(Untitled)

[shallowness]

Argh, so the frustrating part of making this whole attack database thing isn't the actual work (though the HTML/PHP is tedious!), but the classification stuff. I keep reconsidering the classification scheme, and recoiling at my own work because I suspect I've done it wrong somehow, and I could do it better. I want to make it as useful as possible

Comments

[evan]

my intuition is that if you need the name of the table as data you're doing something wrong.

[shallowness]

heh. I don't need the table name as data actually (probably a bad example?), I need to put something in that column that's going to differ based on the source table it came from. But, yeah. I asked someone else and he said sorta the same thing...

I guess it would make sense to merge these tables together outright, and do all queries specifying that differentiating field as WHERE NewColumn="blah" condition. Hmm. Yes!

why such problems with ELFsh ?

[anonymous]

Hi there,

I've seen your post about a bug in the ELF header modification feature. You mentionned a problem about saving your result file, but I never encountered this problem with the one year old public version 0.51b3. If you could provide the original target binary and the elfsh script file (or log of interactive session) that provoked this bug, that could probably help to spot the problem.

--
mayhem@segfault.net
ELFsh crew