Lazarus Release 4.0

[flowCRANE]

I think I've found a problem with IDE coloring - gutter, despite the RGB color set to 10,10,10, has a light gray background for bookmarks and breakpoint markers (see attachment).

Please report as a bug.

Done — https://gitlab.com/freepascal.org/lazarus/lazarus/-/issues/41616

By the way, thanks for the new version of Lazarus — I really like the new features. Good job, as always.

[440bx]

Among the new features listed is "Color config for assembler window".

Where is the color config facility for the assembler window, I've looked everywhere and I can't find it.

Thank you for your help.

[n7800]

Among the new features listed is "Color config for assembler window".

Where is the color config facility for the assembler window, I've looked everywhere and I can't find it.

Thank you for your help.

In the drop-down list among the languages, see the attachment.

[440bx]

I totally missed that.

Thank you @n7800

[n7800]

Thanks for the feedback. I added this information to the wiki so that others can find it too.

[jonyrh]

Great news! Thanks for the great work done!

[Martin_fr]

Thank you Martin for your precious help and for telling the true story.

You may want to comment on the issue itself, to get or give more feedback.
The forum threads are not as closely monitored, and your posts here may or may not be seen by everyone involved.

For me the issue "ended" with the creation of the bug, since the underlying code is not falling into my core areas and I got to many other things to look after.

[simone]

Thank you Martin for your precious help and for telling the true story.

You may want to comment on the issue itself, to get or give more feedback.
The forum threads are not as closely monitored, and your posts here may or may not be seen by everyone involved.

For me the issue "ended" with the creation of the bug, since the underlying code is not falling into my core areas and I got to many other things to look after.

Of course, the bug tracker is, by definition, the place to report problems. At the time, I didn't do it right away because I couldn't make the problem reproducible outside of my complex application. However, shortly after our posts on the forum, from which the solution you found emerged, I saw that Bart kindly opened a report, explaining how to reproduce the problem and providing an accurate analysis (by the way, thanks Bart).

I thought there was nothing else to do at that point.

As far as I'm concerned, I can most likely solve my problem by moving the initialization of some objects, from the OnShow event handler to the OnCreate event handler (which is probably also more conceptually correct, at least in my context).

I wonder if this is a regression and if, consequently, it will be solved in the next releases. Since the order of the events has been changed (now TForm1.Action1Update is triggered before TForm1.FormShow), I think it is a potentially relevant problem for many existing applications.

Thanks again Martin and Bart.

[Martin_fr]

Of course, the bug tracker is, by definition, the place to report problems. At the time, I didn't do it right away because I couldn't make the problem reproducible outside of my complex application. However, shortly after our posts on the forum, from which the solution you found emerged, I saw that Bart kindly opened a report, explaining how to reproduce the problem and providing an accurate analysis (by the way, thanks Bart).

I thought there was nothing else to do at that point.

As far as I'm concerned, I can most likely solve my problem by moving the initialization of some objects, from the OnShow event handler to the OnCreate event handler (which is probably also more conceptually correct, at least in my context).

I wonder if this is a regression and if, consequently, it will be solved in the next releases. Since the order of the events has been changed (now TForm1.Action1Update is triggered before TForm1.FormShow), I think it is a potentially relevant problem many existing applications.

It was fine to check it in the announcement thread first, after all it got attention, and at least narrowed down... (If it hadn't gotten attention, then you would have needed to file the bug)

The if/how to fix would be best discussed with Bart (not sure if he will see this thread, so that is why I pointed to the bug).

[flowCRANE]

I'm glad that it's finally possible to style the contents of the Assembler window, but there are so few styles that I've reported three issues with proposals to expand the coloring capabilities. Here are links to these issues:

https://gitlab.com/freepascal.org/lazarus/lazarus/-/issues/41619
https://gitlab.com/freepascal.org/lazarus/lazarus/-/issues/41620
https://gitlab.com/freepascal.org/lazarus/lazarus/-/issues/41621

Plus problems with the cursor shape for mouse links and with the current line highlight. This window is colored strangely, not very intuitive.

[calm_sea]

Congratulations, very nice version! I have windows and linux in dual boot, both with dark theme. on linux lazarus adapts to the theme, on windows it remains light. is this normal?

[Bart]

I wonder if this is a regression and if, consequently, it will be solved in the next releases. Since the order of the events has been changed (now TForm1.Action1Update is triggered before TForm1.FormShow), I think it is a potentially relevant problem many existing applications.

The if/how to fix would be best discussed with Bart (not sure if he will see this thread, so that is why I pointed to the bug).

IIRC we discussed it on devel ML.
AFAIK there is no offcial documentation about how this order should be.
So, it's not even certain it is a bug.

The behaviour however is not as it used to be, so maybe it should have been documented in the 4.0 release page on the wiki.

I opend a bugreport about it, so that would not be forgotten in case it was deemed to be a bug indeeed.

Further discussion should go to the bugtracker please (at least I will see it then, it's a mere coincidence I saw this post)

Bart

[Hansvb]

on windows it remains light

I had the same yesterday. You get choose force dark theme where you checked the dark theme. and then restart Lazarus.

[flowCRANE]

Congratulations, very nice version! I have windows and linux in dual boot, both with dark theme. on linux lazarus adapts to the theme, on windows it remains light. is this normal?

Yes, it is normal. If you want to have dark IDE on Windows, you have two options.


1. Install MetaDarkStyle package

As suggested by @Hansvb, install the MetaDarkStyle package, go to the IDE Options window, on the left side select the Environment/Dark style branch, in comboboxes choose Force dark and Dark respectively, save changes and reboot the IDE — from now the IDE will be dark.

This package is not ideal, it has few small bugs and glitches, but in overall it is great. I'm using it for a long time and I'm very happy that it is available to use freely. Of course you can use this package in you own projects, if you want them to have dark theme (or your own theme, with custom color palette).


2. Build Lazarus using Qt

You can open Lazarus project and change the widgetset to Qt. You will need Qt libraries to download and place them in the same folder as the Lazarus executable file, and add some code to the Lazarus project to enable dark theme. More work but it can be done easily. The downside is that Qt is slower than the native win32 interface, so the IDE will work slightly slower.

Some time ago I've tested this solution, Lazarus build with Qt widgetset. Worked fine, looked fine, but it worked slower so I've never tried to enable dark theme — I just went back to the win32 widgetset and MetaDarkStyle. If you need inspiration with changing colors for Qt application, check this post.

[zeljko]

Not qt widgetset (qt4), qt5 or even qt6 should be used.