Re: [royale-asjs] branch develop updated: Fixes #24

No. We’re not copying the MouseEvent. There are getters, but they are specific to MouseEvent and KeyboardEvent, rather than BrowserEvent. Whether it’s KeyboardEvent, MouseEvent or BrowserEvent, there is always going to be a constructor called which wraps the good event inside it. Would it

Re: Event Metadata

That is one of the possible approaches. The trade-off is, as you mention, having to filter out the right event vs all of the work re-dispatching events. Right now the strand is the wrapping component. It doesn't have to be and maybe we should change that, but in order for MXML event attributes

Re: [royale-asjs] branch develop updated: Fixes #24

We could, but it would be slightly more expensive. IMO it is a trade-off of usability vs code size. Flex leaned towards usability, but we saw several customers never get off the launch pad because of code size performance. My 2 cents, -Alex On 11/1/17, 12:20 PM, "Idylog - Nicolas Granon"

Re: Trying to commit to royale-asjs repo

Hi Carlos, It looks like it mentions SSO on step 10, but I didn't need to execute step 10. Step 9 was all I needed, then as it says below that, I use the token instead of a password when committing. HTH, -Alex On 11/1/17, 11:26 AM, "carlos.rov...@gmail.com on behalf of Carlos Rovira"

Re: [royale-asjs] branch develop updated: Fixes #24

Aren't we essentially making a copy of each MouseEvent? That doesn't sound right to me. Unless some code is actually checking the type: If (event is MouseEvent) it would seem to me to be far cheaper to just add new getters to BrowserEvent. And if we do want to support type-checking of

RE: [royale-asjs] branch develop updated: Fixes #24

I do not really understand why we do not have a simple "DateChooserEvent" with a "change" type, and "previousSelectedDate" and "selectedDate" properties. Nicolas Granon > -Message d'origine- > De : Piotr Zarzycki [mailto:piotrzarzyck...@gmail.com] > Envoyé : mercredi 1 novembre 2017

RE: Event Metadata

Excuse my naive comment : In my opinion, we have two different situations : Case 1 : I use a component, and I add one or more beads to it (in the source code where the component is used) (inlined). Case 2 : I use a component that is *already* composed of one strand and a number of beads (or

Re: Trying to commit to royale-asjs repo

Does this help? https://help.github.com/articles/error-permission-denied-publickey/ > On Nov 1, 2017, at 8:26 PM, Carlos Rovira wrote: > > Hi Alex, > don't know about this. I create the token. In the

Re: Trying to commit to royale-asjs repo

Hi Alex, don't know about this. I create the token. In the instructions it states that I should have an SSO button before EDIT and DELETE buttons, but I don't have that SSO button Where I could use that token generated? thanks 2017-11-01 15:49 GMT+01:00 Alex Harui : >

Re: package paths

What do we have left in the renaming effort? I can help with that. Harbs > On Nov 1, 2017, at 6:23 PM, Alex Harui wrote: > > Please wait until we finish the replacing of FLEX and FALCON before doing > package renaming.

Re: [royale-asjs] branch develop updated: Fixes #24

The only way to get MouseEvent and KeyboardEvent-specific properties was by digging into the native event buried in the goog event. Additionally, client code is expecting KeyboardEvents and MouseEvents and the expected properties was missing in the BrowserEvents. Before my changes, MouseEvents

Re: [royale-asjs] branch develop updated: Fixes #24

Those look like change logs, but I'm not recalling why you needed to do that. That looks pretty expensive to me. -Alex On 11/1/17, 10:33 AM, "Harbs" wrote: >We discussed this a few months ago: >https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.co

Re: Working on UI Controls styling

Hi, I think I could start to try what Harbs expose, although I think what I will need in the end is to control some SVG parts with variables. Maybe with the showed SVG/CSS relation could be sufficient. I'll be showing how limitations I find. As well as Alex said having inline SVG as HTML would be

RE: Restarting contributions

Hi Erik, Welcome back ! Rather than "ask for" components, we would like to learn how to build components. I don't think that we are good enough (or experienced enough) to contribute to the SDK at this stage. However, we would like to build some components ourselves for our own usage. We will

Re: Back on the list

Hi Nicolas, I'm glad that you back! Looking forward to have questions from you! Piotr On Wed, Nov 1, 2017, 18:22 Idylog - Nicolas Granon wrote: > Hi all, > > Since we are "only" developing applications, I did not interfere too much > in > the threads about SDK renaming

Re: [royale-asjs] branch develop updated: Fixes #24

We discussed this a few months ago: https://github.com/apache/royale-asjs/blob/develop/frameworks/projects/Basic/src/main/royale/org/apache/royale/core/HTMLElementWrapper.as#L68

Re: Working on UI Controls styling

I’m not sure. I haven’t seen problems. The only issues that come to mind are: 1. There’s no load events on SVG images on Microsoft browsers. 2. Chrome has issues with SVG, transforms and fractional pixels. 3. There’s some blending issues that different browsers handle differently depending on

Back on the list

Hi all, Since we are "only" developing applications, I did not interfere too much in the threads about SDK renaming process etc. Also, we do not want to compile the SDK ourselves : we believe that we must stick to the developer's point of view and we only use the binaries distribution. So, we

Re: [royale-asjs] branch develop updated: Fixes #24

How did you "fix" that? The most obvious way that popped into my mind would be pretty expensive. Thanks, -Alex On 11/1/17, 8:45 AM, "Harbs" wrote: >FWIW, I fixed MouseEvents and KeyboardEvents to be the Royale types and >not BrowserEvents as they are in the goog

Re: package paths

Please wait until we finish the replacing of FLEX and FALCON before doing package renaming. The original point of org.apache.flex.html and org.apache.flex.html5 was to thinly wrap existing HTMLElements and give them Flex-like names (Label, TextInput). Other components in Basic are compositions

Re: [royale-asjs] 02/02: Revert "Updating the source of an ArrayList does not change the view of the component without this change."

Hi Yishay, I like your changes :) I just added couple of minor suggestions in the github. Thanks! 2017-11-01 16:50 GMT+01:00 yishayw : > Hi Peter, > > Thanks for the detailed explanation. As I understand it, the vision you > outlined is already implemented, we just need

Re: [royale-asjs] 02/02: Revert "Updating the source of an ArrayList does not change the view of the component without this change."

Hi Peter, Thanks for the detailed explanation. As I understand it, the vision you outlined is already implemented, we just need to listen to more of these fine-grained events in the mappers which can happen DAYG. I personally don't see anything wrong with this architecture, maybe Harbs who's

Re: [royale-asjs] branch develop updated: Fixes #24

FWIW, I fixed MouseEvents and KeyboardEvents to be the Royale types and not BrowserEvents as they are in the goog libraries. OK. Let’s get rid of selectedDateChanged. > On Nov 1, 2017, at 5:41 PM, Alex Harui wrote: > > AIUI, events generated by an HTMLElement are

Re: package paths

Harbs, If you will start working on that do not forget about maven changes. If you don't want to touch it during your work just let me know and I will make the changes there. It would be great also raise Issue on github and make changes under feature/branch - it will be helpful to make changes on

Re: package paths

I find I very often have to open the entire asjs repo so I can search for a specific class. > On Nov 1, 2017, at 4:55 PM, Peter Ent wrote: > > I agree, the namespaces can remain. The packaging is more for > logic/organization to make it easier to get into the code.

Re: package paths

I think the namespaces would probably stay the same. We currently have the following namespaces and I don’t see a need to change them: library://ns.apache.org/royale/basic library://ns.apache.org/royale/svg library://ns.apache.org/royale/express library://ns.apache.org/royale/flat

Re: package paths

This makes a lot of sense to me. I think that if we’re going to do this, the time to do so is now. I’m willing to help with this reorganization. Harbs > On Nov 1, 2017, at 3:33 PM, Peter Ent wrote: > > I'm glad you brought this up. I've been giving some thought to

Re: package paths

I'm glad you brought this up. I've been giving some thought to refactoring Royale into more logical components. We've done this before, but I think some refinement is in order before we could have a 1.0 release that would make sense to the general public. I think streaming and moving things around

Re: Working on UI Controls styling

You can also use a GraphicShape class and set the element to any random SVG. This would work as long as you don’t need access to the SVG structure using normal Royale methods. Harbs > On Nov 1, 2017, at 3:12 PM, Harbs wrote: > > We also have an SVG package. Look in

Re: Working on UI Controls styling

> On Nov 1, 2017, at 2:45 PM, Carlos Rovira > wrote: > > Hi Harbs, > > could you post some basic snippet for this to look at your approach? > > Thanks! > > 2017-11-01 13:42 GMT+01:00 Harbs : > >> You can use SVG files instead of PNG

Re: Working on UI Controls styling

Hi Harbs, could you post some basic snippet for this to look at your approach? Thanks! 2017-11-01 13:42 GMT+01:00 Harbs : > You can use SVG files instead of PNG files. I’m doing that extensively in > my app, but that will only work for components which have src or >

Re: [royale-asjs] branch develop updated: Fixes #24

Harbs, Whole framework is working like that. We are getting in most cases BroserEvent. I'm always in favor of "change" event, but the second one could be also ok. Piotr 2017-11-01 8:37 GMT+01:00 Harbs : > I’m not sure why/how, but I checked and the change event on the