FWD » Core Development » User Interface

10/14/2020

05:06 PM Support #4955: web client/javascript drawing - fast vs crisp

If you provide a popup menu for the virtual desktop background for the performance test page, we should also add a "V... Greg Shah

05:04 PM Support #4955: web client/javascript drawing - fast vs crisp

> Please review this patch for @p2j.logger.js@ to enable its functionality to the virtual desktop. Can I commit these... Greg Shah

04:28 PM Support #4955: web client/javascript drawing - fast vs crisp

Please review this patch for @p2j.logger.js@ to enable its functionality to the virtual desktop. Can I commit these c... Sergey Ivanovskiy

03:01 PM Support #4955: web client/javascript drawing - fast vs crisp

Most rectangles are drawn with 4 lines instead of the rectangle primitive. There are quite a few non-black lines dra... Greg Shah

02:42 PM Support #4955: web client/javascript drawing - fast vs crisp

Constantin Asofiei wrote:
> On Chrome, when opening a widget-heavy customer screen, the @this.offscreenCtx.putImageD... Eugenie Lyzenko

11:47 AM Support #4955: web client/javascript drawing - fast vs crisp

Sergey Ivanovskiy wrote:
> Greg Shah wrote:
> > Sergey Ivanovskiy wrote:
> > > @putImageData@ doesn't take into ac... Sergey Ivanovskiy

10:59 AM Support #4955: web client/javascript drawing - fast vs crisp

Yes, that is OK. Greg Shah

09:14 AM Support #4955: web client/javascript drawing - fast vs crisp

Ok. Following this way I need to add new js module @p2j.perf_tests.js@ to @index.html@ page... Sergey Ivanovskiy

08:37 AM Support #4955: web client/javascript drawing - fast vs crisp

Sergey Ivanovskiy wrote:
> Yes, I am using p2j.canvas_renderer.js and p2j.strokes.js and p2j.js. Why does it requir... Constantin Asofiei

08:04 AM Support #4955: web client/javascript drawing - fast vs crisp

In regard to concurrency, I want to avoid it for now. Web workers come with other issues and certainly add complexit... Greg Shah

07:36 AM Support #4955: web client/javascript drawing - fast vs crisp

> This idea has been already proposed by Sergey, see #4785-831 and related notes. But I'm not sure how far he went wi... Greg Shah

05:19 AM Support #4955: web client/javascript drawing - fast vs crisp

Greg Shah wrote:
> Sergey Ivanovskiy wrote:
> > @putImageData@ doesn't take into account clipping regions. Thus, if... Sergey Ivanovskiy

05:17 AM Support #4955: web client/javascript drawing - fast vs crisp

Yes, I am using p2j.canvas_renderer.js and p2j.strokes.js and p2j.js. Why does it require to run this html page with... Sergey Ivanovskiy

04:41 AM Support #4955: web client/javascript drawing - fast vs crisp

Sergey Ivanovskiy wrote:
> Greg Shah wrote:
> > > what would you like to measure for our js web client. It can degr... Constantin Asofiei

04:30 AM Support #4955: web client/javascript drawing - fast vs crisp

Greg Shah wrote:
> > what would you like to measure for our js web client. It can degrade performance if we add perf... Sergey Ivanovskiy

04:25 AM Support #4955: web client/javascript drawing - fast vs crisp

I didn't go farther because these changes @screenImageWorkingCanvasDrawImageRevertedChanges.patch@ added regressions ... Sergey Ivanovskiy

03:39 AM Support #4955: web client/javascript drawing - fast vs crisp

Greg Shah wrote:
> > If I understand properly your idea this is a kind of "double buffering" concept. We have two bi... Hynek Cihlar

03:31 AM Support #4955: web client/javascript drawing - fast vs crisp

Greg Shah wrote:
> > If I understand properly your idea this is a kind of "double buffering" concept. We have two bi... Sergey Ivanovskiy

12:05 AM Support #4955: web client/javascript drawing - fast vs crisp

Greg Shah wrote:
> Crazy idea number 2: use "WebGL":https://developer.mozilla.org/en-US/docs/Web/API/WebGL_API/Tutor... Eugenie Lyzenko

11:18 AM Feature #4029: implement more widget rendering knowledge within the client drivers

I really like the idea of pushing the text editing down to the JS side. All of our text editing is very slow, I agre... Greg Shah

10:08 AM Feature #4029: implement more widget rendering knowledge within the client drivers

Hynek Cihlar wrote:
> Ovidiu Maxiniuc wrote:
> > BTW, yesterday, because of an incremental compile issue, I had a d... Hynek Cihlar

10:06 AM Feature #4029: implement more widget rendering knowledge within the client drivers

Ovidiu Maxiniuc wrote:
> BTW, yesterday, because of an incremental compile issue, I had a debug session with mouse e... Hynek Cihlar

10:05 AM Feature #4029: implement more widget rendering knowledge within the client drivers

Ovidiu Maxiniuc wrote:
> We live in the era of intensive streaming. Not only Hollywood productions but games also. T... Hynek Cihlar

09:59 AM Feature #4029: implement more widget rendering knowledge within the client drivers

BTW, yesterday, because of an incremental compile issue, I had a debug session with mouse events. I had a hard time t... Ovidiu Maxiniuc

09:48 AM Feature #4029: implement more widget rendering knowledge within the client drivers

We live in the era of intensive streaming. Not only Hollywood productions but games also. This year https://www.nvidi... Ovidiu Maxiniuc

09:30 AM Feature #4029: implement more widget rendering knowledge within the client drivers

Greg Shah wrote:
> Are there some "quick wins" we can achieve immediately using either the higher level primitives o... Hynek Cihlar

08:28 AM Feature #4029: implement more widget rendering knowledge within the client drivers

Are there some "quick wins" we can achieve immediately using either the higher level primitives or the recording appr... Greg Shah

08:26 AM Feature #4029: implement more widget rendering knowledge within the client drivers

The transpiling is a reasonable way to handle the code duplication issue, IF we don't have to edit the result. Any h... Greg Shah

06:51 AM Feature #4029: implement more widget rendering knowledge within the client drivers

I gave all the known options more thought and I still think that some kind of transcompiling the existing widget impl... Hynek Cihlar

10/13/2020

06:54 PM Support #4955: web client/javascript drawing - fast vs crisp

Greg Shah wrote:
> > Every line need separate putImageData.
>
> Why is this needed?
>
This is a kind of opti... Eugenie Lyzenko

06:53 PM Support #4955: web client/javascript drawing - fast vs crisp

> If I understand properly your idea this is a kind of "double buffering" concept. We have two bitmaps one that is cu... Greg Shah

06:52 PM Support #4955: web client/javascript drawing - fast vs crisp

Crazy idea number 2: use "WebGL":https://developer.mozilla.org/en-US/docs/Web/API/WebGL_API/Tutorial/Getting_started_... Greg Shah

06:43 PM Support #4955: web client/javascript drawing - fast vs crisp

Greg Shah wrote:
> > We use putImageData for each and every line draw, and this includes drawing rectangles. If we c... Eugenie Lyzenko

06:35 PM Support #4955: web client/javascript drawing - fast vs crisp

> Every line need separate putImageData.
Why is this needed?
> If we get full image covered by rectangle to dra... Greg Shah

06:22 PM Support #4955: web client/javascript drawing - fast vs crisp

Constantin Asofiei wrote:
> On Chrome, when opening a widget-heavy customer screen, the @this.offscreenCtx.putImageD... Eugenie Lyzenko

06:10 PM Support #4955: web client/javascript drawing - fast vs crisp

> We use putImageData for each and every line draw, and this includes drawing rectangles. If we could use a single pu... Greg Shah

01:52 PM Support #4955: web client/javascript drawing - fast vs crisp

On Chrome, when opening a widget-heavy customer screen, the @this.offscreenCtx.putImageData(imageData, x, y);@ line i... Constantin Asofiei

12:21 PM Support #4955: web client/javascript drawing - fast vs crisp

Sergey Ivanovskiy wrote:
> @putImageData@ doesn't take into account clipping regions. Thus, if @disablePixelManipula... Greg Shah

12:21 PM Support #4955: web client/javascript drawing - fast vs crisp

> what would you like to measure for our js web client. It can degrade performance if we add performance counters for... Greg Shah

12:00 PM Support #4955: web client/javascript drawing - fast vs crisp

@putImageData@ doesn't take into account clipping regions. Thus, if @disablePixelManipulation=true@, then some of the... Sergey Ivanovskiy

11:49 AM Support #4955: web client/javascript drawing - fast vs crisp

>It needs to fix incorrect lines when using drawings on the canvas. I tried to check clipping regions and empty lines... Sergey Ivanovskiy

11:25 AM Support #4955: web client/javascript drawing - fast vs crisp

Greg, what would you like to measure for our js web client. It can degrade performance if we add performance counters... Sergey Ivanovskiy

11:23 AM Support #4955: web client/javascript drawing - fast vs crisp

I think the point here is how fast the @BitBlt@ or line primitives work in graphics driver. Another word what is bett... Eugenie Lyzenko

08:29 AM Support #4955: web client/javascript drawing - fast vs crisp

> I agree that these changes don't improve the UI performance that it can be visible on your computer.
I think thi... Greg Shah

08:19 AM Support #4955: web client/javascript drawing - fast vs crisp

Eugenie Lyzenko wrote:
>Disabled pixel manipulation give the following picture with many GUI line artifacts:
Yes, I... Sergey Ivanovskiy

11:17 PM Support #4955: web client/javascript drawing - fast vs crisp

Another note. Can you estimate the performance gain from @disablePixelManipulation@ approach? On my machine I do not ... Eugenie Lyzenko

09:49 PM Support #4955: web client/javascript drawing - fast vs crisp

@<customer_screen_redacted>@ Eugenie Lyzenko

08:40 AM Feature #4029: implement more widget rendering knowledge within the client drivers

> Maybe you already think about this and ruled it out because of code duplication, but what if we push the drawing pr... Greg Shah

07:07 AM Feature #4029: implement more widget rendering knowledge within the client drivers

Maybe you already think about this and ruled it out because of code duplication, but what if we push the drawing prim... Ovidiu Maxiniuc

05:53 AM Feature #4029: implement more widget rendering knowledge within the client drivers

> The drawing primitives are already batched in single socket message. I don't think there will be much of a performa... Greg Shah

10/12/2020

06:44 PM Support #4955: web client/javascript drawing - fast vs crisp

Committed revision 11711 (3821c) changed @drawLineSegment@ so that if this flag, @disablePixelManipulation@, is set i... Sergey Ivanovskiy

05:27 PM Support #4955: web client/javascript drawing - fast vs crisp

Sergey Ivanovskiy wrote:
> Please review this patch. Can I commit these changes for 3821c?
Yes, no objection. Greg Shah

05:21 PM Support #4955: web client/javascript drawing - fast vs crisp

Please review this patch. Can I commit these changes for 3821c? Sergey Ivanovskiy

04:11 PM Support #4955: web client/javascript drawing - fast vs crisp

I encountered the new issue with @OffscreenCanvas@. If pixels manipulations are disabled, then line styles are not ap... Sergey Ivanovskiy

07:52 AM Support #4955: web client/javascript drawing - fast vs crisp

Let's use @disablePixelManipulation@ instead. The default value is @false@, which means that our current pixel manip... Greg Shah

03:29 AM Support #4955: web client/javascript drawing - fast vs crisp

It shouldn't be hard to implement because @p2j_canvas_rendering.js@ uses Canvas API for filling all drawings and only... Sergey Ivanovskiy

10:14 PM Support #4955: web client/javascript drawing - fast vs crisp

My perception is that the browser/javascript drawing performance is very dependent upon the GPU of the system.
F... Greg Shah

10:02 PM Support #4955 (Closed): web client/javascript drawing - fast vs crisp

Greg Shah

01:36 PM Feature #4029: implement more widget rendering knowledge within the client drivers

Greg Shah wrote:
> I think it may be time to consider a more serious implementation of this idea.
>
> One thing I... Hynek Cihlar

02:57 AM Feature #4029: implement more widget rendering knowledge within the client drivers

Hynek Cihlar wrote:
> Has anybody explored the idea of moving the canvas rendering in a separate web worker? This co... Sergey Ivanovskiy

02:44 AM Feature #4029: implement more widget rendering knowledge within the client drivers

Has anybody explored the idea of moving the canvas rendering in a separate web worker? This could free the main threa... Hynek Cihlar

10:25 PM Feature #4029: implement more widget rendering knowledge within the client drivers

I think it may be time to consider a more serious implementation of this idea.
One thing I don't want to lose is t... Greg Shah

10/09/2020

04:27 PM Bug #4954: Extent field alignment

Testcase:... Stanislav Lomany

04:24 PM Bug #4954 (New): Extent field alignment

Stanislav Lomany

03:54 PM Bug #4953: Frame is not displayed when backing buffer is not available

Testcase:... Stanislav Lomany

03:50 PM Bug #4953 (New): Frame is not displayed when backing buffer is not available

Stanislav Lomany

10/07/2020

11:51 AM Bug #4942: Embbeded web client doesn't work for Hotel GUI

Adrian Lungu wrote:
> The bug was fixed in 3821c. rev. 11675. @GenericWidget.pushWidgetAttr@ is overridden in sever... Constantin Asofiei

09:12 AM Bug #4942 (Closed): Embbeded web client doesn't work for Hotel GUI

Nice! Greg Shah

08:26 AM Bug #4942: Embbeded web client doesn't work for Hotel GUI

Please review only #4942-6 (3821c/11675), as the previous comments are just part of the bug hunting process. Adrian Lungu

08:24 AM Bug #4942 (Review): Embbeded web client doesn't work for Hotel GUI

Adrian Lungu

08:24 AM Bug #4942: Embbeded web client doesn't work for Hotel GUI

The bug was fixed in 3821c. rev. 11675. @GenericWidget.pushWidgetAttr@ is overridden in several other widgets (child... Adrian Lungu

08:20 AM Bug #4942: Embbeded web client doesn't work for Hotel GUI

Hynek/Constantin: Please review. Greg Shah

06:32 AM Bug #4942: Embbeded web client doesn't work for Hotel GUI

The root of the bug is related to FWD's specific @disable-redraw@. Commenting out the @disable-row@ statements from @... Adrian Lungu

10/06/2020

09:17 AM Bug #4942: Embbeded web client doesn't work for Hotel GUI

I could restrict the searching scope somewhere around 11600..11608 revisions. The most of the intermediary revisions ... Adrian Lungu

07:28 AM Bug #4942: Embbeded web client doesn't work for Hotel GUI

I added the fix from rev. 11619 straight to the 11588 revision and tested. It seems like the embedded web client star... Adrian Lungu

06:09 AM Bug #4942: Embbeded web client doesn't work for Hotel GUI

I checked some older revisions of 3821c in order to find which one triggered the bug. I should note that the latest 3... Adrian Lungu

10/05/2020

06:14 AM Bug #4942 (Closed): Embbeded web client doesn't work for Hotel GUI

Adrian Lungu

09/24/2020

11:52 AM Feature #4912: move UI portions of the web client to the server-side

Greg, wouldn't it make more sense to think of this problem the other way around? Keep the client processes as they ar... Hynek Cihlar

11:06 AM Feature #4912: move UI portions of the web client to the server-side

When the original UI architecture of FWD was designed, there were 2 "near term" targets: ChUI and fat client GUI (Swi... Greg Shah

09:58 AM Feature #4912 (New): move UI portions of the web client to the server-side

Greg Shah

09/18/2020

04:26 PM Bug #4907: Conversion bug with button triggers

The problem is most likely in @rules/annotations/triggers_phrase_rewrite.rules@. This code:... Greg Shah

03:58 PM Bug #4907: Conversion bug with button triggers

Greg Shah wrote:
> I guess this is related to the @TRIGGERS@ phrase which is not as commonly used as the @ON@ statem... Vladimir Tsichevski

03:43 PM Bug #4907: Conversion bug with button triggers

I guess this is related to the @TRIGGERS@ phrase which is not as commonly used as the @ON@ statement.
As a workaro... Greg Shah

03:29 PM Bug #4907: Conversion bug with button triggers

The trigger, associated with a button is *not* converted.
The 4gl:
@main.p@... Vladimir Tsichevski

03:24 PM Bug #4907 (New): Conversion bug with button triggers

Vladimir Tsichevski