Breaking News, few guys in China wanna write a new basic language

Quote:

---

Originally Posted by stuck-n-past

Well this is not the first time I've made the existence of my IDE known to others, and while is causes a little bit of a buzz right off the bat, there really isn't much interest on a larger scale. Believe me, if I thought it was something that would take-off so to speak, I'd do something about it.

---

That's why we aren't swimming in a sea of open source projects: There are plenty of people who might like to use it...as long as somebody else writes it AND maintains it. As for committing serious time to something open source....well, that's a tougher sell. I use some open source stuff, to which I contribute financially, but I couldn't afford to contribute any meaningful amount of time.

While the code is a mix of VB and C, mostly VB, I can always port it to another language and at the same time support a different model other then COM. I know that sounds like a huge undertaking, but I've already written code that converts C to VB with remarkable accuracy, it would just mean going the other direction.

And one has to understand that the majority of code deals with loading, saving, editing, lexicon / interpreting, debugging and the like. So it's 'value' encompasses much more then simply being a VB clone.

So other then those times where I am between projects and choose to play with the code, it will stay archived on CD's til needed. I can always hope to find the time and resources to finish the project at a later date.

One might say, therefore, that you have taken your VB out to C, but getting back in may not be a shore thing.

Shaggy - One good things is that I'm always up for a laugh!

Stuck-n-Past

Why don't you use what you have to implement an and IDE for vbrichclient. I am not sure, But i would think alot of the work is done (frm file parsing etc).

Please make contact with Olaf and see if this is feasible.

I think most of on this board would like an vb6 alternative

WP

Quote:

---

Originally Posted by axisdj

Stuck-n-Past

Why don't you use what you have to implement an and IDE for vbrichclient. I am not sure, But i would think alot of the work is done (frm file parsing etc).

Please make contact with Olaf and see if this is feasible.

I think most of on this board would like an vb6 alternative

WP

---

Good suggestion although I am sure Olaf has his own plan, I know it takes like ages for Olaf's compiler and IDE to lift off but I am hopping it will see the light soon!!!

Quote:

---

Originally Posted by stuck-n-past

While the code is a mix of VB and C, mostly VB, I can always port it to another language and at the same time support a different model other then COM. I know that sounds like a huge undertaking, but I've already written code that converts C to VB with remarkable accuracy, it would just mean going the other direction.

And one has to understand that the majority of code deals with loading, saving, editing, lexicon / interpreting, debugging and the like. So it's 'value' encompasses much more then simply being a VB clone.

So other then those times where I am between projects and choose to play with the code, it will stay archived on CD's til needed. I can always hope to find the time and resources to finish the project at a later date.

---

I'd certainly encourage you to make your work available for the rest of us to try.

Am I correct in assuming:-
- you still need the VB6 Runtime
- you compile using the VB6 C2.exe

How do you handle the intrinsic controls that are part of VB6 ?

Quote:

---

Originally Posted by stuck-n-past

Well this is not the first time I've made the existence of my IDE known to others, and while is causes a little bit of a buzz right off the bat, there really isn't much interest on a larger scale. Believe me, if I thought it was something that would take-off so to speak, I'd do something about it.

---

You ever thought about putting it on something like GitHub? You could clean up the code, add some commenting and a few disclaimers and such and just leave it there. Who knows, maybe someday somebody might want to do something interesting with it. A lot of great apps, mostly free ones, were born as forks of something somebody open sourced. I think FireFox began as such. Android was born of Linux. Xamarin was born of Mono. The Entity Framework also began humbly as a small open source framework. It just doesn't happen as quickly as it would if a giant corporation was interested and can afford to put massive resources behind it for quick development. It can take decades before something interesting happens. But in general I think great projects like yours should be made open source, except in cases where you plan to commercialize it for your own ends. That's too much good work to just end up laying on someone's HD rotting away.

I thought I would include a couple more screenshots before cleaning it off my disk. I really wish this site allowed videos as it would be so much easier to see my IDE run instead of all the still-frame shots.

The form I have loaded in my IDE is the 'Tiny' test form which I use for repetitive testing of the some of the most basic functions.

All of the pictures are of the IDE operating in the single-step mode of debugging the Tiny Form's code.

The first picture is just prior to changing the forms Caption. The form's startup name is called Tiny Form

Attachment 153639




In the second picture you can see the caption has changed and the code is preparing to set the width of the form

Attachment 153641




Here you can see the Form's width has been changed and the code is using the saved value of the original forms width to restore it.

Attachment 153643




The Form's width has been restored and the command buttons text is about to be changed from 'Command' to 'Cancel'

Attachment 153645




The command button is now called Cancel and the Immediate window is about to be disabled.

Attachment 153637

You can see here that the immediate window has been disable and we are attempting to output text to it. The enable / disable feature of the immediate window is purely for speed purposes so one can leave debug statements in place but still be able to test what the performance would be if they were removed, or at least close to it. And then they can be turned back on without having go through a process of deleting and then retyping them.

Attachment 153649




Here you can see that the debug text was not displayed and next we will turn the immediate window back on again.

Attachment 153651




The immediate window is now on and a new text message is about to be written.

Attachment 153653




The message has now appeared and we are declaring a string variable called 'text'

Attachment 153655




This picture is showing the use of the intrinsic Time function being assigned to the text variable just declared.

Attachment 153647

In the next picture an attempt was made to leave an unfinished line of code and the IDE flagged it as an error. The lines color has been changed to Red and a Popup Message informing the user of the problem is displayed. The message includes a list of the allowable sub-commands.

Notice there isn't an OK button to click in order to continue programming. All one needs to do is to type another character and the message will be cleared. If the new text typed happens to corrects the syntax error, the code line will return to the proper color.

Attachment 153659




In this picture another error has been detected, but with so many possible sub-commands, the warning message has been limited. There is an option to expand messages to include lengthy responses if one so chooses.

Attachment 153661




You can't actually see the cursor on the screenshots, something I need to fix in my capture program, but the mouse happens to be hovering over the 'text' variable. This has caused the IDE to lookup the value of the variable and display it's contents in a popup. There are situations in the original VB where popup messages overlap one another, so I made sure there is plenty of space between message and the related statement.

Attachment 153663




Here again the IDE is displaying the contents of a variable, this time instead of a string, it's an Integer and it shows both the Decimal and Hex values. There is an option to include Binary as well.

If you look up a couple of lines in the code you can see some of the language enhancements available when assigning variables, similar to what you might find in C.

Attachment 153665




In the last picture as this is taking too much time without being able to use a video, you can see where I tried to exit from the IDE after making changes, without first saving the work. The IDE keeps a list of modified files and notifies the users as to what needs to be saved.

Attachment 153667

I know that some of the examples seem a bit cheesy, like changing the caption on the Form and Command Button, but once the code is in place and capable of gaining access to VB objects and their properties, then the code should be capable of modifying any of them. As long as the Library-Viewer (Object-Browser) is working properly, then the IDE can modify properties of any given object following the COM model.

If I had more time I could demonstrate many additional features, I wasn't able to even reach the part of the test code where there are examples of conditional statements or loops.

I did show a call to an intrinsic function, the IDE is also capable of calling API's with most of them being predefined. No more searching and pulling your hair out when you don't have the proper signature of a DLL call. Well unless it's one that's not defined, but I have assembled quite a large list.

And not only API's, Enums and Types are also built in so you don't have to go looking for these either. Of course you can always override any of them with a local deceleration.

I would love to show more of the IDE's abilities, but it's time to get back to the daily grind.

Quote:

---

Originally Posted by stuck-n-past

I really wish this site allowed videos as it would be so much easier to see my IDE run instead of all the still-frame shots.

---

The next time you have time, just post a video to youtube, et al (private if you wish) and link to it.

Hi stuck-n-past, could you show us how your IDE references third-party COM components? In addition, do you have a complete project developed with your IDE? Thanks.

topshot - while I tend to be a YouTube junkie when it comes to watching, I've never uploaded a video. That sounds like a pretty good solution and yet the thought never even crossed my mind - duh. It's like multiple choice for me, senility, dementia or Alzheimer's, take your pick, either way my mind's definitely not as sharp as it used to be.

dreammanor - I'm not completely sure what your asking in regards to 3rd party COM components.

I treat COM Objects / Components all in the same manner, and all with the same functions and routines. As long as it conforms to the proper standards, the access is consistent across the board.

As for a complete project, I test with VB modules that incorporates every possible statement and construct available. It's a 'test-bed' of sorts for putting the IDE through it's paces with a predetermined data-set output / baseline.

After changes are made to the IDE, I run the test-bed project and it automatically compares it own output to that of the baseline, immediately letting me know if the changes were successful. It's a great way to insure modifications don't take the development effort in a backwards direction, and as to whether or not a bug, unrelated to the changes, was able to sneak in during the process. So I guess you could say the test-bad is a complete project.

Quote:

---

Originally Posted by stuck-n-past

It's like multiple choice for me, senility, dementia or Alzheimer's, take your pick, either way my mind's definitely not as sharp as it used to be.

---

I understand. :p

It's quite simple to upload to your account. Just make sure you record in at least 720p and your final product is in one of the common video formats they support. By default I think it will be public (able to be found in searches). You'll need to set it to private so the direct link is required to view.

Quote:

---

Originally Posted by stuck-n-past

dreammanor - I'm not completely sure what your asking in regards to 3rd party COM components.

I treat COM Objects / Components all in the same manner, and all with the same functions and routines. As long as it conforms to the proper standards, the access is consistent across the board.

As for a complete project, I test with VB modules that incorporates every possible statement and construct available. It's a 'test-bed' of sorts for putting the IDE through it's paces with a predetermined data-set output / baseline.

After changes are made to the IDE, I run the test-bed project and it automatically compares it own output to that of the baseline, immediately letting me know if the changes were successful. It's a great way to insure modifications don't take the development effort in a backwards direction, and as to whether or not a bug, unrelated to the changes, was able to sneak in during the process. So I guess you could say the test-bad is a complete project.

---

If your IDE could reference and use 3rd party COM components just like VB6, then you could use vbRichClient5 to enhance your IDE. We would like to know what effect is after using vbRichClient5?

dreammanor, I wish I had a better answer for you other then the vbRichClient5 should run and function under my IDE without any problem as long as it conforms to the COM model.

I don't have vbRichClient5 downloaded to be able to test with, and I have already removed my IDE from my development machine. I would love nothing more then to spend time on my IDE putting together all of the finishing touches, but it has a fairly low priority with it's relatively low ROI.

On another note, all it can be called is an OCD. I just can't relax and let things go when I know there's a problem. Ever since I brought my IDE back to life a few days back, the tiny-weenie toolbar icons were bugging me to death.

It's like being taken over by my alter ego telling me that I just had to know why. So after looking into it I found that a reference to the VB run-time files was broken causing an error. After correcting the invalid path, I took more screenshots of the toolbar which supports both an icon only view and an icon + title view.

So sorry, I'm sure by now your all completely done with the over kill of IDE screenshots. However... after correcting the toolbar icon problem, and doing the unthinkable of grabbing more pictures, I just couldn't upload them right away being leery about posting even more crap.

Then while deleting the IDE code, I was staring at these two lonely pictures and figured what the heck and well here they are. I promise, no more.

This is the toolbar view of icons only.

Attachment 153717




Here is the list with icons and titles

Attachment 153715

No offense stuck-n-past, but why do I feel you are just teasing us?!!
What do you have to loose for not to post the source code or open source it and let others worry about bugs/problems and you just go relax?!!
I do not see a point of removing your "IDE" from your development machine!!!!, do bugs float on the screen while it is there in folder with dust covering it?! or maybe it bloats your OS and hard drive causing your machine to perform badly?!!!

Again no offense here but I just can't help my mind "bugging" me with different feelings and conflicts about this!!

labmany - First I'd like to apologize to any and all that feel I shouldn't have posted information about my IDE, perhaps I made a mistake in doing so. My intention was not to upset or 'tease' anybody, but to demonstrate to all that creating a clone of VB is more then doable for anybody with the desire to put forth the effort.

What I choose to keep or remove from my PC is entirely a personal choice which is really nobodies business but my own. But if you must know, I perform nightly backups of my current system with daily restore points. By keeping down the number and size of files, the process not only runs faster but uses fewer resources. And keeping code on DVD is much safer then storing it on disk drives.

As for why I have not released my code to the world, I would kindly direct you back to post #22 of this thread.

Out of curiosity though, would you expect for Microsoft to make VB6 open source upon receiving an email request?

Microsoft is a giant with revenue beyond comprehension, able to absorb any type of financial hit from such a release, and yet the probability of VB6 becoming public is all but nonexistent.

I will however be more careful in the future before I make known the existence of code in my possession that is not to be made open source. I'm am not here to tease or upset anybody, and if I have, then I do apologize.

No need to apologize, and you are totally free to do whatever you like and whatever to your best interest.

Best wishes to you.

So are these IDE's an attempt to just replace the VB6 IDE? I mean, is it just to ensure continued existence of the VB6 language? I've heard of a few attempts of people creating their own VB6 compilers and such. Is the goal to also extend the language into a VB7-ish future? Was this already answered somewhere in this giant thread and I missed it?

Quote:

---

Originally Posted by loquat

there were few guys here in China that wanna write a new Basic Language.
And they have already making some efforts of it.

There were two groups of them.

Code:

---

group 1 wanna enhance the function of old VB6.exe, such as develop some extension add-ins or maybe try to take over vb6.exe
group 1 has a great effort of real inherit in VB6... they can inherit method and attributes of another class module unlimitedly. not just inherit simulation by using "implements".
They call this version VB6X

---

Code:

---

group 2 wanna rewrite a fully new basic language using VC6.0, the called "New Basic".
there will be below features:
1.Fully BASIC syntax compatibility
  will use BASIC syntax and will make some progress of VB6
2.Mixed programming, will be able to write almost each and every language in New Basic
  New Basic do not compile other language sentences, but leave compile work to the original language compiler
  and then New Basic will link obj file to make a execute file
3.the features that VB6 does not obtain such as, native multi-threading

group 2 has also make great efforts. they are now write New Basic IDE, and they have one guy to write compiler.

but this group have two guys only. one for IDE and the other compiler.

and they wanna communicate with anyone has experiences of rewriting c2.exe or hacking c2.exe or anything can help.

I know little VB6, I can not help programing, but i can help communication maybe.

if there is someone that is interested in what i said above, you are welcome to PM me or write to loquat520@qq.com

---

---

If your friends can just take and read the VB6 project files (vbp, frm,bas, etc. etc.) and translate the VB6 codes to their language of choice (C, C++, Pascal, etc.) and use their choice of existing compiler to compile the translated code, then this will be a great project. Then there is no need to have dependencies on MSVBVM60.dll and anything VB6. The final compile exe can also have a choice of x86 or x64.

Spent time on the translator instead of a new compiler when there are already so many existing. If the translator is successful, your friends see result immediately.

Maybe in the later stage, your friends can then think of developing their own VB6 work-a-like language (with enhancement) and IDE. Then at that time can wean off VB6 completely.

Another area your friends can look at is developing common controls that look exactly like the UWP XAML controls. This enable using VB6, and your VB6 work-a-like, to develop apps that look like UWP apps.

This and the translator will attract people to your friends' final products. They may even want to look at developing this for UWP too.

Quote:

---

Originally Posted by chosk

If your friends can just take and read the VB6 project files (vbp, frm,bas, etc. etc.) and translate the VB6 codes to their language of choice (C, C++, Pascal, etc.) and use their choice of existing compiler to compile the translated code, then this will be a great project. Then there is no need to have dependencies on MSVBVM60.dll and anything VB6. The final compile exe can also have a choice of x86 or x64.

Spent time on the translator instead of a new compiler when there are already so many existing. If the translator is successful, your friends see result immediately.

Maybe in the later stage, your friends can then think of developing their own VB6 work-a-like language (with enhancement) and IDE. Then at that time can wean off VB6 completely.

---

Quote:

---

Originally Posted by chosk

Another area your friends can look at is developing common controls that look exactly like the UWP XAML controls. This enable using VB6, and your VB6 work-a-like, to develop apps that look like UWP apps.

This and the translator will attract people to your friends' final products. They may even want to look at developing this for UWP too.

---

This is why VB6 successors have had such a hard time getting any kind of decent traction... everyone has their own idea of what it should do, which may or may not jive with the needs/wants of others, both on and off the project.

-tg

Tech, I couldn't agree more.

IMHO, if we could just get a totally open source basic IDE, p-code compiler, and machine code compiler that just worked with existing VB6 code (no "enhancements"), that would get traction. Again, IMHO, the p-code compiler would have to compile to exactly the same p-code, and the COM instantiation engine would have to work exactly the same way. In other words, a very good reverse engineering job.

I suspect the knowledge of how to do this is out there, but there's just not the willingness nor motivation to do it.

Also, yet again IMHO, it'd be cool if 99% of all of this was actually written in VB6. That way, existing VB6 coders could possibly be excited about helping out with the endeavor. In other words, it'd take a good copy of the Microsoft VB6 IDE to compile the open source VB6 IDE.

And, if (and that a HUGE if) that ever got done, then (and only then) we could start talking about enhancements.

But, who knows. Tilting at windmills again.

Y'all Take Care,
Elroy

Quote:

---

Originally Posted by Elroy

Tech, I couldn't agree more.

IMHO, if we could just get a totally open source basic IDE, p-code compiler, and machine code compiler that just worked with existing VB6 code (no "enhancements"), that would get traction. Again, IMHO, the p-code compiler would have to compile to exactly the same p-code, and the COM instantiation engine would have to work exactly the same way. In other words, a very good reverse engineering job.

I suspect the knowledge of how to do this is out there, but there's just not the willingness nor motivation to do it.

Also, yet again IMHO, it'd be cool if 99% of all of this was actually written in VB6. That way, existing VB6 coders could possibly be excited about helping out with the endeavor. In other words, it'd take a good copy of the Microsoft VB6 IDE to compile the open source VB6 IDE.

And, if (and that a HUGE if) that ever got done, then (and only then) we could start talking about enhancements.

But, who knows. Tilting at windmills again.

Y'all Take Care,
Elroy

---

Why exactly are you invested in p-code? wouldn't an out of process, real-time compiler that supports edit and continue be good enough?

As for COM, the bare minimum would be the easy part, it's only when you start supporting all the OLE Interfaces of Controls and Windows etc that it starts getting ugly.

Hi Dex,

Quote:

---

Originally Posted by DEXWERX

wouldn't an out of process, real-time compiler that supports edit and continue be good enough?

---

Personally, I'm not sure I'd know how to get there without p-code. I'm not the expert on the bowels of the VB6 IDE, but it's my understanding that, when you load a project, it basically does two things:

1) Loads and p-code compiles each module.
2) Loads each module as an ANSI file for editing.

And, it's my understanding (subject to correction) that it's fairly easy to keep the compiled p-code perfectly mapped with each statement in the ANSI file. And that's how it all works transitioning from design-time to run-time. Sure, that's the broadest of brushstrokes, but that's always been my understanding.

Therefore, I can't see how we'd ever have an open source VB6 IDE clone that did some degree of "edit and continue" without p-code.

Quote:

---

Originally Posted by DEXWERX

As for COM, the bare minimum would be the easy part

---

As I've personally stated elsewhere, I'd start getting very interested (and possibly willing to throw in my time) if I ever saw a bare-bones VB6 IDE clone. Ultimately, it'd need to compile my primary project (which is huge and uses about every trick we've seen). But I could get motivated if we just got anything reasonable going.

Best Regards,
Elroy

EDIT1: Actually, I'm not totally sure I need a true "edit and continue" function, as I really don't do much of that. I typically go back to design-time to do my editing. However, any IDE clone would need to execute-break-and-step-through-code. And, I guess I still don't see how to do that without p-code.

Quote:

---

Originally Posted by Elroy

Therefore, I can't see how we'd ever have an open source VB6 IDE clone that did some degree of "edit and continue" without p-code.

---

If there was ever a marriage between p-code and edit and continue, then this marriage died so long ago, it might as well be in another life. No Visual Studio version after VS6 has any problem doing edit and continue for any supported language. You can do edit and continue on executable applications being written in C#, VB.Net and any other .Net language implemented in these versions of VS. They execute native code, which is produced by a JIT compiler. They don't need a one to one mapping of single statements in the source language and a unit of executable code in a VM like VB6 and previous versions of BASIC needed.

I don't even think this idea is particularly new, even in VB6's time. I remember being able to break in C++ code being executed as it's compiled native code. However, I don't think I ever tried to edit code after the break. I mostly just checked values. So I can't say for sure whether edit and continue was supported somehow, at least from my own experience. And if it wasn't supported, you can be sure MS was already exploring ways to implement it without an interpreter.

TLDR; P-code and a VM to execute it is not necessary to implement edit and continue in a possible VB6 clone.

Even neighbor VC6 supports edit&continue and of course the debugger is out-of-process.

Anyway, writing compliers starts with writing parsers. VB6 lacks tooling for any type of parsers, no generators, no PEG, nothing. Even regexes are foreign land.

I'm dabbling with Ian Piumarta's peg/leg (mostly making unicode version and impl case-insensitive grammars) and think of fusing it with Mike Pall's DynAsm to produce a JIT compiling parser generator that can callback COM interface functions (sematic predicates, passing matches etc.) in grammar actions. Then probably porting the whole thing to VB6 (including embeded ASM) will be a viable option.

VB6 needs *several* grammars impemented. The language definately needs a two-pass compiler as in a module you can call procedures defined later w/ no forward declaration. First pass needs to collect symbols (mostly consts, UDTs, enums, classes, functions, global, member and local vars) in scopes as to be able to disambiguate language constructs, as parens are heavily overloaded in the language (e.g. what is calc(42) -- function call or array access). A separate grammar is needed for the preprocessor. Another for consts evaluation (which can involve module inter-dependency graph). A really exhaustive typelib importer for symbols in referenced typelibs is needed too.

Basicly building the frontend of the compiler will take some effort as tooling is non-existent, before an usable AST is produced for any moderately complex source project. Writing a dummy interpreter (directly on the AST) seems like a trivial task compared to implementing a reasonably complete frontend.

Also, producing a faithful AST has many bonuses, emulating tools like aivosto's Project Analyzer for instance, like `Find All References` tool in recent VS's, source code refactoring tools like the guys behind Rubberduck are doing for VBA (they are using ANTLR parser generator and are fighting tough battle with the grammar impl)

Backend is a long way to go. VM opcodes (as p-code is not well documented or understood), ANF then SSA form for any optimizations, low-end IR, then codegen, PE/obj file format, relocations, linking, let alone runtime impl, Forms subsystem, IDE, debugger. . . and so on.

cheers,
</wqw>

Quote:

---

Originally Posted by Elroy

EDIT1: Actually, I'm not totally sure I need a true "edit and continue" function, as I really don't do much of that. I typically go back to design-time to do my editing. However, any IDE clone would need to execute-break-and-step-through-code. And, I guess I still don't see how to do that without p-code.

---

You are in luck my friend - break points and line by line stepping are the defining feature of a debugger. Debuggers have existed long before (and outside of) VB6.
As long as you have the symbolic info generated by the compiler - you can correlate your line of code in any language to the machine instruction.

Attachment 154171


edit: Niya beat me to it...

Quote:

---

Originally Posted by DEXWERX

edit: Niya beat me to it...

---

Well I didn't explain the how, which you did. ;)

Quote:

---

Originally Posted by Elroy

Hi Dex,
However, any IDE clone would need to execute-break-and-step-through-code. And, I guess I still don't see how to do that without p-code.

---

You can have a debugger which makes use of 'compiled' source code coupled with Symbolic information to step line by line through code. Typically in this type of debugger, there is limited if any ability to modify the original source code.

Then there are interpreters, which in the true sense of the word, interpret the code without compiling it into machine code. In this model one can not only step through code examining memory locations, but can also modify the source.

My IDE is an interpreter that does not use machine code while developing and debugging. One can step through the execution of the code, examine variables and make changes. Once editing is done you can then continue with the execution of code.

So for a development environment running an interpreter, one does not necessarily have to compile the source into machine code.

The more I think this through, I don't really do editing while in "break" mode. However, I do very often check the value of various variables, and also "crawl back" and see how I got to the procedure I'm in. These things do help me a great deal to perform my Alpha Testing.

IDK, I suppose I've just always liked the idea of p-code. In fact, since everything I do is open source, I wouldn't even mind distributing my executable as p-code, although I never do.

Everyone Have A Nice Weekend,
Elroy

The problem with editing source code at debug-time is that the IDE does not allow saving these files back until ending the debug session.

Since using MZ-Tools to map its `Save File At Debug-Time or Run-time` command to `Ctrl+Shift+S` shortcut I found myself writing most of the code at run-time never fearing to loose a bug-fix. It's a major productivity boost for me as the REPL loop is instantaneous and IDE intellisense does not stutter when auto-completing var types or class methods on large multi-project code bases.

I'm also using VSS explorer outside of source control IDE plugin, so this allows me to check-out files at debug-time too, then edit the source in the IDE at debug-time and save the modifications back with MZ-Tools w/o terminating current debug session.

There are issues w/ forms and user controls that host controls which are currently debugged. All constituent controls have to loaded from binaries, not to lose design-time props when saving at debug-time with MZ-Tools.

cheers,
</wqw>

I have run into similar issues when debugging code, and after long sessions with many changes, there's nothing worse then experiencing a crash before you get the chance to save. There are times where I'll be sitting there staring at the screen in disbelief knowing that I have to click OK and watch the IDE crash without first being able to save. Once in a while I've managed to save the current module using CTRL-S and salvage some work, but it doesn't always work.

In the VMS operating system, there is a builtin file version mechanism which I implemented in my IDE just for saving development version of source code while debugging. All file types can be saved while in 'break' mode, including forms. As with VMS you can have up to 32,767 versions of a file and there are support utilities for managing the versions, purge, keep etc.

Each time you save, a new version is automatically incremented so it's quick and easy to save as many times as you want without the worry of coming up with unique file names. Once the debug session is complete, you can choose to 'purge' all of the unneeded versions with a single command, Or you can go back to any of the saved versions.

Sorry for waking up a relatively old thread, I am unsure as to the posting etiquette/rules here.

First of all I can't believe certain things in this thread:

1. stuck-n-past's IDE project - I can't believe that I missed it - I am ALWAYS on the lookout for a VB6-like replacement IDE, let alone a working interpreter/compiler. If it could be made language-aware I would use such an IDE for all my development.

2. I can't believe that stuck-n-past has actually made so much progress in creating what seems to be an almost-working IDE when others whom I had expected to come up with something similar have yet to produce!

3. I cannot believe that we are being teased with images of an IDE yet no binary is forthcoming to allow us to at least determine if it is actually 'real' - it isn't that I don't trust you stuck-n-past, it is just that we never believe anything until it is in our actual hands and we can squeeze it!

4. I cannot believe that yet another fundamentally beautiful project is closed-source and in the hands of one developer who has insufficient time and money to complete it! I have knelt and prayed for so many key one-man projects to come to fruition and NONE of them ever have. In all cases the well-wishing developer has either died, given up, gone on sabbatical, abandoned or just lost interest and refuses to make it open-source. The end result is that time passes, the project inevitably dies and everyone leaves dejected at such an opportunity lost. If a project isn't open-sourced and there is no organisation/team behind it then the chance of it surviving is low. That is the truth and it is a REAL pity.

(I have donated money to various projects that would have given me a IDE for Rapid application development, I donated $50 to KBASIC and the subsequent product that was replacing it. It still notionally exists but development came to nothing. I did the same donation for Xwidgets, which had its own graphical IDE with a choice of VBscript/JScripting but the developer there would not let his 'baby' go for various reasons and it also died a slow painful death, it still exists but is now unused by anyone on Windows.)

-oOo-

stuck-n-past - I won't argue your technical competency or ability to complete such a project but I will argue that it is too big a project for one man to complete in within an acceptable time frame. That should be obvious to anyone. To build VB6 MS could have required a team of 50 or more developers and the additional team to support users/ handle bugs and give feedback would have been potentially as large. You have made great in-roads into a VB7 but one man won't complete that project, that is certain.

To make it usable within a reasonable timeframe with achievable goals requires you to utilise resources that are available to you for free, namely other developers brains and time.

In my opinion you should do the following:

1. You could release a binary that shows proof-of-concept, that you are serious (I don't doubt it but it needs to be proven to everyone) and that your IDE works to a fashion...

2. You then open source the code post it on gitlab/hub and redact that commercial code that comprises those core items you don't currently wish to expose.

3. You then define the inputs/outputs and functionality that those missing modules are meant to achieve and then act as the project co-ordinator/tester, getting others to do the coding work that you have no time for. Create a mini requirements document or some pseudo-code for each so we can see what is missing.

(Getting others to contribute code is the thing here. I doubt they would even be minded to contribute if the code wasn't open source)

4. Accept contributions from people like Olaf even though it may seem like others are attempting to take control. I know that your IDE is your baby and it can remain that way, your involvement is up to you but remember, if you don't have the time then your baby will never grow. Take a 'snapshot' of your code and if you want it can grow slowly in your own hands, then you can do exactly that in your own time - just give others the chance to let it grow a little faster, you never know, it may grow into the child you always wanted!

-oOo-

Don't be scared of exposing your code to others, I have heard other developers state exactly that, they were embarrassed and let their baby die as a result. What you have built will stand up by itself, others will always be able to criticise and they always will, who cares? Just ignore and contribute. You gave birth to it, no-one can ever deny that.

That's it, that's my penn'orth. If I had chance to contribute to a project like yours I would, another $50 would be coming your way, perhaps in $10 dribbles as I saw my previous $100 contributions to other projects disappear and I don't want that to happen again. I would also contribute in testing, artwork if required, forum contributions with certain useful technical knowledge and general enthusiasm. I was a project manager for years so I know what is needed to complete a project, I could even bugfix if some of the IDE was actually written in VB6. I created many projects in VB6 in the past and would love the opportunity to use those old skills to contribute again in some new way. I have webdev skills in various CMS and the underlying technologies, PHP/.js &c - so a website and forum, bug reporting systems are well within my capabilities. Pull a team together!

The ReactOS o/s team is making leaps and bounds on their alternative o/s to Windows. I have always felt that ReactOS as an alternative platform for VB6/7 would be a winning combination. You could bundle your VB7 IDE with a compatible o/s that supports all the technologies that VB requires. The two could be a match made in heaven if only both would make progress. Some sort of unoffical tie-in would benefit both projects but you'd need to have something to show first...

Attachment 159707
That is VB5 running on ReactOS, not Windows.

stuck-n-past - if you let this project die my curses will sound loud and clear, I will be be snorting and shouting like a madman for a week. I can curse in many languages. I have just found your VB7 project - I don't want to watch it die by inactivity, keep posting images of its progress, whether others think it is teasing or not - and provide a feckless binary! if only to test - PLEASE.

GIMME, GIMME, GIMME, GIMME, GIMME.

PS. As a VMS System Administrator of many years I understand and appreciate your addition regarding the filenaming conventions that allow you to simulate version numbers for simple source version control. I always emulate this myself on Windows by copying/renaming to add a .1 .2. 3 suffix to all my source prior to making changes. The '.' suffix being an allowable character in a Windows filename whereas the ';' is not.

filename.vbp becomes filename.vbp.1, filename.vbp.2 &c.

PPS. I post teaser images of my own personal pet projects on the relevant forums just as you have done, there is nothing wrong with it - keep it going, some of mine are here as an example of me doing the same: https://www.rj-texted.se/Forum/viewt...hp?f=21&t=4411

PPPS. Sorry for the single swear word. I am an angry, passionate soul screaming out in this torturous Microsoft mediocrity... :)

It's not THAT old a thread, and seems like a good addition...except that avoiding the swear word censor by altering the spelling is still technically against the AUP, so I edited your post to turn your one swear word into something ridiculous...you can edit it to something that makes sense, if you so choose.

Stuck-n-past how much do you want for the source code?

What is it written in?

Is there a version I could evaluate?