Re: [DISCUSS] Discuss Release Apache FlexJS 0.8.0 RC2

Hi Om, I think you need this one [1], but you have to also prepare for you settings file [2] with active profile "staged-artifacts". I was testing those artifacts in following way: 1) Removed all folder 0.8.0 from ".m2\repository\org\apache\flex\flexjs\" 2) Try to compile simple app or complex [

Re: [DISCUSS] Discuss Release Apache FlexJS 0.8.0 RC2

I am trying to test the release artifacts, but there is no mention of maven build in the README. Are there instructions somewhere? Thanks, Om On Jun 23, 2017 12:29 AM, "Alex Harui" wrote: > I wrote this up recently [1]. It would be great if you can give it a try. > Let me know if you have qu

Re: [DISCUSS] Discuss Release Apache FlexJS 0.8.0 RC2

I wrote this up recently [1]. It would be great if you can give it a try. Let me know if you have questions. [1] https://cwiki.apache.org/confluence/display/FLEX/ApacheFlexBuild+CI+Server+ Maintenance -Alex On 6/22/17, 11:39 AM, "piotrz" wrote: >Alex, > >How can I restart this server ? This

Re: [DISCUSS] Discuss Release Apache FlexJS 0.8.0 RC2

Alex, How can I restart this server ? This app slave.jar cannot connect with the server. Thanks, Piotr - Apache Flex PMC piotrzarzyck...@gmail.com -- View this message in context: http://apache-flex-development.247.n4.nabble.com/DISCUSS-Discuss-Release-Apache-FlexJS-0-8-0-RC2-tp62498

Re: [FlexJS] TileLayout

In these situations (assuming I understand the problem), the first consideration is "what does the browser do?". So if I take a bunch of divs, give them explicit sizes and give them margins do you get what you expect/want? There are some funny rules on how margins work in CSS in the browser, so s

RE: [FlexJS] TileLayout

It’s a bit different because horizontal/vertical layouts don’t size the children. Tile does according to the set number of rows and columns. Peter, FlexBox works fine, the question is how to calculate the childrens’ size. It looks to me like another PAYG question. Is the basic tile functionalit

Re: [FlexJS] TileLayout

Hi Yishay, I think Josh wrote some bead - not sure whether it is related to your problem [1][2] [1] https://github.com/apache/flex-asjs/blob/release0.8.0/frameworks/projects/Basic/src/main/flex/org/apache/flex/html/beads/layouts/HorizontalLayoutWithPaddingAndGap.as [2] https://github.com/apache/f

Re: [FlexJS] TileLayout

You could write a new TileLayoutWithGapAndPadding to handle that specific situation. On the JS side, TileLayout uses Flexbox and perhaps you need more control. ‹peter On 6/22/17, 10:14 AM, "yishayw" wrote: >Adding margins adds a gutter on the right side of the host. This is >probably >because e

Re: [FlexJS] getElementIndex()

I was wondering why React keeps its own DOM and why it is so fast. Perhaps this is the reason. On 6/22/17, 10:56 AM, "Alex Harui" wrote: >Feel free to test your theory. It sounds like duplication of work to me >as in: The DOM does have some sort of list so why keep our own? Making >and manipul

Re: [FlexJS] getElementIndex()

Feel free to test your theory. It sounds like duplication of work to me as in: The DOM does have some sort of list so why keep our own? Making and manipulation Array is supposedly also not very performant. Regarding text nodes, I haven't looked to see when text nodes are intentionally children o

Re: [FlexJS] TileLayout

Adding margins adds a gutter on the right side of the host. This is probably because each child's width is calculated according to the parent's width divided by the number of children, disregarding the margins. I'm using padding instead, but that has the drawback that in a tiled list, for example,

Re: [FlexJS] TileLayout

Hi, Use margins on the items. ‹peter On 6/22/17, 9:48 AM, "yishayw" wrote: >What is the preferred way to set gaps between columns and between rows? > > > >-- >View this message in context: >https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fapache-fle >x-development.247.n4.nab

[FlexJS] TileLayout

What is the preferred way to set gaps between columns and between rows? -- View this message in context: http://apache-flex-development.247.n4.nabble.com/FlexJS-TileLayout-tp62568.html Sent from the Apache Flex Development mailing list archive at Nabble.com.

[FlexJS] getElementIndex()

Is there a reason we are using the underlying node lists for our collection of elements? I would think that it would be more performant (at the possible cost of some extra memory) to keep a list of FlexJS elements. childNodes are live collections which need to constantly be resolved so they sh

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

The changes are quite extensive, but I’ll break it up into a lot of smaller commits. > On Jun 22, 2017, at 12:04 PM, piotrz wrote: > > Harbs, > > Make sure that those changes are simply, cause I will do cherry pick up. > > Piotr > > > > - > Apache Flex PMC > piotrzarzyck...@gmail.com >

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

Harbs, Make sure that those changes are simply, cause I will do cherry pick up. Piotr - Apache Flex PMC piotrzarzyck...@gmail.com -- View this message in context: http://apache-flex-development.247.n4.nabble.com/Re-git-commit-flex-asjs-refs-heads-tlf-Reverted-strict-equality-tp62526p6

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

+1. If we have a sample app, we can analyze where performance problems might be and make an educated decision based on that. I am making more reverts (and optimizations) in the TLF branch. If there proves to be a compelling reason to use strict equality I will be happy to audit my code. > On

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

I think if changes in UIBase making app unusable and cause crash +1 for reverting it. - Performance is on the second place. I'm willing to put Harb's changes in develop, but Justin, Could you create simple app and raise jira which shows this performance issue? Since we did talk about raising j

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

And the only thing you changed was strict equality? I find that VERY hard to believe. I see you made lots of other changes. For example, I see you removed a try/catch in SimpleCSSValuesImpl. That seems like a much more likely culprit for performance improvement. > On Jun 22, 2017, at 10:43 AM,

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

Your link is actually arguing for use for non-strict equality… But again: Some places you need == and some places you need ===. Bugs can be introduced both ways. This is getting tiring, and I have work to do. > On Jun 22, 2017, at 10:38 AM, Justin Mclean wrote: > > HI, > > Point 1 may explai

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

Hi, > Again; this means nothing to me. What is your app doing and how much CPU time > was spent? Startup up and showing a few windows, changing a few tabs within a reasonably complex layout some web service call going on in the background. Initial cpu time was of the order of 480ms from memory

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

> On Jun 22, 2017, at 10:07 AM, Justin Mclean wrote: > > or do your own tests No need: https://jsperf.com/triple-equals-vs-twice-equals https://jsperf.com/triple-equals-vs-double-equals/3

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

HI, Point 1 may explain the issue clearer [1] there are situation we’re use of == and != is going to cause bugs. Thanks, Justin 1. https://herringtondarkholme.github.io/2016/11/05/how-to-write-copy-paste-friendly-code/

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

Again; this means nothing to me. What is your app doing and how much CPU time was spent? > On Jun 22, 2017, at 10:33 AM, Justin Mclean wrote: > > Hi, > >>> I got performance gains of 30-40%. >> >> Of what? Percentages when dealing with such cheap operators are generally >> meaningless. > >

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

Hi, >> I got performance gains of 30-40%. > > Of what? Percentages when dealing with such cheap operators are generally > meaningless. Of total CPU time spent on application for a few minutes. Measured using the Chrome profiler. Thanks, Justin

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

Justin, I'm wondering what area of your code get speed up. Maybe there is something which you can do it differently and gain that? Does your changes was visible in your benchmarks only? Does user complains about hanging some things or slowing? - Is it even visible to user? Piotr - Apa

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

> On Jun 22, 2017, at 10:14 AM, Justin Mclean wrote: > > Hi, > >> The main argument for using strict equality is because it does type >> conversions which can cause bugs. > > Incorrect strict equality doesn’t do type conversion, but I assume you meant > to say "it doesn’t do type conversion”

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

Hi, > The main argument for using strict equality is because it does type > conversions which can cause bugs. Incorrect strict equality doesn’t do type conversion, but I assume you meant to say "it doesn’t do type conversion”. > Since we are using a strongly typed language, type safety is gene

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

> On Jun 22, 2017, at 10:07 AM, Justin Mclean wrote: > I got performance gains of 30-40%. Of what? Percentages when dealing with such cheap operators are generally meaningless.

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

I just looked at the code and I see at least one possible bug. Assigning the source to undefined will fail. > On Jun 22, 2017, at 9:44 AM, piotrz wrote: > > Hi Harbs, > > That's a strong point to your changes. Compiler protect us. I'm wondering > there were also changes in ArrayList - Do you e

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

Hi, > You willy-nilly changed equality to strict equality. That cost me time and > money. I vetoed your change. Period. I have no issue with you fixing a bug or even asking me to fix it but please put my changes back. > Changing all equality to strict equality is (IMO) pedantic. In most cases

Re: git commit: [flex-asjs] [refs/heads/tlf] - Reverted strict equality

Hi Harbs, That's a strong point to your changes. Compiler protect us. I'm wondering there were also changes in ArrayList - Do you experience with that area some problems ? Thanks, Piotr - Apache Flex PMC piotrzarzyck...@gmail.com -- View this message in context: http://apache-flex-develop