The Sinclair QL Forum

For the last few weeks(months), I have been doing an update for this decompiler

I have added support for some more Turbo versions I have acquired.

I have also rewritten a lot of the procedure and function handling, to improve it. So that the program now identifies procedure and function definitions. Giving DEFine PROCedure, and DEFine FuNction. Rather than just creating a generic DEFine PROC/Fun line.

I have tried to make the user interface a bit more user friendly, and made various other small improvements.

I thought it was about time I gave the program a proper version number, rather than just a release date. So this is now going to be Version 1.00

The supplied zip files are Windows zip's, intended for a QPC2 DOS drive. The only file that might be a problem on other systems is the Talent/Quanta assembler/monitor program which has a Qemulator executable header. But I have also supplied a QDOS zipped copy of this file in case you lose the header.

Thank you for this work. When I get back to my laptop, I'll be having a play!

Cheers,
Norm.

Here's another update to the decompiler. This fixes a problem in the SuperCharge decompiler with FOR loops

It was happy with a simple FOR x=1 to 10 STEP 2, but could not handle things like FOR x=1,3 TO 7,10

I'm amazed I have never hit this problem before, I implemented it in the Turbo decompiler, but must have forgotten the SuperCharge one.

I have attached a full update to V1.02, and just the just the files that have changed from V1.00 to V1.02

It looks like a great and solid piece of work! Great detective work too. I guess that was a bit of a challenge.

I just skimmed through some of the docs. The whole thing is too massive to merely take a casual interest in, but I'll know where to find it should I ever need it!

The number of Turboed and SuperCharged programs left to decompile must be an ever diminishing prospect. Any big ones in the pipeline?

Hi,

I have been disassembling the ICE ROM which is written is assembley language, which I am currently trying to understand the coding. For my own enjoyment... I do not get out much!

I have also been looking at the ICE application software, like ARTICED which is a Supercharge v1.16 compiled basic programme.

I have been asking Martin for help with the discharge of this software, which he has been updating Discharge to deal with Supercharged programmes.

I am learning how to use the software, so early days.

Derek_Stewart wrote: Sun Dec 01, 2024 6:44 pm like ARTICED which is a Supercharge v1.16 compiled basic programme.

This is the program that highlighted the FOR problem. Right after I uploaded the previous update.

I still have some ideas for further improvements, and I could do with doing an update to the Qliberator decompiler. As I have come across some programs compiled with older versions of Qliberator, that are silghty different than the version 3's.

I've spent a lot of time on this latest update, and I could do with a break from staring at it.

As far as programs waiting to be decompiled, I think there is still a lot of comerical stuff that has not been done (like the Digital Precsion stuff). That cannot be released. Some things I have used test the decompilers, but I have not made fully working decompiles of them.

Hi Martin,

I think the work you have done up to now is tremendous.

I would like to do some work on the decompiling of software, so will have a look at the compiled programmes I have to decompile.

Hi Martin,

Is the decompiler for Q_Liberator ready to be published? Then Dilwyn could put it on his homepage.

RalfR wrote: Thu Jan 09, 2025 5:10 pm Hi Martin,

Is the decompiler for Q_Liberator ready to be published? Then Dilwyn could put it on his homepage.

I have not got around to looking at the Qliberator de-compiler yet (working on another little project at the moment). I plan to stick a version number on it like DisCharge, And I have come across a couple of old Qliberated programs that are a little bit different internally to the ones I have come across so far.

This has highlighted an issue that the later compiled programs I can find the names of any externally loaded SuperBASIC extensions from within the compiled program. But in these older versions, I have to actually load these extensions into BASIC before decompiling. And the de-compiler has to search the actual BASIC name table to find them. At least I think is was something like that, it was a while ago that I was looking at it.