Clarion News

Clarion Product News

Search

Navigation

Subscriptions

Alpha Phase 1 completing

There have been three builds released to the Alpha group including the initial build. And there have been over 125 bug fixes, 35 new features implemented based on requests by alpha testers, and about two dozen adjustments made to improve usability of the IDE components.

We'll be closing down Phase 1 soon with the introduction of full Unicode support in the next build. Along with Unicode support comes the support for using any Clear Type fonts in ENTRY controls. Though we had originally announced in the Clarion roadmap that version 7 would have better support for Unicode (double-byte character sets), and we would phase in additional Unicode support in the subsequent Comanche release, we decided that Unicode, along with complete support for Clear Type fonts was too important to delay. With Vista shipping with Clear Type turned on by default, and old CRT monitors being rapidly replaced with flat panel LCDs we knew it would be an important requirement for Clarion developers. Coupled with the fact that the number of Clarion developers who require double-byte character support is rapidly increasing, made the decision an easy one.

Text rendering in Clarion entry controls with support for all the possible attributes for Picture tokens, Masks, Alignments, Fonts, Fore/Back colors, and Caret positioning required an implementation that preserves all of the aforementioned functionality, and it was a technical challenge and achievement to preserve all the required behaviors.

The core Unicode support was implemented using the Uniscribe API. Although natively supported on Windows NT 5.0, the Uniscribe DLL may also be distributed for use on Windows NT 4.0, Windows 95, and Windows 98-based systems. But because this DLL may not be available on older systems we also implemented fallback code, so in the event that USP10.Dll cannot be loaded the RTL uses the GetCharacterPlacement function. According to the GetCharacterPlacement function's description it can have problems with caret placement on old Windows versions for certain languages, but it's good enough in most cases and it's available for any 32 bit Windows version. Having support for Unicode implemented with the fallback code was extra work, but it means we don't have to apply any additional requirements to which systems which can run Clarion programs.

If you're interested in having an idea of what Uniscribe is all about then have a look at :
http://msdn2.microsoft.com/en-us/library/ms776469.aspx

On to another topic, the Project System is Clarion 7 has had a major overhaul. Though the underlying project system in version 6 (and prior versions) was quite powerful, it wasn't often used near to its potential. In version 7 you'll see a lot of changes for the better. Rather then try to describe them all here, I'll post a link to a movie that was created to jumpstart the Alpha testers into accessing some of the new functionality
http://www.softvelocity.net/videos/alphap1/MultiDLL/multiDll.htm

The new Visual Styles available in version 7 can be configured in code using the provided classes, but it's really easy to use a combination of Clarion 6/ Code generation / Templates, and the Clarion 7 RTL. That's exactly what this movie shows you how to do :

http://www.softvelocity.net/videos/alphap1/VisualStyles/VisualStyles.htm

There is a lot to show, and we'll be posting more movies. Watch for our next progress report via email and/or in this blog space, and thank you all for your patience and support.

As I've said before, be sure to keep an eye on ClarionMag, Dave Harms is publishing some in-depth articles on C7 and soon on Clarion.Net.

posted on Wednesday, March 28, 2007 7:52 PM by Robert Zaunere

# Unicode support and Microsoft compatibility @ Thursday, March 29, 2007 1:15 AM

Bob,

Thanks a lot for this good news about Unicode - we will at last be able to let our clients using Chinese or japanese caracters in their english or french clarion application be able to sell to their clients in the respective languages :)

One question though please in this respect of better compatibility : I am desperately trying to get my application certified Microsoft SQL 2005 and I have found out that all is perfect except one point :

Make any single exe in C6 ABC like 'hello wold' one and that exe will not pass the Microsoft verification tool : handles are used from the Clarion exe calling old API's (like the ones used for GETINI etc) which MS decided are obsolete!

Could you please look into this seriously as having our apps certified MS would definitely enable a MUCH wider market for us and for your Clarion baby <BG>

Merci again

JP

jean-pierre gutsatz

# re: Alpha Phase 1 completing @ Thursday, March 29, 2007 7:47 AM

JP,

Which certification tool shows the problem you mentioned, can you post a link to it?

Robert Zaunere

# re: Alpha Phase 1 completing @ Thursday, March 29, 2007 9:14 AM

Hi Bob,

Glad to note that the Alpha 1 stage is nearly over. Keep up the good work.

Any chance that CSP subscribers would receive the final Alpha 1 build when Alpha 2 testing starts since the IDE and RTL seem to be quite solid and near-production quality? That would give us all a chance to lessen the learning curve of the new IDE when it goes Beta or Gold.

Best Regards

S.Jayashankar

# Microsoft Verification tool @ Thursday, March 29, 2007 9:21 AM

Bob,

here is the MSDN link to it
http://msdn2.microsoft.com/en-us/library/aa480483.aspx

One has simply to check in the right pane
Handles
Heaps
Locks

plus turn on logging from the menu and then run the exe and check for errors ...

I can privately let you have all other infos if need be of course

Merci

JP

jean-pierre gutsatz

# re: Alpha Phase 1 completing @ Saturday, March 31, 2007 4:56 AM

Unicode is a very important enhancement (for me at least), thanks a lot!

Vadim Berman

# re: Alpha Phase 1 completing @ Monday, April 02, 2007 3:56 PM

Bob,

A question about Unicode - are you going to incorporate Unicode support in your regular expression processor?..

I mean those \u tokens that are able to identify non-Latin characters, etc.

And, while we're on it - are there plans to enhance the regular expression processor to more modern standard? In C6.3 it lacks support of {minCount,maxCount}, zero-width assertsions, and a lot of grouping constructs.

Vadim Berman

SoftVelocity Inc.