New testing release of UNDELETE: undelete-20oct02
Eric Auer <[email protected]>
| Newsgroups | gmane.os.freedos.devel |
|---|---|
| Message-ID | <[email protected]> |
Hi all, I am releasing a new testing release of my UNDELETE tool. Changes: - fixed a bug that broke in accessing 16..32 MB sized partitions (?) - improved help screen and error message usability - changed the default size for all modes to autodetect, no need to specify 0 for "autodetect" anymore. Be warned that autodetect in follow mode may give you too long or too short recovered files! - added EXTRACT mode: While FOLLOW mode avoids used clusters and is used to recover deleted files (it even skips over used clusters: you can use cluster 2 and size 100 to recover the contents of the first 100 unused clusters on the drive), EXTRACT does the opposite and tries to stick within used clusters. Useful if you have a still existing but truncated file, I hope (will at least recover the part between end of the file and end of the cluster where the file ends) - made DIRSAVE mode (hopefully) autodetect if you try to recover/read a deleted versus existing directory and act appropriately. You can find the source (now with full GPL disclaimers and full text of the GPL included, sigh....) at the usual place: http://www.coli.uni-sb.de/~eric/stuff/soft/ undelete-20oct02.zip This zip file also contains the DOS (16bit) and Linux binaries, object files and Turbo C setup stuff. For DOS, you need NASM if you are not satisfxied with bigdisk.asm / bigdisk.obj as well, but for Linux, gcc -Wall *.c -s -o undelete.elf is enough :-). See the various included other information (howto, notes, relnotes...) as well. This is a TESTING release: I hope, but am not sure, that everything still works, and even more. Please try to figure out if everything does work as expected. The worst that can happen is that you will get a huge file with useless data instead of what you really wanted to recover. Or that the program crashes / hangs / loops. No error I can think of could cause any filesystem damage apart from the usual risk that a crash poses to your filesystem, so you should be very safe :-)). Have a nice day (I think I will, its my birthday the 20th), Eric PS Tom/Bart: It seems that http://www.coli.uni-sb.de/~eric/stuff/soft/ sysmem.zip and specials/tempfile.com still report only 64k available RAM for .sys files and still report the use of a file name extension with tempfiles with Kernel 2027te. Please do include the patches we have discussed for these issues with Kernel 2027. Thanks! [CCing a few people that are on the list - evil trick to make them get a copy of my mail outside their "mailing lists" folder :-)...]