Leaderboard

Hi, just for you guys to weigh in, we have located a bug in the Delphi 10.4 compiler, that removes a totally valid assign statement in RELEASE mode, believing, that the code will never be executed. The actual method is from the mORMot library in releases before last week (unit SynCommons.pas, method TSynAnsiFixedWidth.AnsiBufferToUTF8. A full discussion is available at the mORMot-Forum: https://synopse.info/forum/viewtopic.php?id=5520 . I have reported the bug to Embarcadero: https://quality.embarcadero.com/browse/RSP-30088 . Please check out my report and try to confirm and weigh in there. It should be reproducible. --- Note, the compiler will tell you (Hint H2077), that the value assigned is never used and thus, in RELEASE mode, with optimizations enabled, drop the code completely. Thanks for your time and support 👍

Check out how you can use Delphi's new Custom Managed Records feature to wrap C(++) APIs. https://blog.grijjy.com/2020/07/20/wrapping-c-apis-with-custom-managed-records/

Why, oh why is there no description on github to say what is so great about the project? Sigh... even the project that this originated from says nothing, except showing some screenshots. I guess we programmers are not marketers 🙂

I very rarely use with. Too many pitfalls.

Given that so much that is actually important is broken, this should be way down the list of priorities.

I use Spring4D extensively (but not all of it obviously) - Base and Core and Extensions compiled fine for me with Update 2.

I can't really say, I installed it, my code failed to compile and I did something else for the weekend. I would love it if the debugger was improved but haven't got to that point yet, perhaps others can chime in.

Spring4D is not negatively affected by Patch2

At Export() that 5000 records doesn't count because the report is prepared. How many pages are? How many items per page? There are also vector images like charts or EMF, WMF? I tested for example 40 pages (200 records with one image and 30 text items per record) with booth FR Export and printing to "Microsoft print to PDF". The speed is about 3 seconds for both on a i7-8700K CPU. This is only Export because I do this at preview where the report is already prepared. P.S. You are too scarce with details and want a suggestion.....

"with as do" is not a thing in its own right. There is with (expr) do. And here we see as used in an expression. We are just composing different aspects of the language.

Yes it did!

All problems relating to with could be solved by the compiler warning about such collisions.

I have written a blog article about this problem. https://blog.synopse.info/?post/2020/07/20/Special-Care-of-Delphi-10.4 Hope the issue is fixed soon in the next Delphi patch - after the holidays I guess. 🙂

It still does not work in 10.4

Details are important. Without accurate details, you might as well be posting "something is wrong".

Hi... ...I hope not. WITH is out of date. ...this has always been so. That´s the reason why i never had a WITH in my code.

I'd go ahead and install it, those debugger issues are a real pain. There's an uninstall for patch 2 if it does break your code. So far the only patch 2 specific compiler issue reported is related to using record constraints on generics where the record contains a managed type such as a string or a dynamic array. If you don't use record constraints on your generics (and you probably don't) then you're likely fine. To my knowledge the only library that's definitely broken so far is VSoft.Messaging. I haven't read anything that leads me to believe that Spring4D is broken.

@Kas Ob. I noticed that you take optimization very seriously. That's really good but just as an advice from someone that did, don't be driven too much. Behind that optimization door resides the evil ! In fact, I spent much time reading Intel doc/books, comparing different compilers (gcc, clang, ...). At the end I become more obsessional and less productive as I started paying much attention to my codes than what is required like taking some time to decide whether I should write if a = b or if a <> b.

That sounds a lot easier than it is. Especially since the last assembler code that i really understood was 68000.

ok, I've sorted this. I should have started with the basic application first. I needed to add /default before the /helloworld said I'd missed something obvious!