Getting the ball rolling. Which VB6 projects are you working on?

[Niya]

There's a lot of new information to absorb about .NET concepts I've never heard of before which I need to adapt into VB6.

Not so much .Net as it is WPF. My guess is that you had to learn about flow based positioning, control templating, triggers etc? Well those started with WPF which itself is based on HTML's way of doing things. The idea is to create UIs declaratively rather than programmatically.

[VanGoghGaming]

Yes, all those and a lot more, information overload at its finest. There are too many acronyms to be honest, as far as I've gathered, WinRT is UWP, not WPF and there are quite a few incompatibilities between the two. I had to repeatedly tell Copilot to stop giving me examples with triggers as those don't work in UWP but usually there are workarounds for every incompatibility. I can't say I'm a fan of this declarative model but it certainly has its advantages.

And there's also the Binding concept that was new to me as well. Many properties of any control can be bound to some source and pull data from there automatically (and the binding can work two-way as well). Then there is the concept of Commands that can be used instead of events for some controls. Like I said, a lot of information overload...

[fafalone]

See I've been going the opposite way, flood of details for a different way of doing things for drivers. So when I look at that extremely abstracted high level stuff, it's more difficult to absorb when I've been immersed in handling all every byte of memory allocation and freeing myself along with all that other low level stuff.

[yereverluvinuncleber]

Not working on the RC5/6 desktop widgets at the moment until RichClient's capability for upgrade to 64bits is determined. Just a pause until I determine the sense of continuing in that direction. Don't want to get locked into a 32bit past when the future appears decidedly 64bits in capability with TwinBasic.

Digging into my old SteamyDock with renewed vigour, sorting out all the remaining bugs (and there are lots of those), upgrading my code incorporating what I now know, removing all those slow GetPrivateProfileString/WritePrivateProfileString APIs and replacing the settings.ini with random access data.

Getting it to a state where it is usable for anyone else and not just me.

[luanshizhimei]

I used VB6 to write a small utility program for merging character portrait images in games.



Github project address: https://github.com/luanshizhimei/AzurLanePaintComposite

This might be slightly off-topic, but I'm particularly passionate about using VBScript to implement various small games.

[yereverluvinuncleber]

It is not off-topic. VBScript is near-as-damnit VB6.

[yereverluvinuncleber]

You must be thinking about converting your VBScript games to VB6 or TwinBasic.

[VanGoghGaming]

@VanGoghGaming: XAML controls in conjunction with the "Windows App SDK" would be the next step. However, I'm afraid that the "Windows App SDK" is too complicated for VB6, since it's actually only intended for .NET. First, you need to install the "Windows App SDK" in 32-bit and/or 64-bit versions. For VB6, you need to copy the installed files of the "Windows App SDK" from the "C:\Program Files\WindowsApps" folder, because some winmd files are required there. Then you need to download the NuGet package "microsoft.windowsappsdk.[version number]" and unpack it with 7-Zip, for example, and copy the DLL "Microsoft.WindowsAppRuntime.Bootstrap.dll" into the VB6 project folder. The unpacked NuGet package also contains some winmd files that must be extracted using the .NET tool "winmdidl.exe" to access the included idl files. The "Windows App SDK" can then be initialized and used via the MddBootstrapInitialize API of Microsoft.WindowsAppRuntime.Bootstrap.dll. All of this together is simply too complicated.

Hey Franky, have you tried to see if WinUI 3 controls can be loaded using the "registration-free" method as described in this blog post?

I have tried it with a WinUI 2 control, such as the NumberBox for example, by copying the "Microsoft.UI.Xaml.dll" file in the current folder and updating the manifest:

  <file name="Microsoft.UI.Xaml.dll">
    <activatableClass name="Microsoft.UI.Xaml.Controls.NumberBox" threadingModel="both" xmlns="urn:schemas-microsoft-com:winrt.v1" />
  </file>

Then the NumberBox control can be created successfully in code like this:

    hRes = RoGetActivationFactory(StrRef("Microsoft.UI.Xaml.Controls.NumberBox"), VarPtr(IID_NumberBoxFactory), NumberBoxFactory)
    Set NumberBox = NumberBoxFactory.CreateInstance

The object is valid, you can call its methods and set its properties but DesktopWindowXamlSource.Content simply refuses to render it. I am thinking that WinUI 3 controls behave differently in this regard since they don't need DesktopWindowXamlSource at all.

[-Franky-]

Hey Franky, have you tried to see if WinUI 3 controls can be loaded using the "registration-free" method as described in this blog post?

I haven't tried that yet. My first attempt with the Windows App SDK was the "Title bar customization" (https://learn.microsoft.com/en-us/wi...elop/title-bar) and how to get it working in an unpacked, non-.NET Win32 application. If I remember correctly, the Microsoft.WindowsAppRuntime.Bootstrap.dll is only included in the corresponding NuGet package and not in the Windows App SDK installation. At least I was able to customize the title bar and create new windows/dialogs, etc., using it. I was also able to create WinUI3 buttons, similar to WinRT. However, other WinUI3 controls only worked partially or couldn't be created. It might actually work better if you use a RegFree option here as well. I'm currently busy with other things, so I don't have time to write a few lines of code right now.

[yereverluvinuncleber]

I am enjoying testing TwinBasic, importing various projects, so far all are importing successfully, with some problems in the designer, but compiling and running well.



Above are my dock, the clock/calendar and a sample Panzer gauge chosen at random for migration. They all function at runtime. All running with TwinBasic and competently too. Each a little slower that its VB6 equivalent, noticeably on startup where all the gubbins is sorted out beforehand. In operation, very similar.

The designer, however had a wobble with recognising controls on the standard VB6 form with one of Krool's slider controls on the form. I can probably find an alternative method of adding one of his controls, done currently by a single OCX but when it comes to importing TB has an issue that I have raised here: https://github.com/twinbasic/twinbasic/issues/2211



It isn't meant to look like that.

[CrazyDude]

Hi all, been wanting to post in this thread for a while. This is a VB6 app project of mine. Useful for finding devices on your network and pulling info from them. Free for most home networks with up to 50 network devices.

The indicator grids at the top & main data listview is by 10Tec and the 3D Graph is from Codejock

Full Image

[Niya]

That looks very nice. Impressive.

[yereverluvinuncleber]

It does look impressive, so good to see these VB6 apps doing real-life jobs. I would imagine this is rather complex code. The name Slitheris does give a rather scary aspect though. It certainly doesn't appear to come from the better side of Hogwarts.

Reminds me of this:

[CrazyDude]

Thanks. Yeah, we’re looking to change it to simply Komodo Labs Network Discovery. We just haven’t gotten around to it. And yes, very complex. Many years of work. Probably why we haven’t been able to change the name yet.

It does look impressive, so good to see these VB6 apps doing real-life jobs. I would imagine this is rather complex code. The name Slitheris does give a rather scary aspect though. It certainly doesn't appear to come from the better side of Hogwarts.

[yereverluvinuncleber]

Tried it, tested it on VirusTotal first, all good. Very impressed by its capability to sniff out a network and identify the components so well. That could easily come part of my tool box. That is a credit to you and VB6.

Have you tried conversion to TwinBasic yet? You'll find that it will test TwinBasic's import capabilities and give TB some good technical feedback in the process.

[fafalone]

10Tec's iGrid was one of the first commercial products written in VB6 to be converted to tB with x64 support. (Also most CodeJock components have x64 versions as they were made in C++ and also designed to work in .NET)

So room for updates there too

[Niya]

(Also most CodeJock components have x64 versions as they were made in C++ and also designed to work in .NET)

Probably C++/CLI. I forgot that even existed.

[yereverluvinuncleber]

I have just knocked together a new "getting the ball rolling" thread solely for TwinBasic Projects, same rules as this thread.

You'll find it here: https://www.vbforums.com/showthread....you-working-on

From this point onward I will keep my TwinBasic project show-and-tells on that thread, please join me there. If we assume that TB is the natural successor to VB6 and that TB usage may grow, it makes sense to start that thread now and see how it grow in the futures. I will have plenty to show there.

Personally, I'd like to see some of the old theTrick, Wqweto, LaVolpe, Elroy and even Niya's old VB6 creations converted to TwinBasic and displayed there. Faf. of course will pay a regular visit, I hope.

I may drop a link here to that thread if something relevant to VB6 pops up.

[CrazyDude]

Tried it, tested it on VirusTotal first, all good. Very impressed by its capability to sniff out a network and identify the components so well. That could easily come part of my tool box. That is a credit to you and VB6.

Have you tried conversion to TwinBasic yet? You'll find that it will test TwinBasic's import capabilities and give TB some good technical feedback in the process.

Thanks. Glad you like it!

Yeah, went through a lot of trouble for the software to pass anti-virus checks. The EXE EV (Extra Validation) cert definitely helps.

Haven't gotten to try TwinBasic yet. I keep hearing about RAD Basic too. Could see those 2 devs working together.

Like fafalone said, 10Tec has worked with the TwinBasic dev to get their iGrid to work well with together. Highly recommend 10Tec's iGrid.. been using it since its creation in the early 2000's.

[yereverluvinuncleber]

Haven't gotten to try TwinBasic yet. I keep hearing about RAD Basic too. Could see those 2 devs working together.

The two chaps/teams on TB/RB are working using different approaches, one seems to be considerably more advanced than the other. At the moment, I have some fairly complex programs that work out-of-the-box with TwinBasic so I would strongly suggest you give it a try.

[Episcopal]

The two chaps/teams on TB/RB are working using different approaches, one seems to be considerably more advanced than the other. At the moment, I have some fairly complex programs that work out-of-the-box with TwinBasic so I would strongly suggest you give it a try.

What's RB doing? I only hear from TB. I recently saw a video on YouTube showing RB using MDIForms.

[Niya]

RADBasic? People are still pinning hopes on that fiction?

[fafalone]

RB is never going to be useful. It's barely a step above Hello World apps. You still can't call APIs, use Variants, make UDTs beyond ones with simple integral types, set/modify binary properties like Picture in controls, or compile anything at all besides very minor changes to their included demos.

I consider it a scam as non-paywalled information is extremely misleading, going so far as outright false claims, as to it being still in the 'unfit for any purpose whatsoever' phase. You're required to subscribe to even see basic blog posts, subscribe at a higher level to get any access to development versions at all, and even higher than that to receive the 'privilege' of being able to report bugs.

If tB ceased development today, it would likely take a decade plus for RB to get to where tB was in early 2022, than another decade or two to approach tB's current state, assuming development progress at the same rate it has for the last 4y.

It's very disappointing because I think legitimate competition would have elevated both products.

[Niya]

It's very disappointing because I think legitimate competition would have elevated both products.

Getting older isn't all bad. Sure, you get aches and pains, but that buys you wisdom. As little as three years ago, I would have wholeheartedly agreed, but now I understand that this notion is nothing more than propaganda we've been fed all our lives so that we not only give the cutthroat, morally bankrupt, predatory practices of greedy capitalists a pass, but we worship them for it.

Anyways, I don't really want to get into all that. Point is, TwinBASIC has everything it needs to succeed: A very passionate userbase with people like yourself doing everything you can for it to succeed, and a beast of a developer who seems unstoppable in his determination to make a production ready VB6 replacement a reality. You guys have everything you need. A cooperative effort of passionate people ensures that any endeavour can succeed. tB doesn't need a competitor. It just needs the people behind it to keep believing in it, and as far as I can tell, it has that.

The only thing that is of concern is the potential for work on tB to be stalled on account of something happening to Wayne, like sickness or a family tragedy.

[SearchingDataOnly]

I didn't follow RADBasic anymore.

My only interest in RADBasic is that it is developed with .Net(C# or VB.Net). IMO, it is much easier to develop a VB6 alternative with .Net (C# or VB.Net) than with C++. It's just that for those who want VB6's successor to be cross-platform, .Net isn't what they want. Maybe Microsoft has made .Net some cross-platform features, but I don't seem to see anyone developing with .Net on MacOS.

[Niya]

You know to be honest, I've never really bought into the idea that VB6 could be truly cross-platform. I've heard both Olaf and Wayne claim that it is possible, and they are certainly quite talented enough to pull something like this off, but I still have this nagging feeling that when that time comes, it will be far more difficult than they expect.

.Net's approach to start from scratch I think, was the better path. But I could be wrong. I can't wait to see how Wayne will do it. That will be very eye-opening if he actually succeeds.

[yereverluvinuncleber]

Not a better path for us, the VB6-ers with all that huge codebase and nowhere to go - but as YOU say, let's not dig into that here. Let's get the thread back on track for VB6 Projects that we are you working on.

So, Niya, when are we going to see one of your old VB6 projects converted to TB? Let's take this conversation over to the other thread as you make progress.

[yereverluvinuncleber]

So, after my failure to import my old social media platform, Firecall (built just for the two of us, my mate and I, using different languages on different platforms), I decided to give the code a long-needed revamp.



I hacked this thing together using VB6 and Photoshop in less than a week and achieved rather a lot in a short time, it was a birthday present. It is however, quite complex code in parts and the problem is I hacked it with no concern for readability nor maintenance. Lots of code was duplicated and altogether it was a general braindump with ever-present bugs.

Failing to migrate it to TB made me look at the code and begin to tidy it with some solid bugs to address - Making progress on the former.

It uses Dropbox for the inter-user network and encryption, it writes utf-8 files for compatibility with Mac os/x and other Unix systems, it handles attachments and emojis, records and transmits voice messages and displays/transfers all types of images. Not meant for general use, it is for fun between two old bonkers chaps with time on our hands and the source code to do what we want.

[yereverluvinuncleber]

Hi all, been wanting to post in this thread for a while. This is a VB6 app project of mine. Useful for finding devices on your network and pulling info from them. Free for most home networks with up to 50 network devices.

The indicator grids at the top & main data listview is by 10Tec and the 3D Graph is from Codejock

Full Image

Are you using an old version of RichClient5? 3,764kb (0047) vs 3,908kb (0067)

[yereverluvinuncleber]

I am just mucking around with a new but very simple one-image desktop widget. First one built using RC6 and my new standard codebase. It looks OK but it does not do much, just some eye-candy that can be assigned a function, like a big icon. All my older single-image widgets were built using RC5 and were basic in operation with old code. The good thing about the old RC5 widgets was that they all operated well on XP, supposedly! - I haven't tested this for a while so I am unsure of that last claim.



With that in mind I have created RC5 and RC6 versions of the same simple one-image widget. This template will act as an upgrade path for the older RC5 widgets.

I'm just playing. The point is that I am just about to start doing some testing on ReactOS new version (32bit) and so I will need XP versions of my widgets. If I can upgrade them at the same time then that it is a benefit.



I have converted them both to TwinBasic and they migrate without issue. I have some conditional compiler statements for working around known TwinBasic issues but they are trivial.



I have an XP machine dedicated for the purpose of testing, I'll test there first and then test on ReactOS in Virtualbox 5 later.

[yereverluvinuncleber]

A desktop crop

[yereverluvinuncleber]

Now that Olaf is alive (again) I may have found my Mojo to create more programs and to delve into RC some more.

[Niya]

Err....It's nice to know that he is ok but based on what he said, it seems that he is completely burned out, something that is very surprising. I never thought he of all people could burn out. The guy is a beast. Nonetheless, I know what that is like and sometimes, you never really get back around to your unfinished work. You keep saying you'll do it tomorrow and before long it's been 15 years since you last touched your code.

Point is, I'm not sure this is good news of RichClient.

[yereverluvinuncleber]

Hmmmm, interesting. He is alive and he is here. He will recover from the burn out. I suggest small amounts of alcohol and fun.

Your thoughts are appreciated though and I should say that the offer of assistance to Olaf still stands. I that I think we have a quorum for a team that has the will and the capability to take that task on, the upgrade to 64bit, I mean.

If he is burnt out out as much as you suggest then a perfect way to get help is to take it.

Olaf, if you are reading this, we could offer to start work on the upgrade of your code to 64bit operation, if you'll let us have that code! We could start with the RC5 source if you were happy to release that alone and we could promise to limit exposure of your code to a select few. NDAs could be signed &c. It all depends upon how much you are prepared to release.

You know the people here that are technically capable of the task (not necessarily me) but there others that would most likely be willing to dive in, me included - and others better than me, of course.

We seriously do not want RichClient to wither on the vine due to lack of attention.

[yereverluvinuncleber]

Working on my single image RC5 and RC6 widgets. Making them rotate as required so the desktop trinket goes where you want at any size and rotation.

[CrazyDude]

Hi all, been wanting to post in this thread for a while. This is a VB6 app project of mine. Useful for finding devices on your network and pulling info from them. Free for most home networks with up to 50 network devices.

The indicator grids at the top & main data listview is by 10Tec and the 3D Graph is from Codejock

Full Image

Are you using an old version of RichClient5? 3,764kb (0047) vs 3,908kb (0067)

I've been pretty sick, so couldn't reply. :/

About RichClient5.. It's kind of a long story. Started using RichClient for the SQLite capabilities. It was left in to try out the direct 2D hardware motion features, which look awesome. Would definitely like to use it some how. Schmidt is an amazing coder.

[yereverluvinuncleber]

We all stand on the shoulders of giants...

[CrazyDude]

Tell me about it. Without iGrid, the software I work on would've never been possible.. not to mention VB6, best language ever. And without RichClient, yours wouldn't be possible. Giants indeed..

[yereverluvinuncleber]

I've built all my designs previously using other languages, all of which were built by better men than me.

o Konfabulator by Arlo Rose and Ricky Romero built a runtime environment using javascript that was the perfect one for such desktop trinkets. It died.
o A Chinese chap by the name of Tony built Xwidgets development and runtime environment using Jscript which looked good for a while. It died.
o I'm now building using VB6 and RichClient, VB6 built by some clever chaps that knew what they were doing. However, VB6 has already been killed and is probably on the last gasps of life support.
o TwinBasic, by God it is nice to be working with something that is new - whilst I remain clinging to the hope that RichClient is about to have some new life injected into it yet.

Just because it is built by giants doesn't mean it can't fail.

[Niya]

It died....It died....It died.

That's always the risk when we invest in niche technologies.

I've always wished I had taken a different path and gone heavily into C and C++ instead of sticking so devoutly to Microsoft and Visual Basic. I must have been out of my mind. Too late now, I don't have the gusto or the time to completely switch tracks at this stage in my life. I could dabble here and there, but I cannot truly change now. Had I done that, I wouldn't have to worry about whether or not MS will kill whatever technology I fell in love with and they killed a lot of good technologies before they had time to truly mature. Silverlight, WPF are two examples of good MS tech that died before their time. Meanwhile, you have C/C++ stacks pushing decades of longevity out here. I must have been crazy indeed.