Lost Source Code (AREV Specific)
At 19 OCT 1998 08:12:47AM Jocelyn Amon wrote:
Is it possible to recover lost source code in some format by analysing the object code? Does software exist to do this?
Failing that, is there anyway to tell what routines are initiated from within a routine by analysing the object code? I know it is possible to view all literals used but is there a method of knowing whether they are routine names or not, by analysing any associated object code?
At 19 OCT 1998 08:22AM Andrew P McAuley wrote:
Such technology exists, though proving that you have the right to use it is generally the issue. Revelation Software have been known to help out with this sort of request in the past when it was appropriate.
If you need more details drop me an email and I can point you inthe right direction.
World Leaders in all things RevSoft
At 19 OCT 1998 01:35PM Dave Bennett wrote:
In most cases you can look at the last line of the object code for some of the code and hope you recongize your subroutine.
At 19 OCT 1998 02:17PM Eric Emu wrote:
Owing to a disk error some years ago, I accidentally corrupted my copy of the AREV editor. The only remaining option was to use edlin.com, arev's dump command, or debug.com to assemble program object code one instruction at a time. This can be done using Pcode or 8086 assembler. For examples \CD21\ is a DOS interrupt 21h \CD03\ is a debug int 3h \B84C01CD21h\ means your program exits with an error code of 1. After a while this becomes second nature.
You can remove any instructions that you don't understand with \90\ (in 8086 assembler objects) or \01\ in pcode objects. This usually simplifies the application code to the point where just about all of my AREV.EXE is patched as \90\ and all my pcode (and AREV's) is \01\. This results in massive speed improvements, maximum compression under PKZIP, and in the case of much of my code, not much drop in functionality.
To make it hard for people to patch your object code, never hard-code any literals as literals - always use something indirect like
A=char(15)
B=char(100)
LIMIT=SEQ(A)*SEQ(B)
LIMIT :=D'
instead of
LIMIT=1500D'
The encryption (from within AREV) of object code and templates or windows or menus is also an alternative to slow would-be hackers, or enraged moas.
Eric
At 21 OCT 1998 04:19PM Michael Slack wrote:
If you have lost source code and only have the object code and would like to recover the source code, I know of someone that does that for a fee. I've asked a question of this forum about recovering source code from object code or object code and middle code. One of the responses was someone giving me the name of a guy in Chicago (I believe) that do. He is still at the number I was given and will recover source code from Arev object files.
Michael Slack
At 31 OCT 1998 04:26PM Molly Moa wrote:
Encryption!?! So that's what you did! You're such an ostrich! I could tar and defeather you! Oh! I'm so mad!
At 12 NOV 1998 10:07PM S. C. Wilson wrote:
Yes we have a piece of software that we use to re-create
source code from $object code in versions of Arev from
1.16 through to 3.12.
We're willing to help you. Contact me: [email protected]
At 11 JUN 2000 10:06PM S Wilson wrote:
Yes we have a piece of software that we use to re-create
source code from $object code in versions of Arev from
1.16 through to 3.12.
We're willing to help you. Contact me: [email protected]