Can someone tell me what goes into the FormatSpecifiers of IOTAThread50.Evaluate? Or point me to the Helpentry if there is any. The Comment in XE talks about it, but i haven't any help installed right now.

ObjectPascal is case insentisive?


Yes but not the IDE!!!!

Working with XE7 and Berlin

refactoring a really big Delphi Application.

First I have done some Documentaion (XML Style) in common used Units;

But if you working in Source that uses a common function like

function ISearchhelper.mapToScreen(const Value : Tvec3d) : Tpoint;

and in the Unit it is used as

apoint := Searchhelper.MapToscreen(actPoint);

the Documentation Hint will not be shown...

The next thing are the Hints and Warnings

You can disable single Warnings but not Hints

so for the same reason you get a lot of (Sorry in German)
[dcc32 Hinweis] cad.Layer.pas(65): H2365 Schreibweise der Überschreiben-Methode TcadLayer.HandlEevent muss exakt ihrem Vorfahren TcadGroup.HandleEvent entsprechen

these made me mad.
I will not change the source only for make the IDE and compiler happy. To much differences in commit for compare


The IDE should help in refactoring process but it is a nightmare.

Also if the source could not compile, (normal on time during refactoring)
the IDE is unstable as Hell. If all compiles it is ok but what is a tool for, if it not work on refactoring?

the best tool for now is Notepad++








Really, really tired of fighting broken tools at every turn. And no, updating Delphi is not the answer. First, because in a major product, it's a decision not up to individual devs, and second because you trade one set of defects for another.

I am seriously overdue for this rant.

Error Insight is seriously defective, and has been for years. The red squiggles are often incorrect, anyway, and it seems it never gets any improvements. I can only wonder what other effects it contributes to the instability of the IDE.

I can't seriously consider it a feature of the product.

Post has attachment
When Did This Happen?

The last time I checked Interbase was foolishly expensive at about $4400 for unlimited users on 8 cores. I checked this evening and now it's ludicrously expensive at over $7000!!! Did Idera do this or what?

When it was $4400 you could get a commercial version of PostrgreSQL, with many advanced features lacking in Interbase such as windowing functions, stored procedures in about a dozen different languages, a variety of replication features, foreign data wrappers to interface with outside data as if it were a database (e.g. query your GMail or Google Finance), scalable read performance to at least 64 cores, serializable snapshot isolation, etc. along with EnterpriseDB's commercial Oracle compatibility tools for about $400 less. You also got 5 years of support vs. 3 and a proven track record of significant yearly feature updates and a community with thousands of open source plugins.

But at $7000, Interbase is now absurdly uncompetitive. When did they give it a 75% price increase? When it was at $4000 Tim del Chiaro couldn't explain what the buyer would look like (falling back to something along the lines of "Those who see value in our product will buy it and those who don't won't" when asked to explain where interbase fit in with Oracle on one end and open source on the other). How does one make a case for paying $7000?!?

Are they intentionally trying to kill of Interbase or what? They might have been milking locked-in Eastern European users before but at $7000 it seems like they intend to shoot the cow and make steaks. :-(

Post has attachment
The importance of ^
In both screenshots, I just clicked on Inspect.

><
Photo
Photo
Delphi 10 Seattle
2 Photos - View album

OK, so EMBT's been sold, so I guess now marketing can rest? David Millington's Webinar STILL (3 weeks after the fact) not linked online, nor is Nick's from a week later.

Annoying. Apparently they assume that as they have multiple live viewing times, one of those will do for everyone.

There are some entries in Quality Portal about enabling/disabling the Code Navigation toolbar:

https://quality.embarcadero.com/browse/RSP-12398
https://quality.embarcadero.com/browse/RSP-12511

And guess what:
"Closed", "Works As Expected".

<quote>
Comment from internal system: Yes in XE8 the Editor Navigator Toolbar should be visible or hidden, but it has been changed. It works as expected now in Seatle 10. It will work in this way in the next versions
</quote>

English is not my native language, so perhaps I don't understand that comment. In XE8 there was a working option to enable or disable that toolbar; that option was removed in D10, and it will not come back with an update? 

And BTW:
Even Embarcadero stuff has problems with the product name "Seattle". See quote above.

Post has attachment
Am I being unreasonable in expecting FireDAC explorer (DX) to not fail with the following message:

"No peer with the interface with guid {9BB0BE5C-9D9E-485E-803D-999645CE1B8F} has been registered." 

... when you (attempt to) set up and then connect to a standard port 211 connection to a DataSnap server?  

Digression: This seems to perhaps have to do with the FireDAC app perhaps [still?] not having the unit IPPeerClient in the uses clause.  This was a required change some versions back in Delphi [XE3] see e.g. http://blogs.embarcadero.com/pawelglowacki/2013/01/10/39958 and http://www.developpez.net/forums/d1346815/environnements-developpement/delphi/bases-donnees/xe4-firedac-datasnap/ and http://www.devsuperpage.com/search/Articles.aspx?G=2&ArtID=139718 

I would contend that it should, at the very least, offer a better error message...  and ideally would support probing the Datasnap server to the extent possible/reasonable.  Listing procedures would seem to be a reasonable expectation, which is indeed what Delphi itself allows you to do with the built in Data Explorer.

Aside: I've managed to severally crash and/or cause access violations in it in trying to use it, -- am I just unlucky? This basically while adding/deleting definitions and so on.  

Further aside, I'd be curious to know whether you also first run into an issue whereby the FireDAC explorer looks in the wrong place for FDConnectionDefs.ini.  When I tried the above on my colleague's PC and he had this additional issue as well... I seem to remember having had this additional issue myself as well a long time ago and at the time fixing this via some registry hack.  I'm kind of amazed that this issue is still present, if indeed this is a general issue and not just some leftover from the various upgrades these PC's have been through.

Post has attachment
Why oh why aren't I authorised to get http://cc.embarcadero.com/Item/30378 ?

I have a registered version of Seattle (RS Pro edition) and it have been a good six hours since I first tried and when I tried a second time, which should be more than enough time to allow me... ARGH!
Wait while more posts are being loaded