should be modified to strip blanks after an *. Granted you might want something like rm -r a* b*, but even that is dangerous. There already exists a routine in fileutils to strip_trailing_slashes, I think strip_trailing_blanks_from_asterisks would be a good thing. But it seems that fileutils may have problems with file names containing blanks also. This would be OT for LyX except that LyX does not provide a delete file option, so users are stuck with rm. Garst
file recovery and format
Several months ago, in one of those tragic typing accidents, I inserted a space into "rm -r *~" on my laptop. I ceased using it immediately, and eventually used dd and strings to strip the hard drive, hoping to recover a half-written paper. Nothing *should* have been written to the hard disk after the castrophe, but I am finding two plain text versions, and no lyx files of the paper. The end of the longer paper may or may not be truncated by a postscript file. I suppose I could go ahead with just this, but does anyone have any suggestions? It seems to me that I *should* have a full dump of any ascii, but I'm just not finding the lyx files--or *any* lyx files, for that matter. grep -i "This..ile" xa* | grep -i creat comes back with three hits, none of which are what I"m after. rick
