In my upcoming assignment (the one that made me canceled my leave), I was asked to "verify" that a set of assemblies are indeed the compiled product of a set of source codes. *Gee! I'm a walking assembler or what?*
Anyway, here's how it can be done. You will need two tools - fc.exe (to diff files) and ildasm.exe (to disassemble a .NET assembly). You will also need the original source codes.
First, compile the source codes to produce the assembly (.dll or .exe). We will call this our source assembly. The target assembly will be the one already present. Next, use ildasm.exe to disassemble both assemblies and output the result to a file. Example:
ildasm /out=source.txt c:\source\MyAssembly.exe
ildasm /out=target.txt c:\target\MyAssembly.exe
To compare both files, use the fc.exe tool. Example:
fc source.txt target.txt
In the event where both assemblies are indeed produced by the same source codes, the comparison result will only show a minor difference on a comment line. It will be something like:
***** source.txt
.corflags 0x00000001 // ILONLY
// Image base: 0x00250000
***** target.txt
.corflags 0x00000001 // ILONLY
// Image base: 0x009F0000
Take note that the base address will be different on your machine. If no other changes are reported, then it is quite likely that both assemblies come from the same source code.
To compare the differences (i.e line-by-line) at a programming language level, I recommend Reflector and the Reflector.Diff add-in.
Why Not Do A Binary Comparison?
The fc.exe tool has a /b flag to perform binary comparisons. The reason why this will not work is because .NET assemblies are always different whenever they are being recompiled. Therefore, while a binary comparison on an assembly that was originally produced by the same compilation would yield no differences, a binary comparison on an assembly that was compiled on different compilations will reveal differences.
I would like to thank Tom Hollander for bringing up the disassemble and compare idea.
Popular Post
-
V12 / 5.576 cc / 385 PS / 406 ft/lb (550 Nm) @ 4.000 / 0 - 62 mph (100 km/h): 6,6 s / Vmax: 155 mph (250 km/h) (click images for a larger ...
-
4 x V8 / 27.204 cc (4 x 6.801) / supercharger / AWD / Vmax: 406,6 mph (654 km/h) / World Record 1960 (click images for a larger view)
-
I 'm currently preparing the second edition of my Application Architecture workshop which is a revision to the previous Application Arch...
-
Sept 1st and 2nd Ventura Nationals http://www.venturanationals.com/ Sept 8th Kustom Kulture San Diego Old Town 10 am 6pm http://www.fiestade...
-
on December 31, 2011, a little past two in the afternoon, on the corner of Bixel Avenue and Wilshire Boulevard as I waited in my car for the...
-
Most developers when developing Windows Communication Foundation (WCF) services, will take security for granted and deploy their web service...
-
Santa Monica AltFuel Expo was by all standard measure an overwhelming success. Thousands packed the hangar to meander through exhibits by c...
-
T he READY Worldwide Launch at KLCC Convention Centre on Tuesday was quite an exciting event for me. Although it was pretty tiring, I'...
-
Brass guns, top hats, and goggles... right on. Why am I posting about non vehicular movies? It's goofy fun gearhead maker stuff. Or, as ...
No comments:
Post a Comment