Attila Kovacs

http://docwiki.embarcadero.com/Libraries/Rio/en/System.Rtti.TRttiType.QualifiedName "QualifiedName is only available for public types declared in the interface section of a unit." FQN makes only sense for public declarations = interface section, so no FQN for dpr members, and no lookup-table entry for those non-public types. TRealPackage.MakeTypeLookupTable makes it clear as it starts with "units := GetUnits;". At the end, we are still discussing the same what was said to me 3 days ago.

found by accident

Ignoring Mahdis corrected answer (which was actually the answer, he nailed it, even though my question was very lazy and broad) and acting like a spring, chanting the same in reply to his first comment is nothing else just trolling and does not help here.

There is a qualified name, the RTTI lib looks for <projectname.Type> in case of "FindType", however TRealPackage.MakeTypeLookupTable / GetUnits returning only units, not the dpr itself (it's just not in the the PPackageTypeInfo). So yes, there is RTTI but no TypeInfo. Maybe because (as you mention) the <projectname.type> is not a valid qualified name or just forgotten or because it's in the implementation section and does not belong to the PackageTypeInfo anyway (which I think is the reason). But to get a resolution to this, strictly answering my question you are right. And the answer to my problem was what Mahdi said. I could not ask better without knowing a lot of implementation detail. Thank you all.

The question is answered as far as I can see, Ctx.FindType(TypeName) fails on types declared in the dpr, doesn't matter what directives I put there, @Mahdi Safsafi's answer is very plausible for me, with not having interface section, no public pool for them. Not a big deal, just the usual lazy prototyping in a console dpr... thx

I have to disagree, it got much better. Sorry for the interruption, please continue.

You could play around with the LockWindowUpdate(Form.Handle); / LockWindowUpdate(0); and/or with the Form.Perform(WM_SETREDRAW, Winapi.Windows.wParam(LongBool(False)), 0); / Form.Perform(WM_SETREDRAW, Winapi.Windows.wParam(LongBool(True)), 0); pairs. I'm using them in OnDPIChange. It will also give a delay for showing the form.

Then it must be something other than Delphi.

I can remember. It's every time Dmitry.

ctrl+alt+f12, not the best hotkey though.

Doesn't this network license cost you mobility? What about notebooks, travelling, home office, etc..? Do you carry a shadow VM with your notebook which runs in the background? What is EMBT's problem with a 20 year old application? They should have make it free/abandonware long time ago if they don't want to take care of it anymore. But wait! There was a campaign in July 2015, buy Delphi XE8 and get an old version free of choice for free with it! Including Delphi 7! So what exactly is EMBT's problem? Nothing, which could't be managed in 5 minutes with some internal changes.

Ah, I see, but, Picture.Assign in this case calls TGraphic.Assign(Source) where source is your "jpeg", and it assigns by reference, so you must not free it, it will be free'd on reassigning or destroying the component. As far as I can see in the sources.

you changed Picture.Graphic.Assign(DataLink.Field) to Picture.Assign(...); ?

https://lmgtfy.com/?q=GDPR

after accidentally turning off the "View references before refactoring" option on the form. (ctrl-shift-e) btw. if ctrl-shift-e stops working just trigger the context menu of the IDE code editor once. (right click)

"And I have CommonProjectDefinitions.pas where all types and common methods are defined." You either stick with this "very oldschool" method or you explode this unit separated by the modules using it, but splitting it after public/bird/dog/whatever is definitely a way to nowhere.

you have to edit VTV sources and before every DrawText(W) do a DrawFormat := DrawFormat or DT_EXPANDTABS or DT_TABSTOPS; DrawFormat := DrawFormat or (FTabwidth shl 8); and you can create a property of FTabwidth, (plus one for Expandtabs yes/no).

I understand his curiosity. It's only matters if there was an error which was not handled but it should have been. In 10.1 and 10.2 (the last release I have) the above example does not yield any error, GetLastError returns 0. You have to step through the app and look where it happens if you wan't to know. "Is this a known issue?" As others has also mentioned, it's not necessarily an issue, but a state.

I don't know FMX, but normally a StringGrid is just a display and you sort the underlying data. But I don't know FMX.

turn off you disk cache and check it again

it's in the comments

@Kas Ob. Everything fine with your test. Reverted back to 'a'+c+'xxx..' and the results are same as yours. However, changing 'xxx..' to something longer changes the game. It performs better on larger memory chunks, but therefore also uses a lot more RAM. Btw, now I understand why was it slower on the first couple of runs, it did not reach the barrier. FMM5 performs more homogeneous, seamless transition between the block sizes.

@Kas Ob. Okay, I don't know how interesting is it. I've downloaded the SMM2 sources, changed the 'xxx..' stuff in the test dpr for some lorem ipsum cantus and my CPU is i7-3930K The two fastest runs: // Parallel For used : 8857900 ticks 32-bit // Parallel For used : 7113633 ticks 64-bit Edit: Under Berlin U2.

for me SMM2 is faster on x64 as on x86, also SMM2 uses much more RAM as FMM5, so it's possible that FMM5 could be tuned to use more RAM and even less CPU (?).

copy paste this post as a comment into your frame