A third-party Visual FoxPro decompilation program from China.
Not much is known about it so far except that there only seems to be a Chinese version.
Here is a .RAR download of UnFoxAll 3.0 from www.vfptop.com -- If somebody tries this, let us know how it goes :-)-- Steven Black
One report tells us that this is a Demo version of Unfox All Professional v3.0, and it decompiles Foxbase,FoxPro 2.x for DOS/Windows, VFP 3, VFP 5 and VFP 6. It's developed by Henan Yuneng Information LTD, with support contact at firstname.lastname@example.org
Well, I downloaded and installed the unfoxall pro version 3 demo and that was about it. I could not open files, I could thus not decompile things (my own build exe's BTW :-) )so bye bye unfoxall, no good for me...
First of all, unless there's something somewhere saying the company no longer care, this archive is warez and contains a keygen.
I'm not too bothered but maybe Steven (as host) wants to remove the link.
The program cannot open any files until the program is registered, although you can read the help files [which also claims decompiling of the same versions mentioned above.] Nice demo ;-)
To register you need to run the installer, then extract the keygen to the Unfox All program directory. Enter your user information as you have it in the unfoxall.ini and click Generate. This will add a working serial number to the .ini for the user details.
When you first run it you may want to click the language button and swap to something else. English is there (as well as an Engrish helpfile.)
I've only got FPW2.6a and VFP3 to test on.
I threw a couple of 2.6a .exes at it and it seems to have some problems:
It correctly shows all the source files and identifies FPW2.6 but when moving on to the decompile stage, the program hangs with an hourglass and 100% CPU load (but low priority? Doesn't seem to slow anything else down much.) The toolbar still works and clicking the "Exit" button causes an address violation. After that, clicking the window's close program dialogue does the same. Task manager kills it without any great pain.
Apparently it has decompiled all the reports correctly (they look OK and are editable in FPW) but nothing else appeared in the output directory.
VFP3 was more successful:
It seemed to decompile my .exe successfully (again not well tested. I suppose I could have tried to recompile it in VFP3 but I don't happen to have it to hand right now..)
It also created a couple of "list" like files, a .lst (text file of what was decompiled) and a .pj_ (similar but with a bunch of quotes and commas.)
Hi i'am Victor from Lima - Peru - 2006/09/08
Well i've tried Unfoxall with VFP7 exes even branded with password by REFOX and Unfox ALL decompiled ALL , also i used my system made with VFP9 SP1 and i got the same result; decompiled all the forms , programs , classes , reports, i am very scared , because this guys of china are very good , and i think we don't have any protection of this decompiler , i think the guys of refox must to do they homework and fix this because the branded level protection only work with refox but not with unfoxall. And i'm asking my self if the china guys know this... imagine is they decide make a new version.
Victor E. Torres Tejada
6/2/2009: I ran this on an exe built with VFP8. It was able to recover most of my project files except for the .pj_ files. However, it took several iterations of running the program to do so. It would display a memory error and the hourglass, but clicking the Exit button worked fine -- no need to kill it in Task Manager. I would then start the program again, uncheck any files it had managed to recover, and it would do a few more. Eventually everything came back excepting the .pj_. Their project rebuild did not work at all, but I was able to locate backup copy of those files.
The .prgs are rather mangled -- all comments are lost, everything is in ALL CAPS, and it does some very weird stuff reordering the SELECT and inserting some funky ???EClause[D7] stuff -- but fortunately I have backup of those too. It was the .scxs I had lost; it found those and did not mangle any of the code.
Still, I'm really glad this tool exists. The link above doesn't seem to work anymore, but I found it archived at:
04/01/2011: Hi, this is excellent!!! The tool works perfectly for VFP6 files. It extracted all my files in one go, even the .pj_ files. Follow the steps mentioned by Victor to register the tool by keygen. Without registering it wont open any files. Download it from the link mentioned above in http://www.woodmann.com. The original does not work.
Thanks everybody for your inputs.
04/28/2011: I used this tool to decompile VFP 5 exe's. Tool does a great got job of extracting programs, classes and screens. The only thing I have noticed is it does not extract the main program in the exe. It will list it correctly in the list of files but will not extract it. I tried it with multiple exe's and it won't extract the main file in each exe. Does anyone else have the same problem? Any workarounds?
It's possible that the EXE has been obfuscated with some other tool. -- Alan Bourke
2011-07-24: @Roger / Alan: No, I also tried it with several of my exe files and the main program could never be decompiled. I think this is due to a bug: if You have a look at the filetype in the list it says "other file" while all other prg files in the exe are properly identified as program files.
They list the main program as ".prg(Main file)" and obviously try to work with that extension.
I searched the whole web and found other installs as well. Internally, the ones I looked ad hold the same version.
But I found a Version 2.00 on a chinese page:
It's all in chinese and a little bit tricky to find the right links (have a look at the status bar)
The setup and EXE are chinese as well, but with a version 3 aside, You get along.
This older version properly decompiled my main program at least with the save as.. option
2011-07-25: Update: the main program is properly decompiled if You run Version 3.0 in simple chinese. It's obviously a bug in the "english" mode.
As it's based on an older VFP version, it seems to be unable to properly decompile newer language enhancements. I found an entry "???FUN[EAEE] (m.lcText,2)"
in the decompiled code that I could easily decode it to "getWordNum(m.lcText,2)", as it was my own code, but You should check for those (fails to compile though, so You will fall over it).
Also sometimes decimals are separated by colons instead of points (x = x + 0,25). Also not passing the compiler and easy to fix.
See also other Vfp Decompilers.
Category Third Party Products
( Topic last updated: 2011.07.25 09:59:40 AM )