Re: [FDK 4.12.0] - wrong info in flex-config.xml about RSL
Hi All, Yes, I'm Aware of that Apache RSL's are not signed. But, one of our product have 1 main application, 8 sub modules, 9 * 1 style module and 9 * 5 resource modules. user may visit any module at any time in any locale. and our client want to support more locale and wanted to have more style/theme (style modules) options as user preference. and eventually that product will grow further with new modules. So we prefer RSL instead of binding the same core classes into all the modules and make them fat. Also, I'm very excited to try the latest Apache Flex Installer 3. *Regards,S. Jagan Langa* -- 2014-03-21 19:56 GMT+05:30 Tom Chiverton t...@extravision.com: On 21/03/14 14:22, Alex Harui wrote: I assume you are aware that Apache Flex RSLs are not signed and therefore are really only helpful if you are deploying more than one Flex app from a domain. Or if you change your core app more frequently than SDK Tom
Re: [FDK 4.12.0] - wrong info in flex-config.xml about RSL
Hi, When I tried downloading the SDK by [RC] Installer 3 ((have configured the download cache option)), due to MD5 mismatch installation was aborted. The problem/feedback's I noticed here are, - there is no option to retry, I have to close the app. and restart it. - even after I restart the app. since i enabled the cache, it is reusing it from the cache and aborting the installation for the same reason. (I guess app. should delete (only) the corrupted file from the cache, instead of expecting the user to delete the cache by himself.) *Regards,S. Jagan Langa* *|* *Project Manager* *Sybrant Technologies Pvt. Ltd.* ** Mobile: +91 98430 50897 Skype: seesamjagan -- In The Media: http://www.pr-inside.com/motorola-solutions-showcases-sybrant-s-mobile-r3745747.htm *About Sybrant:* *Sybrant is a premier web/mobile/UIUX expert. Most admirable IT firms in India too, leverage Sybrant's expertise for their Web UIUX needs. We are recognized for working with our clients right from the product conceptualization till the execution stage offering end-to-end solutions that sets us apart from other firms in this space. Right from developing a learning management system for a French company to helping a Non-Profit with a Mobile application till developing UX for a IT firm, Sybrant is known for delighting Customers* -- 2014-03-24 11:57 GMT+05:30 Jagan Langa Sami Durai jaga...@sybrant.com: Hi All, Yes, I'm Aware of that Apache RSL's are not signed. But, one of our product have 1 main application, 8 sub modules, 9 * 1 style module and 9 * 5 resource modules. user may visit any module at any time in any locale. and our client want to support more locale and wanted to have more style/theme (style modules) options as user preference. and eventually that product will grow further with new modules. So we prefer RSL instead of binding the same core classes into all the modules and make them fat. Also, I'm very excited to try the latest Apache Flex Installer 3. * Regards,S. Jagan Langa* -- 2014-03-21 19:56 GMT+05:30 Tom Chiverton t...@extravision.com: On 21/03/14 14:22, Alex Harui wrote: I assume you are aware that Apache Flex RSLs are not signed and therefore are really only helpful if you are deploying more than one Flex app from a domain. Or if you change your core app more frequently than SDK Tom
Re: [FlexJS] Tasks before first Release Candidate
On 21/03/14 19:19, Alex Harui wrote: folks to try. Yes, the changes above might have backward compatibility impact on folks using these first releases, but IMO, if we set expectation correctly, folks will be flexible. Release early, release often would be my suggestion. Package or XMLNS changes are just a global search and replace down the road, and the change to Falcon is just a tweak to the the build process (of users of the SDK). Not very much work. Tom
RE: Search in Apache Flex Docs?
I could look into it, but it wouldn't be my highest priority. Thanks Alex, I have raised a JIRA ticket, for future reference: https://issues.apache.org/jira/browse/FLEX-34183 Do you think I can start working on it when I have time, maybe on a simpler implementation, of better to wait for Adobe's answer? Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : lundi 24 mars 2014 04:20 À : dev@flex.apache.org; n...@joeflash.ca Objet : Re: Search in Apache Flex Docs? On 3/23/14 11:41 AM, Maurice Amsellem maurice.amsel...@systar.com wrote: The Adobe ASLR does have search. I wonder what it would take to implement it. It does not seem to be that complicated ( file asdoc.js, doing a regexp match on a tags in the left list, and setting display style accordingly, to show or hide an item). Sorry for insisting, but could we get that asdoc extensions donated by Adobe, or at least permission to use them in Apache Flex ASDOC, or do we have to rewrite them? I could look into it, but it wouldn't be my highest priority. -Alex
Re: RTL support in mobile apps
Hi, if there are plans to introduce TLF on mobile TextInput this will change my priorities about change the internals of MaskedTextInput component and will only make it to preserve slot positions. IMO, if now TLF give us a good performance in mobile this days it will be very useful to make it happen since this will be more aligned to the Flex philosophy. Right now we need to deal in different ways with TextInput in mobile and browser and this defeat the code once run everywhere. So +1 to TLF support on mobile is performance is good! :) Please let me know if that's are the plans. Thanks! Carlos 2014-03-23 21:08 GMT+01:00 Maurice Amsellem maurice.amsel...@systar.com: Found a number of tickets on this topic: https://issues.apache.org/jira/browse/FLEX-26365 (closed as later) https://issues.apache.org/jira/browse/FLEX-34145 (closed) https://issues.apache.org/jira/browse/FLEX-34181 (In progress) https://issues.apache.org/jira/browse/FLEX-33750 (open) https://issues.apache.org/jira/browse/FLEX-28107 (later) https://issues.apache.org/jira/browse/FLEX-28103 (later) https://issues.apache.org/jira/browse/FLEX-26169 (later) https://issues.apache.org/jira/browse/FLEX-24502 (later) Maurice -Message d'origine- De : Maurice Amsellem [mailto:maurice.amsel...@systar.com] Envoyé : samedi 22 mars 2014 01:14 À : dev@flex.apache.org Objet : RE: RTL support in mobile apps Yes, that might me the answer: so I need to cancel the flipping like I did for StageText (and like is done in spark Label). I will try this tomorrow. Still does not explain why TextField accepts bidi text now ? Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : samedi 22 mars 2014 01:09 À : dev@flex.apache.org Objet : Re: RTL support in mobile apps Again, I was not highly involved in this code, but IIRC, the TextLines are never flipped, so if you choose a flipped layoutDirection the TextLines are re-flipped. But if you start flipping TextFields without embedded text they go blank. Does that explain what you're seeing? -Alex On 3/21/14 5:00 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: Thanks Alex. That was also my understanding. Regarding TextInput / TextArea, there is no issue with regard to RTL in using StageText ( embedded in StyleableStageText or ScrollableStageText) . Now something strange that gets me puzzled. I did some experiments with mobile components that use TextField (actually StyleableTextField) and I managed to displayed Arabic/Hebrew (list , titles and nav bar) https://www.dropbox.com/s/4e4untcp3f4jeb2/List_arabic_LTR.png But this works only if the surrounding View or the application layoutDirection is set to ltr. And indeed, you notice that the text is RTL but the layout is still LTR. Now, if I set layoutDirection to RTL either at the Application or View , then everything disappears: https://www.dropbox.com/s/jzu1veecjm64m51/list_Arabic_RTL.png I thought that layoutDirection = RTL was merely applying a mirroring transform to the display. I am confused. Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : samedi 22 mars 2014 00:44 À : dev@flex.apache.org Objet : Re: RTL support in mobile apps I wasn't on the mobile components team (I did some mobile work but mostly worked on other SDK stuff), but fundamentally, if there's a TextField involved, then there is no RTL support. You need TextLines for RTL. You may be able to swap in the desktop skins for TextInput/TextArea and pay the performance and memory hit to get RTL text, but then I'm not sure how well StageText will work with that, if at all. Essentially, the mobile team traded off RTL support for better performance. Now, that was several years ago and phones and tablets are faster, so it might be worth revisiting that decision. -Alex On 3/21/14 3:41 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: Hi Team, Ori Segal has reported a problem in TextInput default skin with RTL (Hebrew, arabic) layout. I have fixed this problem. Now he has reported a problem in TextInput prompt text not being displayed in RTL. So I did a small test: set layoutDirection=rtl to a sample mobile app (with buttons, mobilegrid, etc..) and almost every text disappeared. The only texts that seem to be displayed correctly are: - TextInput / TextArea with the default text (that is using native StageText) - spark Label, that is using TextLine (and the new FTE/TLF engine). Everything else, that uses the mobile-optimized StyleableTextField, will not display RTL (apparently because it's based on the old TextField engine). Reading the articles below, it seems clear enough that RTL is NOT supported on AIR mobile (with a few exceptions): http://sourceforge.net/adobe/flexsdk/wiki/Mobile%20Text%20Components/ http://help.adobe.com/en_US/flex/using/WS02f7d8d4857b1677-165a04e11269 5
AW: [ANNOUNCE] Welcome Christofer as the newest PMC member
Hi guys (Back from my skiing Holiday ;-)) Thanks for the kind words :-) Really looking forward to reanimating some parts called dead (BlazeDS) and helping giving birth to some new parts (FlexJS). Think we'll really be doing some kick-ass stuff in the near future ;-) Chris Von: Kessler CTR Mark J mark.kessler@usmc.mil Gesendet: Freitag, 21. März 2014 11:56 An: dev@flex.apache.org Betreff: RE: [ANNOUNCE] Welcome Christofer as the newest PMC member Congratulations! -Mark -Original Message- From: Justin Mclean [mailto:jus...@classsoftware.com] Sent: Wednesday, March 19, 2014 8:30 PM To: dev@flex.apache.org Subject: [ANNOUNCE] Welcome Christofer as the newest PMC member Hi, For all his hard work and recent contributions to Apache Flex the Apache Flex PMC has voted Christofer Dutz in as a PMC member. Congratulations! Justin
AW: [ANNOUNCE] Welcome Carlos as the newest PMC member
Congrats from me too :-) Von: Kessler CTR Mark J mark.kessler@usmc.mil Gesendet: Freitag, 21. März 2014 11:56 An: dev@flex.apache.org Betreff: RE: [ANNOUNCE] Welcome Carlos as the newest PMC member Congratulations! -Mark -Original Message- From: Justin Mclean [mailto:jus...@classsoftware.com] Sent: Wednesday, March 19, 2014 8:30 PM To: dev@flex.apache.org Subject: [ANNOUNCE] Welcome Carlos as the newest PMC member Hi, For all his hard work and recent contributions to Apache Flex the Apache Flex PMC has voted Carlos Rovira in as a PMC member. Congratulations! Justin
AW: How to get the attachments of old Adobe JIRA / imported Apache JIRA?
When migrating the Flexmojos Issues from Sonatype to the new Atlassian Jira, I wrote a Manual Migration tool that also transfered attachments using the REST api If there is some stuff missing, we could code a similar tool to Transfer missing attachments. Chris Von: Justin Mclean jus...@classsoftware.com Gesendet: Freitag, 21. März 2014 13:39 An: dev@flex.apache.org Betreff: Re: How to get the attachments of old Adobe JIRA / imported Apache JIRA? Hi, Just post the issue number(s) and Alex should be able to find it for you. Justin On 21 Mar 2014, at 9:21 pm, Marcus Fritze marcus.fri...@googlemail.com wrote: Hi list, unfortunately the attachments to the issues are still not accessible in the JIRA. Is there a easy way to get attachments of some (3-5) issues? Thanks Marcus
RE: RTL support in mobile apps
Hi Carlos, 1) It's not proven yet that TLF is fast enough on mobile, especially when there are lots of text to display, such as in lists of datagrid. Plus I have discovered that the old TextField is actually capable to display RTL , but the Flex positioning is broken, so the text does not appear (probably because it was not supposed to work that way). So IMO, the question is still open, and I won't rush into replacing TextField by TLF on mobile. It would be probably much simpler to fix the layout. Right now we need to deal in different ways with TextInput in mobile and browser and this defeat the code once run everywhere. What do you mean? From the SDK developer standpoint, or from the end-user developer stand point ? From the SDK standpoint, the difference is only on the skin, the 'host' component is the same. From the end-user developer, you must use TextInput in both cases, so where's the difference ? The behavior is different, but that's inherent to mobile vs desktop (eg. you don't have softkeyboard or restricted keyboards on desktop). Please explain Maurice -Message d'origine- De : carlos.rov...@gmail.com [mailto:carlos.rov...@gmail.com] De la part de Carlos Rovira Envoyé : lundi 24 mars 2014 10:54 À : dev@flex.apache.org Objet : Re: RTL support in mobile apps Hi, if there are plans to introduce TLF on mobile TextInput this will change my priorities about change the internals of MaskedTextInput component and will only make it to preserve slot positions. IMO, if now TLF give us a good performance in mobile this days it will be very useful to make it happen since this will be more aligned to the Flex philosophy. Right now we need to deal in different ways with TextInput in mobile and browser and this defeat the code once run everywhere. So +1 to TLF support on mobile is performance is good! :) Please let me know if that's are the plans. Thanks! Carlos 2014-03-23 21:08 GMT+01:00 Maurice Amsellem maurice.amsel...@systar.com: Found a number of tickets on this topic: https://issues.apache.org/jira/browse/FLEX-26365 (closed as later) https://issues.apache.org/jira/browse/FLEX-34145 (closed) https://issues.apache.org/jira/browse/FLEX-34181 (In progress) https://issues.apache.org/jira/browse/FLEX-33750 (open) https://issues.apache.org/jira/browse/FLEX-28107 (later) https://issues.apache.org/jira/browse/FLEX-28103 (later) https://issues.apache.org/jira/browse/FLEX-26169 (later) https://issues.apache.org/jira/browse/FLEX-24502 (later) Maurice -Message d'origine- De : Maurice Amsellem [mailto:maurice.amsel...@systar.com] Envoyé : samedi 22 mars 2014 01:14 À : dev@flex.apache.org Objet : RE: RTL support in mobile apps Yes, that might me the answer: so I need to cancel the flipping like I did for StageText (and like is done in spark Label). I will try this tomorrow. Still does not explain why TextField accepts bidi text now ? Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : samedi 22 mars 2014 01:09 À : dev@flex.apache.org Objet : Re: RTL support in mobile apps Again, I was not highly involved in this code, but IIRC, the TextLines are never flipped, so if you choose a flipped layoutDirection the TextLines are re-flipped. But if you start flipping TextFields without embedded text they go blank. Does that explain what you're seeing? -Alex On 3/21/14 5:00 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: Thanks Alex. That was also my understanding. Regarding TextInput / TextArea, there is no issue with regard to RTL in using StageText ( embedded in StyleableStageText or ScrollableStageText) . Now something strange that gets me puzzled. I did some experiments with mobile components that use TextField (actually StyleableTextField) and I managed to displayed Arabic/Hebrew (list , titles and nav bar) https://www.dropbox.com/s/4e4untcp3f4jeb2/List_arabic_LTR.png But this works only if the surrounding View or the application layoutDirection is set to ltr. And indeed, you notice that the text is RTL but the layout is still LTR. Now, if I set layoutDirection to RTL either at the Application or View , then everything disappears: https://www.dropbox.com/s/jzu1veecjm64m51/list_Arabic_RTL.png I thought that layoutDirection = RTL was merely applying a mirroring transform to the display. I am confused. Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : samedi 22 mars 2014 00:44 À : dev@flex.apache.org Objet : Re: RTL support in mobile apps I wasn't on the mobile components team (I did some mobile work but mostly worked on other SDK stuff), but fundamentally, if there's a TextField involved, then there is no RTL support. You need TextLines for RTL. You may be able to swap in the desktop skins for TextInput/TextArea and pay the performance and memory hit to get RTL text, but then I'm
Re: How to get the attachments of old Adobe JIRA / imported Apache JIRA?
Oh, I missed your answer. (I don't know why, but many of your e-mails were automatically marked as junk in my mail program. Maybe, you are sending too much e-mails. ;-) ) Ok, it would be nice to have the attachments for the bugs: https://issues.apache.org/jira/browse/FLEX-22758 https://issues.apache.org/jira/browse/FLEX-20272 (if source is attached and not only the swf) https://issues.apache.org/jira/browse/FLEX-14784 https://issues.apache.org/jira/browse/FLEX-20755 Thanks. Marcus Am 21.03.2014 um 13:39 schrieb Justin Mclean jus...@classsoftware.com: Hi, Just post the issue number(s) and Alex should be able to find it for you. Justin On 21 Mar 2014, at 9:21 pm, Marcus Fritze marcus.fri...@googlemail.com wrote: Hi list, unfortunately the attachments to the issues are still not accessible in the JIRA. Is there a easy way to get attachments of some (3-5) issues? Thanks Marcus signature.asc Description: Message signed with OpenPGP using GPGMail
Re: AW: How to get the attachments of old Adobe JIRA / imported Apache JIRA?
Essentially, every attachment is missing. When Infra did the import of the Flex issues, they had the attachments, but assumed that they could add the attachments after import (which worked in prior versions of JIRA) and then found out that wasn't true in the current version. There is about 7GB of attachments and Infra does not want us to use REST or SOAP to re-insert the attachments because that's a lot of data and because there is no way to recover if we get it wrong. Flex JIRA issues are in the same database as all of the other Apache projects and shutting them down to recover the Flex JIRA is not desirable. The plan is that someday, I (or someone) will find time to write a utility that renames the attachments to their proper names (JIRA just stores them as 1, 2, 3, etc, no suffixes or other hints as to the content, the name is determined from an XML file), and then Apache will post those files somewhere. Given that in two years we've only been asked for less than 10 attachments, it is low on my list of priorities. -Alex On 3/24/14 3:16 AM, Christofer Dutz christofer.d...@c-ware.de wrote: When migrating the Flexmojos Issues from Sonatype to the new Atlassian Jira, I wrote a Manual Migration tool that also transfered attachments using the REST api If there is some stuff missing, we could code a similar tool to Transfer missing attachments. Chris Von: Justin Mclean jus...@classsoftware.com Gesendet: Freitag, 21. März 2014 13:39 An: dev@flex.apache.org Betreff: Re: How to get the attachments of old Adobe JIRA / imported Apache JIRA? Hi, Just post the issue number(s) and Alex should be able to find it for you. Justin On 21 Mar 2014, at 9:21 pm, Marcus Fritze marcus.fri...@googlemail.com wrote: Hi list, unfortunately the attachments to the issues are still not accessible in the JIRA. Is there a easy way to get attachments of some (3-5) issues? Thanks Marcus
RE: FlexUnit was Unit tests in TLF
BTW, was one of the enhancements automated test generation? Maybe from metadata on the class? There was a bigger project on that, yes, although we didn't get particularly far before the Apache switch so most of it was put on hold. Also, Flash Builder would create stubs for you based on the code model it had, so we don't go that route. Mike
RE: [DISCUSSION] Release Apache Flex FlexUnit 4.2.0 RC2
I'd like Labriola to verify that DP should be in the NOTICE and whether anyone else like Adobe should also be in there. I will make time to review but may have some questions as I feel a bit mal-informed on procedure on these fronts. Members of Adobe Consulting did write and donate the UIRunner that you see.
Re: How to get the attachments of old Adobe JIRA / imported Apache JIRA?
OK, added attachments. There doesn't seem to be source for FLEX-20272 From: Marcus Fritze marcus.fri...@googlemail.commailto:marcus.fri...@googlemail.com Reply-To: dev@flex.apache.orgmailto:dev@flex.apache.org dev@flex.apache.orgmailto:dev@flex.apache.org Date: Monday, March 24, 2014 4:56 AM To: dev@flex.apache.orgmailto:dev@flex.apache.org dev@flex.apache.orgmailto:dev@flex.apache.org Subject: Re: How to get the attachments of old Adobe JIRA / imported Apache JIRA? Oh, I missed your answer. (I don't know why, but many of your e-mails were automatically marked as junk in my mail program. Maybe, you are sending too much e-mails. ;-) ) Ok, it would be nice to have the attachments for the bugs: https://issues.apache.org/jira/browse/FLEX-22758 https://issues.apache.org/jira/browse/FLEX-20272 (if source is attached and not only the swf) https://issues.apache.org/jira/browse/FLEX-14784 https://issues.apache.org/jira/browse/FLEX-20755 Thanks. Marcus Am 21.03.2014 um 13:39 schrieb Justin Mclean jus...@classsoftware.commailto:jus...@classsoftware.com: Hi, Just post the issue number(s) and Alex should be able to find it for you. Justin On 21 Mar 2014, at 9:21 pm, Marcus Fritze marcus.fri...@googlemail.commailto:marcus.fri...@googlemail.com wrote: Hi list, unfortunately the attachments to the issues are still not accessible in the JIRA. Is there a easy way to get attachments of some (3-5) issues? Thanks Marcus
Re: How to get the attachments of old Adobe JIRA / imported Apache JIRA?
Thanks for the explanation. Is it possible to download the 7GB and the XML to search manually? And maybe - if it's not to complicated - to write that utility. Thanks Marcus Am 24.03.2014 um 17:18 schrieb Alex Harui aha...@adobe.com: Essentially, every attachment is missing. When Infra did the import of the Flex issues, they had the attachments, but assumed that they could add the attachments after import (which worked in prior versions of JIRA) and then found out that wasn't true in the current version. There is about 7GB of attachments and Infra does not want us to use REST or SOAP to re-insert the attachments because that's a lot of data and because there is no way to recover if we get it wrong. Flex JIRA issues are in the same database as all of the other Apache projects and shutting them down to recover the Flex JIRA is not desirable. The plan is that someday, I (or someone) will find time to write a utility that renames the attachments to their proper names (JIRA just stores them as 1, 2, 3, etc, no suffixes or other hints as to the content, the name is determined from an XML file), and then Apache will post those files somewhere. Given that in two years we've only been asked for less than 10 attachments, it is low on my list of priorities. -Alex On 3/24/14 3:16 AM, Christofer Dutz christofer.d...@c-ware.de wrote: When migrating the Flexmojos Issues from Sonatype to the new Atlassian Jira, I wrote a Manual Migration tool that also transfered attachments using the REST api If there is some stuff missing, we could code a similar tool to Transfer missing attachments. Chris Von: Justin Mclean jus...@classsoftware.com Gesendet: Freitag, 21. März 2014 13:39 An: dev@flex.apache.org Betreff: Re: How to get the attachments of old Adobe JIRA / imported Apache JIRA? Hi, Just post the issue number(s) and Alex should be able to find it for you. Justin On 21 Mar 2014, at 9:21 pm, Marcus Fritze marcus.fri...@googlemail.com wrote: Hi list, unfortunately the attachments to the issues are still not accessible in the JIRA. Is there a easy way to get attachments of some (3-5) issues? Thanks Marcus signature.asc Description: Message signed with OpenPGP using GPGMail
Re: Search in Apache Flex Docs?
It would save me much time if you find a different solution. Looks like some folks have suggested that we simply have the files indexed by a web search engine. -Alex On 3/24/14 2:38 AM, Maurice Amsellem maurice.amsel...@systar.com wrote: I could look into it, but it wouldn't be my highest priority. Thanks Alex, I have raised a JIRA ticket, for future reference: https://issues.apache.org/jira/browse/FLEX-34183 Do you think I can start working on it when I have time, maybe on a simpler implementation, of better to wait for Adobe's answer? Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : lundi 24 mars 2014 04:20 À : dev@flex.apache.org; n...@joeflash.ca Objet : Re: Search in Apache Flex Docs? On 3/23/14 11:41 AM, Maurice Amsellem maurice.amsel...@systar.com wrote: The Adobe ASLR does have search. I wonder what it would take to implement it. It does not seem to be that complicated ( file asdoc.js, doing a regexp match on a tags in the left list, and setting display style accordingly, to show or hide an item). Sorry for insisting, but could we get that asdoc extensions donated by Adobe, or at least permission to use them in Apache Flex ASDOC, or do we have to rewrite them? I could look into it, but it wouldn't be my highest priority. -Alex
Re: [FDK 4.12.0] - wrong info in flex-config.xml about RSL
Please file a bug. Thanks, -Alex On 3/24/14 1:40 AM, Jagan Langa Sami Durai jaga...@sybrant.com wrote: Hi, When I tried downloading the SDK by [RC] Installer 3 ((have configured the download cache option)), due to MD5 mismatch installation was aborted. The problem/feedback's I noticed here are, - there is no option to retry, I have to close the app. and restart it. - even after I restart the app. since i enabled the cache, it is reusing it from the cache and aborting the installation for the same reason. (I guess app. should delete (only) the corrupted file from the cache, instead of expecting the user to delete the cache by himself.) *Regards,S. Jagan Langa* *|* *Project Manager* *Sybrant Technologies Pvt. Ltd.* ** Mobile: +91 98430 50897 Skype: seesamjagan -- In The Media: http://www.pr-inside.com/motorola-solutions-showcases-sybrant-s-mobile-r37 45747.htm *About Sybrant:* *Sybrant is a premier web/mobile/UIUX expert. Most admirable IT firms in India too, leverage Sybrant's expertise for their Web UIUX needs. We are recognized for working with our clients right from the product conceptualization till the execution stage offering end-to-end solutions that sets us apart from other firms in this space. Right from developing a learning management system for a French company to helping a Non-Profit with a Mobile application till developing UX for a IT firm, Sybrant is known for delighting Customers* -- 2014-03-24 11:57 GMT+05:30 Jagan Langa Sami Durai jaga...@sybrant.com: Hi All, Yes, I'm Aware of that Apache RSL's are not signed. But, one of our product have 1 main application, 8 sub modules, 9 * 1 style module and 9 * 5 resource modules. user may visit any module at any time in any locale. and our client want to support more locale and wanted to have more style/theme (style modules) options as user preference. and eventually that product will grow further with new modules. So we prefer RSL instead of binding the same core classes into all the modules and make them fat. Also, I'm very excited to try the latest Apache Flex Installer 3. * Regards,S. Jagan Langa* - - 2014-03-21 19:56 GMT+05:30 Tom Chiverton t...@extravision.com: On 21/03/14 14:22, Alex Harui wrote: I assume you are aware that Apache Flex RSLs are not signed and therefore are really only helpful if you are deploying more than one Flex app from a domain. Or if you change your core app more frequently than SDK Tom
Re: [FDK 4.12.0] - wrong info in flex-config.xml about RSL
On 3/23/14 11:27 PM, Jagan Langa Sami Durai jaga...@sybrant.com wrote: Hi All, Yes, I'm Aware of that Apache RSL's are not signed. But, one of our product have 1 main application, 8 sub modules, 9 * 1 style module and 9 * 5 resource modules. user may visit any module at any time in any locale. and our client want to support more locale and wanted to have more style/theme (style modules) options as user preference. and eventually that product will grow further with new modules. So we prefer RSL instead of binding the same core classes into all the modules and make them fat. OK. Keep in mind that using the RSL is not optimal. It will take more time to configure, but you could create a shared code module and have a more optimal setup. -Alex
Re: How to get the attachments of old Adobe JIRA / imported Apache JIRA?
Am 24.03.2014 um 17:43 schrieb Alex Harui aha...@adobe.com: OK, added attachments. There doesn't seem to be source for FLEX-20272 Thank you very much. signature.asc Description: Message signed with OpenPGP using GPGMail
AW: AW: How to get the attachments of old Adobe JIRA / imported Apache JIRA?
Well how about creating a too that uses the REST api to fetch the attachments from the Adobe JIRA and to save them using their real names in directories matching the issue ID? Something like this: target-dir/ target-dir/FLEX-123 target-dir/FLEX-123/sample.mxml target-dir/FLEX-123/as-sample.as target-dir/FLEX-123/sometext.txt target-dir/FLEX-425 target-dir/FLEX-425/data.dmp ... Chris Von: Alex Harui aha...@adobe.com Gesendet: Montag, 24. März 2014 17:18 An: dev@flex.apache.org Betreff: Re: AW: How to get the attachments of old Adobe JIRA / imported Apache JIRA? Essentially, every attachment is missing. When Infra did the import of the Flex issues, they had the attachments, but assumed that they could add the attachments after import (which worked in prior versions of JIRA) and then found out that wasn't true in the current version. There is about 7GB of attachments and Infra does not want us to use REST or SOAP to re-insert the attachments because that's a lot of data and because there is no way to recover if we get it wrong. Flex JIRA issues are in the same database as all of the other Apache projects and shutting them down to recover the Flex JIRA is not desirable. The plan is that someday, I (or someone) will find time to write a utility that renames the attachments to their proper names (JIRA just stores them as 1, 2, 3, etc, no suffixes or other hints as to the content, the name is determined from an XML file), and then Apache will post those files somewhere. Given that in two years we've only been asked for less than 10 attachments, it is low on my list of priorities. -Alex On 3/24/14 3:16 AM, Christofer Dutz christofer.d...@c-ware.de wrote: When migrating the Flexmojos Issues from Sonatype to the new Atlassian Jira, I wrote a Manual Migration tool that also transfered attachments using the REST api If there is some stuff missing, we could code a similar tool to Transfer missing attachments. Chris Von: Justin Mclean jus...@classsoftware.com Gesendet: Freitag, 21. März 2014 13:39 An: dev@flex.apache.org Betreff: Re: How to get the attachments of old Adobe JIRA / imported Apache JIRA? Hi, Just post the issue number(s) and Alex should be able to find it for you. Justin On 21 Mar 2014, at 9:21 pm, Marcus Fritze marcus.fri...@googlemail.com wrote: Hi list, unfortunately the attachments to the issues are still not accessible in the JIRA. Is there a easy way to get attachments of some (3-5) issues? Thanks Marcus
Re: [DISCUSSION] Release Apache Flex FlexUnit 4.2.0 RC2
On 3/24/14 9:28 AM, Michael A. Labriola labri...@digitalprimates.net wrote: I'd like Labriola to verify that DP should be in the NOTICE and whether anyone else like Adobe should also be in there. I will make time to review but may have some questions as I feel a bit mal-informed on procedure on these fronts. Members of Adobe Consulting did write and donate the UIRunner that you see. Interesting. Did you have an agreement with Adobe as to code ownership or copyright of contributions by the AC folks? In an Apache-like donation model, the AC contributions would still be copyright Adobe but licensed to DP or the FlexUnit owner, and then NOTICE file should have both DP and Adobe in it. And any other major contributors. In a work-for-hire model, it isn't really a donation, and the hiring party generally owns the code unless there is an agreement that says otherwise. -Alex
RE: Search in Apache Flex Docs?
It would save me much time if you find a different solution. Looks like some folks have suggested that we simply have the files indexed by a web search engine. I understand, but that's a different usage. Adobe ASDOC has both: quick search field on top of the class list, and global search box top right of the page, near the Adobe logo. http://help.adobe.com/en_US/FlashPlatform/reference/actionscript/3/ But I don't want to add you more work (and to me as well), so I leave this for now. Thanks for your advice. Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : lundi 24 mars 2014 17:45 À : dev@flex.apache.org; n...@joeflash.ca Objet : Re: Search in Apache Flex Docs? It would save me much time if you find a different solution. Looks like some folks have suggested that we simply have the files indexed by a web search engine. -Alex On 3/24/14 2:38 AM, Maurice Amsellem maurice.amsel...@systar.com wrote: I could look into it, but it wouldn't be my highest priority. Thanks Alex, I have raised a JIRA ticket, for future reference: https://issues.apache.org/jira/browse/FLEX-34183 Do you think I can start working on it when I have time, maybe on a simpler implementation, of better to wait for Adobe's answer? Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : lundi 24 mars 2014 04:20 À : dev@flex.apache.org; n...@joeflash.ca Objet : Re: Search in Apache Flex Docs? On 3/23/14 11:41 AM, Maurice Amsellem maurice.amsel...@systar.com wrote: The Adobe ASLR does have search. I wonder what it would take to implement it. It does not seem to be that complicated ( file asdoc.js, doing a regexp match on a tags in the left list, and setting display style accordingly, to show or hide an item). Sorry for insisting, but could we get that asdoc extensions donated by Adobe, or at least permission to use them in Apache Flex ASDOC, or do we have to rewrite them? I could look into it, but it wouldn't be my highest priority. -Alex
Re: FlexUnit was Unit tests in TLF
On 3/24/14 9:22 AM, Michael A. Labriola labri...@digitalprimates.net wrote: BTW, was one of the enhancements automated test generation? Maybe from metadata on the class? There was a bigger project on that, yes, although we didn't get particularly far before the Apache switch so most of it was put on hold. Also, Flash Builder would create stubs for you based on the code model it had, so we don't go that route. I'll have to ponder whether it would be faster to manually write and maintain these tests or write a test generator. It seems like you could annotate with metadata and auto-generate a fair number of tests that way. BTW, a key motivating factor in creating a set of FlexUnit tests is to try to conform to that government standard you told me about. Do you have a 5 sentence summary of what a test suite needs to cover? Is it measured purely by code coverage or do you have to have boundary tests, negative tests, etc? -Alex
Re: Unit tests in TLF
On 3/24/14 9:26 AM, Michael A. Labriola labri...@digitalprimates.net wrote: OK, I didn't see any non-Flex apps around the runners. The deal is the mx and Spark apps can't wrap the runners and load in FlexJS classes at the same time. I think I'll just have to replace the Spark app with a FlexJS app. No problem. We even got FlexUnit working in Flash pro at one point. I don't know of any resultant JS at this point. I did wonder about cross-compiling FlexUnit and the tests via FalconJX/FlexJS. Do you think that would work? Perhaps, the biggest problem is the metadata introspection. If you have a way to mimic that in JS, there would be no issue... but that's the blocker that stopped me from going that route FlexJS will keep metadata on the JS side, probably as simple properties on the classes and/or functions. I imagine we'll write a library that abstracts the fetching of metadata and returns it as an array of strings or maybe as JSON. If we could easily retrofit the metadata fetching in FlexUnit, would going this route become more interesting? Or should we actually use some JS testing framework for the hand-written JS pieces? I would recommend looking into Jasmine for this. It's what we did on the Randori front OK, I'll keep Jasmine in mind. IMO, being able to write JS tests with a similar pattern/workflow as FlexUnit would be good. A quick survey didn't really turn up anything that looked like FlexUnit. QUnit is about the closest OK, thanks. Any idea why the JS community did not leverage Junit? Did they all just fall in love with newer approaches like behaviors? -Alex
Re: [FlexJS] Checkintest now runs on both AS and JS
Alex, I'm trying to keep up with developments, and I'll jump in when I have a few cycles, hopefully in the next few days. EdB On Fri, Mar 21, 2014 at 8:05 PM, Alex Harui aha...@adobe.com wrote: Hi folks, I just checked in changes to flex-falcon and flex-asjs so that the checkintests target in flex-asjs uses Mustella to run BasicTests in Flash and then cross-compiles it and runs it in natively in the browser (currently assumes you have FireFox). Please try it so we can work out any kinks. I heavily borrowed from the Marmotinni folder to get this to work, so thanks to Erik for figuring out how to get Selenium to work. Hopefully Erik can take a look and tell me how I should have used more from Marmotinni. I ended up not using the -marmotinni compiler option or many of the tasks used in this demo. I'm not sure if that's because the tooling evolved such that those tasks I didn't need are obsolete, or if there's some goodness about Marmotinni I'm missing. Of course, BasicTests needs more tests so if anyone has time, please help out. Thanks, -Alex -- Ix Multimedia Software Jan Luykenstraat 27 3521 VB Utrecht T. 06-51952295 I. www.ixsoftware.nl
Re: [FlexJS] Tasks before first Release Candidate
1) Do some package reorganization: Currently most components are in org.apache.flex.html.staticControls. However these controls aren't quite so static, so I'm thinking of removing the staticControls folder and flattening the hierarchy by one level. All for it. Be sure to freeze the tree and notify Peter, at least, when you start this, as we'll land in merge hell pretty quickly otherwise, git or no git. 2) Choose a different/better xml prefix: Right now we are using xmlns=basic. That's five characters which is longer that I'd like. I've thought about replacing with fjs or 'af' or 'js' or even 'fxjs' but a shorter prefix means less typing. I still have some hope that once FlexJS is off to a good start I'll get to try my VanillaSDK approach, so my vote goes to 'fjs' - very short and still sorta descriptive; it also leaves 'vjs' available ;-) 4) Take a look at not using MXMLC and going for it with Falcon. This would make the install simpler and might make it easier for other IDEs to work with FlexJS, but we could end up dealing with a lot of Falcon issues. I'd say those issues would be part of the whole FlexJS release cycle, and if we don't do it then, we'll probably end up postponing it again and again. Just make the transition part of the first release. Exciting stuff! EdB -- Ix Multimedia Software Jan Luykenstraat 27 3521 VB Utrecht T. 06-51952295 I. www.ixsoftware.nl
Re: How to get the attachments of old Adobe JIRA / imported Apache JIRA?
XML file: https://www.dropbox.com/s/rhnonjo9p09krws/FlexSDKENTDMV.zip Attachments: https://www.dropbox.com/s/c0b2h93rbjqwcdz/FlexAttachments.zip The XML file format is not well documented, but I eventually figured it out. Tags reference other tags and eventually map the file name to the attachment number which is in a folder name that is the same as the issue number. Good luck and thanks for looking into it, -Alex On 3/24/14 9:44 AM, Marcus Fritze marcus.fri...@googlemail.com wrote: Thanks for the explanation. Is it possible to download the 7GB and the XML to search manually? And maybe - if it's not to complicated - to write that utility. Thanks Marcus Am 24.03.2014 um 17:18 schrieb Alex Harui aha...@adobe.com: Essentially, every attachment is missing. When Infra did the import of the Flex issues, they had the attachments, but assumed that they could add the attachments after import (which worked in prior versions of JIRA) and then found out that wasn't true in the current version. There is about 7GB of attachments and Infra does not want us to use REST or SOAP to re-insert the attachments because that's a lot of data and because there is no way to recover if we get it wrong. Flex JIRA issues are in the same database as all of the other Apache projects and shutting them down to recover the Flex JIRA is not desirable. The plan is that someday, I (or someone) will find time to write a utility that renames the attachments to their proper names (JIRA just stores them as 1, 2, 3, etc, no suffixes or other hints as to the content, the name is determined from an XML file), and then Apache will post those files somewhere. Given that in two years we've only been asked for less than 10 attachments, it is low on my list of priorities. -Alex On 3/24/14 3:16 AM, Christofer Dutz christofer.d...@c-ware.de wrote: When migrating the Flexmojos Issues from Sonatype to the new Atlassian Jira, I wrote a Manual Migration tool that also transfered attachments using the REST api If there is some stuff missing, we could code a similar tool to Transfer missing attachments. Chris Von: Justin Mclean jus...@classsoftware.com Gesendet: Freitag, 21. März 2014 13:39 An: dev@flex.apache.org Betreff: Re: How to get the attachments of old Adobe JIRA / imported Apache JIRA? Hi, Just post the issue number(s) and Alex should be able to find it for you. Justin On 21 Mar 2014, at 9:21 pm, Marcus Fritze marcus.fri...@googlemail.com wrote: Hi list, unfortunately the attachments to the issues are still not accessible in the JIRA. Is there a easy way to get attachments of some (3-5) issues? Thanks Marcus
RE: Unit tests in TLF
FlexJS will keep metadata on the JS side, probably as simple properties on the classes and/or functions. I imagine we'll write a library that abstracts the fetching of metadata and returns it as an array of strings or maybe as JSON. If we could easily retrofit the metadata fetching in FlexUnit, would going this route become more interesting? It could be only as it will keep the parallelism. I am of two minds about it. Ultimately in some of our previous cross-compilation work we decided to cross-compile on top of a native javascript library OK, thanks. Any idea why the JS community did not leverage Junit? Did they all just fall in love with newer approaches like behaviors? Yes to behaviors but also because they are not cross-compiling so they don't have Annotations and the like which JUnit relies upon. Also, honestly, when writing JS by hand you wouldn't write it in the same ways that a Flex or Java app is written. It tends to be more functional and less OO. Further, the need for mocking and other unit testing ideas in Java are less needed in a JS environment without access restrictions and with first class functions. Mike
RE: FlexUnit was Unit tests in TLF
BTW, a key motivating factor in creating a set of FlexUnit tests is to try to conform to that government standard you told me about. Do you have a 5 sentence summary of what a test suite needs to cover? Is it measured purely by code coverage or do you have to have boundary tests, negative tests, etc? It was mostly governed by unit code coverage. I can do a little work and see if I can find the terminology again. Some areas are more strict than others. Mike
ApacheCon Slides
ApacheCon required that we upload our slide decks by today. I just posted my slides. If you have time, please take a look and let me know if you think I should make changes. The FlexJS session is essentially the same as the 360|Stack presentation from last year. What's Happening in the Apache Flex Project http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_Fle x.pdf FlexJS: Flex For javaScript: http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_Fle xJS.pdf Thanks, -Alex
Re: ApacheCon Slides
I would like to see mention of all the activities around Flex Mobile: - New components/features (Mobile Grid, Scrollable text) - iOS7 status bar support - Support for low res (120DPI) and hi-res (480, 640 DPI) devices - CSS-Media queries for OS version - CSS-Media queries for device dimensions - New Android 4.x skins (ongoing work) - New iOS7 skins (coming up next) Thanks, Om On Mon, Mar 24, 2014 at 1:03 PM, Alex Harui aha...@adobe.com wrote: ApacheCon required that we upload our slide decks by today. I just posted my slides. If you have time, please take a look and let me know if you think I should make changes. The FlexJS session is essentially the same as the 360|Stack presentation from last year. What's Happening in the Apache Flex Project http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_Fle x.pdf FlexJS: Flex For javaScript: http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_Fle xJS.pdf Thanks, -Alex
Re: ApacheCon Slides
Looks good @Alex. I just hope that the full session, including the apps on your demo for both sessions would be available to everyone after the event. On Mon, Mar 24, 2014 at 2:12 PM, OmPrakash Muppirala bigosma...@gmail.comwrote: I would like to see mention of all the activities around Flex Mobile: - New components/features (Mobile Grid, Scrollable text) - iOS7 status bar support - Support for low res (120DPI) and hi-res (480, 640 DPI) devices - CSS-Media queries for OS version - CSS-Media queries for device dimensions - New Android 4.x skins (ongoing work) - New iOS7 skins (coming up next) Thanks, Om On Mon, Mar 24, 2014 at 1:03 PM, Alex Harui aha...@adobe.com wrote: ApacheCon required that we upload our slide decks by today. I just posted my slides. If you have time, please take a look and let me know if you think I should make changes. The FlexJS session is essentially the same as the 360|Stack presentation from last year. What's Happening in the Apache Flex Project http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_Fle x.pdf FlexJS: Flex For javaScript: http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_Fle xJS.pdf Thanks, -Alex
Re: ApacheCon Slides
For FlexJS, it would be nice if you can use the beads/strands diagrams from here [1] They do a good job of explaining the concepts visually. Thanks, Om [1] https://cwiki.apache.org/confluence/display/FLEX/Creating+Components On Mon, Mar 24, 2014 at 1:12 PM, OmPrakash Muppirala bigosma...@gmail.comwrote: I would like to see mention of all the activities around Flex Mobile: - New components/features (Mobile Grid, Scrollable text) - iOS7 status bar support - Support for low res (120DPI) and hi-res (480, 640 DPI) devices - CSS-Media queries for OS version - CSS-Media queries for device dimensions - New Android 4.x skins (ongoing work) - New iOS7 skins (coming up next) Thanks, Om On Mon, Mar 24, 2014 at 1:03 PM, Alex Harui aha...@adobe.com wrote: ApacheCon required that we upload our slide decks by today. I just posted my slides. If you have time, please take a look and let me know if you think I should make changes. The FlexJS session is essentially the same as the 360|Stack presentation from last year. What's Happening in the Apache Flex Project http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_Fle x.pdfhttp://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_Flex.pdf FlexJS: Flex For javaScript: http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_Fle xJS.pdfhttp://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_FlexJS.pdf Thanks, -Alex
RE: ApacheCon Slides
Flex.pdf Page 6: Eclipse-based IDE = not only (IDEA, FlashDevelop, etc..) Overall, the impression it gave me is that it's not describing what's happening in Apache Flex Project, but rather an advertisement to attract new developers by telling them Hey, there is not only AS in Flex, there are also other technologies, Java, Maven, so don't be afraid to come help us. Should we be ashamed of Flex being Flex, ActionScript and Flash ? FlexJS.pdf: I think there is a major problem with these slides, starting from page 1: Flex without Flash Flash based solutions no longer desirable. Etc.. For many people people Flex == Flash. So if you say flash is not desirable anymore, they will understand Flex is not desirable anymore. I don't think this is what we are want. Sorry for being harsh, but I think we have had enough with Steve Jobs and Abobe PR. Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : lundi 24 mars 2014 21:03 À : dev@flex.apache.org Objet : ApacheCon Slides ApacheCon required that we upload our slide decks by today. I just posted my slides. If you have time, please take a look and let me know if you think I should make changes. The FlexJS session is essentially the same as the 360|Stack presentation from last year. What's Happening in the Apache Flex Project http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_Fle x.pdf FlexJS: Flex For javaScript: http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_Fle xJS.pdf Thanks, -Alex
Re: ApacheCon Slides
On 3/24/14 1:30 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: Flex.pdf Page 6: Eclipse-based IDE = not only (IDEA, FlashDevelop, etc..) Overall, the impression it gave me is that it's not describing what's happening in Apache Flex Project, but rather an advertisement to attract new developers by telling them Hey, there is not only AS in Flex, there are also other technologies, Java, Maven, so don't be afraid to come help us. Should we be ashamed of Flex being Flex, ActionScript and Flash ? No, it isn't about being ashamed. Given the kinds of people I met at the one ApacheCon I went to, I biased this presentation towards folks who don't know Flex, AS, MXML, and Flash. FlexJS.pdf: I think there is a major problem with these slides, starting from page 1: Flex without Flash Flash based solutions no longer desirable. Etc.. For many people people Flex == Flash. So if you say flash is not desirable anymore, they will understand Flex is not desirable anymore. I don't think this is what we are want. Sorry for being harsh, but I think we have had enough with Steve Jobs and Abobe PR. Hmm. I think the whole point of FlexJS is to create a framework that runs without Flash and a key factor is the impact from decisions and statements made by Jobs and Adobe. Are you saying that enough time has passed that these decisions and statements are no longer a factor? I'm not hearing that. Folks still seem to want to make sure their future isn't tied to Flash. But I probably don't talk to as many customers as you might. -Alex
RE: ApacheCon Slides
+1 for the mobile improvements -Message d'origine- De : omup...@gmail.com [mailto:omup...@gmail.com] De la part de OmPrakash Muppirala Envoyé : lundi 24 mars 2014 21:13 À : dev@flex.apache.org Objet : Re: ApacheCon Slides I would like to see mention of all the activities around Flex Mobile: - New components/features (Mobile Grid, Scrollable text) - iOS7 status bar support - Support for low res (120DPI) and hi-res (480, 640 DPI) devices - CSS-Media queries for OS version - CSS-Media queries for device dimensions - New Android 4.x skins (ongoing work) - New iOS7 skins (coming up next) Thanks, Om On Mon, Mar 24, 2014 at 1:03 PM, Alex Harui aha...@adobe.com wrote: ApacheCon required that we upload our slide decks by today. I just posted my slides. If you have time, please take a look and let me know if you think I should make changes. The FlexJS session is essentially the same as the 360|Stack presentation from last year. What's Happening in the Apache Flex Project http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon1 4_Fle x.pdf FlexJS: Flex For javaScript: http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon1 4_Fle xJS.pdf Thanks, -Alex
RE: RTL support in mobile apps
I just did a quick test to compare TLF and TextField on mobile. Basically, replaced StyleableTextField cell renderer on MobileGrid by spark Label-based renderer. Test results: - iPad 3 (retina) - slow iOS packaging , GPU rendering - Mobile grid with 4 columns of text, and 200 rows StyleableTextField = 25 fps when scrolling Spark Label = 1 to 3 fps when scrolling ( UI is very slow, almost frozen). So of course mobile grid displays a lot of text, including multi-line, but that's where performance is needed, not on button and titles, IMO. I could also have used TextLine, but it does not support multi-line, which TextField does, so it's not equivalent. So for me, spark Label is not good enough on mobile, even on recent devices. I will explore the other track (RTL using TextField). What do you think? Maurice -Message d'origine- De : Maurice Amsellem [mailto:maurice.amsel...@systar.com] Envoyé : lundi 24 mars 2014 11:19 À : dev@flex.apache.org Objet : RE: RTL support in mobile apps Hi Carlos, 1) It's not proven yet that TLF is fast enough on mobile, especially when there are lots of text to display, such as in lists of datagrid. Plus I have discovered that the old TextField is actually capable to display RTL , but the Flex positioning is broken, so the text does not appear (probably because it was not supposed to work that way). So IMO, the question is still open, and I won't rush into replacing TextField by TLF on mobile. It would be probably much simpler to fix the layout. Right now we need to deal in different ways with TextInput in mobile and browser and this defeat the code once run everywhere. What do you mean? From the SDK developer standpoint, or from the end-user developer stand point ? From the SDK standpoint, the difference is only on the skin, the 'host' component is the same. From the end-user developer, you must use TextInput in both cases, so where's the difference ? The behavior is different, but that's inherent to mobile vs desktop (eg. you don't have softkeyboard or restricted keyboards on desktop). Please explain Maurice -Message d'origine- De : carlos.rov...@gmail.com [mailto:carlos.rov...@gmail.com] De la part de Carlos Rovira Envoyé : lundi 24 mars 2014 10:54 À : dev@flex.apache.org Objet : Re: RTL support in mobile apps Hi, if there are plans to introduce TLF on mobile TextInput this will change my priorities about change the internals of MaskedTextInput component and will only make it to preserve slot positions. IMO, if now TLF give us a good performance in mobile this days it will be very useful to make it happen since this will be more aligned to the Flex philosophy. Right now we need to deal in different ways with TextInput in mobile and browser and this defeat the code once run everywhere. So +1 to TLF support on mobile is performance is good! :) Please let me know if that's are the plans. Thanks! Carlos 2014-03-23 21:08 GMT+01:00 Maurice Amsellem maurice.amsel...@systar.com: Found a number of tickets on this topic: https://issues.apache.org/jira/browse/FLEX-26365 (closed as later) https://issues.apache.org/jira/browse/FLEX-34145 (closed) https://issues.apache.org/jira/browse/FLEX-34181 (In progress) https://issues.apache.org/jira/browse/FLEX-33750 (open) https://issues.apache.org/jira/browse/FLEX-28107 (later) https://issues.apache.org/jira/browse/FLEX-28103 (later) https://issues.apache.org/jira/browse/FLEX-26169 (later) https://issues.apache.org/jira/browse/FLEX-24502 (later) Maurice -Message d'origine- De : Maurice Amsellem [mailto:maurice.amsel...@systar.com] Envoyé : samedi 22 mars 2014 01:14 À : dev@flex.apache.org Objet : RE: RTL support in mobile apps Yes, that might me the answer: so I need to cancel the flipping like I did for StageText (and like is done in spark Label). I will try this tomorrow. Still does not explain why TextField accepts bidi text now ? Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : samedi 22 mars 2014 01:09 À : dev@flex.apache.org Objet : Re: RTL support in mobile apps Again, I was not highly involved in this code, but IIRC, the TextLines are never flipped, so if you choose a flipped layoutDirection the TextLines are re-flipped. But if you start flipping TextFields without embedded text they go blank. Does that explain what you're seeing? -Alex On 3/21/14 5:00 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: Thanks Alex. That was also my understanding. Regarding TextInput / TextArea, there is no issue with regard to RTL in using StageText ( embedded in StyleableStageText or ScrollableStageText) . Now something strange that gets me puzzled. I did some experiments with mobile components that use TextField (actually StyleableTextField) and I managed to displayed Arabic/Hebrew (list , titles and nav bar) https://www.dropbox.com/s/4e4untcp3f4jeb2/List_arabic_LTR.png
Re: ApacheCon Slides
Thanks for the feedback. That diagram is a bit lower-level than I was going to go, but I'm interested to hear why that's an important piece for folks attending a 45 minute session. I do have another diagram from my 360|Stack slides that tries to show the class substitution done by the compiler that I might add. -Alex On 3/24/14 1:19 PM, OmPrakash Muppirala bigosma...@gmail.com wrote: For FlexJS, it would be nice if you can use the beads/strands diagrams from here [1] They do a good job of explaining the concepts visually. Thanks, Om [1] https://cwiki.apache.org/confluence/display/FLEX/Creating+Components On Mon, Mar 24, 2014 at 1:12 PM, OmPrakash Muppirala bigosma...@gmail.comwrote: I would like to see mention of all the activities around Flex Mobile: - New components/features (Mobile Grid, Scrollable text) - iOS7 status bar support - Support for low res (120DPI) and hi-res (480, 640 DPI) devices - CSS-Media queries for OS version - CSS-Media queries for device dimensions - New Android 4.x skins (ongoing work) - New iOS7 skins (coming up next) Thanks, Om On Mon, Mar 24, 2014 at 1:03 PM, Alex Harui aha...@adobe.com wrote: ApacheCon required that we upload our slide decks by today. I just posted my slides. If you have time, please take a look and let me know if you think I should make changes. The FlexJS session is essentially the same as the 360|Stack presentation from last year. What's Happening in the Apache Flex Project http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_ Fle x.pdfhttp://events.linuxfoundation.org/sites/events/files/slides/Apache Con14_Flex.pdf FlexJS: Flex For javaScript: http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_ Fle xJS.pdfhttp://events.linuxfoundation.org/sites/events/files/slides/Apac heCon14_FlexJS.pdf Thanks, -Alex
Re: RTL support in mobile apps
I assume this was a release version and not a debug version? Either way, I don't think TLF will get out to 25fps. I'd suggest doing a simple test to see if TextField really can do RTL (text starting from the right edge) or just knows how to place characters in a string based on some positioning information. -Alex On 3/24/14 1:46 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: I just did a quick test to compare TLF and TextField on mobile. Basically, replaced StyleableTextField cell renderer on MobileGrid by spark Label-based renderer. Test results: - iPad 3 (retina) - slow iOS packaging , GPU rendering - Mobile grid with 4 columns of text, and 200 rows StyleableTextField = 25 fps when scrolling Spark Label = 1 to 3 fps when scrolling ( UI is very slow, almost frozen). So of course mobile grid displays a lot of text, including multi-line, but that's where performance is needed, not on button and titles, IMO. I could also have used TextLine, but it does not support multi-line, which TextField does, so it's not equivalent. So for me, spark Label is not good enough on mobile, even on recent devices. I will explore the other track (RTL using TextField). What do you think? Maurice -Message d'origine- De : Maurice Amsellem [mailto:maurice.amsel...@systar.com] Envoyé : lundi 24 mars 2014 11:19 À : dev@flex.apache.org Objet : RE: RTL support in mobile apps Hi Carlos, 1) It's not proven yet that TLF is fast enough on mobile, especially when there are lots of text to display, such as in lists of datagrid. Plus I have discovered that the old TextField is actually capable to display RTL , but the Flex positioning is broken, so the text does not appear (probably because it was not supposed to work that way). So IMO, the question is still open, and I won't rush into replacing TextField by TLF on mobile. It would be probably much simpler to fix the layout. Right now we need to deal in different ways with TextInput in mobile and browser and this defeat the code once run everywhere. What do you mean? From the SDK developer standpoint, or from the end-user developer stand point ? From the SDK standpoint, the difference is only on the skin, the 'host' component is the same. From the end-user developer, you must use TextInput in both cases, so where's the difference ? The behavior is different, but that's inherent to mobile vs desktop (eg. you don't have softkeyboard or restricted keyboards on desktop). Please explain Maurice -Message d'origine- De : carlos.rov...@gmail.com [mailto:carlos.rov...@gmail.com] De la part de Carlos Rovira Envoyé : lundi 24 mars 2014 10:54 À : dev@flex.apache.org Objet : Re: RTL support in mobile apps Hi, if there are plans to introduce TLF on mobile TextInput this will change my priorities about change the internals of MaskedTextInput component and will only make it to preserve slot positions. IMO, if now TLF give us a good performance in mobile this days it will be very useful to make it happen since this will be more aligned to the Flex philosophy. Right now we need to deal in different ways with TextInput in mobile and browser and this defeat the code once run everywhere. So +1 to TLF support on mobile is performance is good! :) Please let me know if that's are the plans. Thanks! Carlos 2014-03-23 21:08 GMT+01:00 Maurice Amsellem maurice.amsel...@systar.com: Found a number of tickets on this topic: https://issues.apache.org/jira/browse/FLEX-26365 (closed as later) https://issues.apache.org/jira/browse/FLEX-34145 (closed) https://issues.apache.org/jira/browse/FLEX-34181 (In progress) https://issues.apache.org/jira/browse/FLEX-33750 (open) https://issues.apache.org/jira/browse/FLEX-28107 (later) https://issues.apache.org/jira/browse/FLEX-28103 (later) https://issues.apache.org/jira/browse/FLEX-26169 (later) https://issues.apache.org/jira/browse/FLEX-24502 (later) Maurice -Message d'origine- De : Maurice Amsellem [mailto:maurice.amsel...@systar.com] Envoyé : samedi 22 mars 2014 01:14 À : dev@flex.apache.org Objet : RE: RTL support in mobile apps Yes, that might me the answer: so I need to cancel the flipping like I did for StageText (and like is done in spark Label). I will try this tomorrow. Still does not explain why TextField accepts bidi text now ? Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : samedi 22 mars 2014 01:09 À : dev@flex.apache.org Objet : Re: RTL support in mobile apps Again, I was not highly involved in this code, but IIRC, the TextLines are never flipped, so if you choose a flipped layoutDirection the TextLines are re-flipped. But if you start flipping TextFields without embedded text they go blank. Does that explain what you're seeing? -Alex On 3/21/14 5:00 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: Thanks Alex. That was also my understanding. Regarding TextInput / TextArea, there is no issue with regard to RTL
Re: [DISCUSSION] Release Apache Flex FlexUnit 4.2.0 RC2
Hi, In a work-for-hire model, it isn't really a donation, and the hiring party generally owns the code unless there is an agreement that says otherwise. Assuming this is the case Alex can you get Adobe to officially donate it? However this may not be an issue as FlexUnit (and I assume this code) is under a BSD license. http://sourceforge.net/adobe/flexunit/wiki/License/ Can anyone confirm this? BSD is a category A license and thus all that would be required to to add Adobe to NOTICE file. FlexUnit has been lingering far too long in limbo and I'd like to see it released. Thanks, Justin
AW: ApacheCon Slides
Hi Guys, I just uploaded my stuf a few minutes ago: http://events.linuxfoundation.org/sites/events/files/slides/Building%20Enterprise%20Level%20Applications%20with%20Apache%20Flex%20and%20Java-v9-20140313_1547.pdf (Noticed too late that I should have shortended the file name ;-)) I thought I'd concentrate on why Flex is great for building Enterprise applications and only briefly go into Flexmojos as this is an Apache Conference. Doing a full Flexmojos introduction would have exploded the timeframe, so I decided to describe the ecosystem, what the different fragments are and how they all together create the big picture. Hopefully it will make a few people more interesed in Flex and guide them to the other talks (I added references tot he other talks at the end) Feedback greatly appreciated. Chris -Ursprüngliche Nachricht- Von: Alex Harui [mailto:aha...@adobe.com] Gesendet: Montag, 24. März 2014 21:03 An: dev@flex.apache.org Betreff: ApacheCon Slides ApacheCon required that we upload our slide decks by today. I just posted my slides. If you have time, please take a look and let me know if you think I should make changes. The FlexJS session is essentially the same as the 360|Stack presentation from last year. What's Happening in the Apache Flex Project http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_Fle x.pdf FlexJS: Flex For javaScript: http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_Fle xJS.pdf Thanks, -Alex
RE: ApacheCon Slides
No, it isn't about being ashamed. Given the kinds of people I met at the one ApacheCon I went to, I biased this presentation towards folks who don't know Flex, AS, MXML, and Flash. Ok, so it's on purpose. Hmm. I think the whole point of FlexJS is to create a framework that runs without Flash and a key factor is the impact from decisions and statements made by Jobs and Adobe. Are you saying that enough time has passed that these decisions and statements are no longer a factor? I'm not hearing that. Folks still seem to want to make sure their future isn't tied to Flash. But I probably don't talk to as many customers as you might. It's a difficult matter, not easy to discuss by email, even more so writing not in my mother tongue, so sorry in advance if I get misunderstood. You can say that FlexJS will work also on JavaScript, and I think that alone is enough to attract people, you don't need to shoot down on Flash. FlexJS is the power of MXML and AS brought to the javascript world And this has nothing to do with Flash, actually, but rather on the current weaknesses of javascript. Once again, for many non developers around me, they don't make the difference between Flex and Flash, and statements like this one, especially coming from Apache Flex PMC himself, could be harmful if they fall in non-educated ears. Will it decrease the value of FlexJS, if you don't say that Flash is not desirable anymore? What do others think? Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : lundi 24 mars 2014 21:42 À : dev@flex.apache.org Objet : Re: ApacheCon Slides On 3/24/14 1:30 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: Flex.pdf Page 6: Eclipse-based IDE = not only (IDEA, FlashDevelop, etc..) Overall, the impression it gave me is that it's not describing what's happening in Apache Flex Project, but rather an advertisement to attract new developers by telling them Hey, there is not only AS in Flex, there are also other technologies, Java, Maven, so don't be afraid to come help us. Should we be ashamed of Flex being Flex, ActionScript and Flash ? No, it isn't about being ashamed. Given the kinds of people I met at the one ApacheCon I went to, I biased this presentation towards folks who don't know Flex, AS, MXML, and Flash. FlexJS.pdf: I think there is a major problem with these slides, starting from page 1: Flex without Flash Flash based solutions no longer desirable. Etc.. For many people people Flex == Flash. So if you say flash is not desirable anymore, they will understand Flex is not desirable anymore. I don't think this is what we are want. Sorry for being harsh, but I think we have had enough with Steve Jobs and Abobe PR. Hmm. I think the whole point of FlexJS is to create a framework that runs without Flash and a key factor is the impact from decisions and statements made by Jobs and Adobe. Are you saying that enough time has passed that these decisions and statements are no longer a factor? I'm not hearing that. Folks still seem to want to make sure their future isn't tied to Flash. But I probably don't talk to as many customers as you might. -Alex
Re: ApacheCon Slides
Hi, Hmm. I think the whole point of FlexJS is to create a framework that runs without Flash and a key factor is the impact from decisions and statements made by Jobs and Adobe. Are you saying that enough time has passed that these decisions and statements are no longer a factor? IMO 40,000+ installs of the Flex SDK say otherwise and not all of those users are going to be interested in FlexJS particularly as it's not going to be a 1 for 1 replacement. Slide 8 you might want to add something about bug fixed/improvements over Adobe 4.6. Thanks, Justin
Re: [DISCUSSION] Release Apache Flex FlexUnit 4.2.0 RC2
On 3/24/14 2:02 PM, Justin Mclean jus...@classsoftware.com wrote: Hi, In a work-for-hire model, it isn't really a donation, and the hiring party generally owns the code unless there is an agreement that says otherwise. Assuming this is the case Alex can you get Adobe to officially donate it? I will probably have to find time for that (and the FlexUnit1 code as well). However this may not be an issue as FlexUnit (and I assume this code) is under a BSD license. http://sourceforge.net/adobe/flexunit/wiki/License/ Can anyone confirm this? BSD is a category A license and thus all that would be required to to add Adobe to NOTICE file. Yes, FlexUnit1 is BSD. I'm not sure it matters what FlexUnit4 was, since in theory it is now under AL. The issue is that there code is in our repo that might have needed a more official contribution agreement when it went from Adobe to DP to Apache. FlexUnit has been lingering far too long in limbo and I'd like to see it released. And I'd like to close the books on Installer 3.0. Please try to find time to look at the one issue you raised. If I don't hear from you by tonight I'll send results out in my morning anyway. But as agreed we won't announce until 4.12.1. -Alex
Re: [FlexJS] Tasks before first Release Candidate
Interesting. Two replies with different view points. I have another factor that occurred to me. I think we should try to get a release out ASAP so that folks can have a chance to look at it and get interested and consider a last minute signup for 360|Flex (assuming there are still spots). That said, I think I'll spend this week on some of these items and then start on an RC. -Alex On 3/24/14 10:13 AM, Erik de Bruin e...@ixsoftware.nl wrote: 2) Choose a different/better xml prefix: Right now we are using xmlns=basic. That's five characters which is longer that I'd like. I've thought about replacing with fjs or 'af' or 'js' or even 'fxjs' but a shorter prefix means less typing. I still have some hope that once FlexJS is off to a good start I'll get to try my VanillaSDK approach, so my vote goes to 'fjs' - very short and still sorta descriptive; it also leaves 'vjs' available ;-) I am still hopeful that, instead of trying to replicate Spark exactly, that you'll settle for something short of that and see how much you can build out of the FlexJS pieces we have. ;-)
RE: RTL support in mobile apps
I assume this was a release version and not a debug version? Damn, I fell in the trap again. Thanks for reminding me. I have re-done the tests with release packaging, almost same results: 21- 25 FPS for TextField 1 ~ 4 for spark Label. Either way, I don't think TLF will get out to 25fps. Yes, 15 would have been fine. but 4 fps is really too bad. I'd suggest doing a simple test to see if TextField really can do RTL (text starting from the right edge) or just knows how to place characters in a string based on some positioning information. Yes, that was my intention. Crossing fingers that it works. Thanks Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : lundi 24 mars 2014 21:56 À : dev@flex.apache.org Objet : Re: RTL support in mobile apps I assume this was a release version and not a debug version? Either way, I don't think TLF will get out to 25fps. I'd suggest doing a simple test to see if TextField really can do RTL (text starting from the right edge) or just knows how to place characters in a string based on some positioning information. -Alex On 3/24/14 1:46 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: I just did a quick test to compare TLF and TextField on mobile. Basically, replaced StyleableTextField cell renderer on MobileGrid by spark Label-based renderer. Test results: - iPad 3 (retina) - slow iOS packaging , GPU rendering - Mobile grid with 4 columns of text, and 200 rows StyleableTextField = 25 fps when scrolling Spark Label = 1 to 3 fps when scrolling ( UI is very slow, almost frozen). So of course mobile grid displays a lot of text, including multi-line, but that's where performance is needed, not on button and titles, IMO. I could also have used TextLine, but it does not support multi-line, which TextField does, so it's not equivalent. So for me, spark Label is not good enough on mobile, even on recent devices. I will explore the other track (RTL using TextField). What do you think? Maurice -Message d'origine- De : Maurice Amsellem [mailto:maurice.amsel...@systar.com] Envoyé : lundi 24 mars 2014 11:19 À : dev@flex.apache.org Objet : RE: RTL support in mobile apps Hi Carlos, 1) It's not proven yet that TLF is fast enough on mobile, especially when there are lots of text to display, such as in lists of datagrid. Plus I have discovered that the old TextField is actually capable to display RTL , but the Flex positioning is broken, so the text does not appear (probably because it was not supposed to work that way). So IMO, the question is still open, and I won't rush into replacing TextField by TLF on mobile. It would be probably much simpler to fix the layout. Right now we need to deal in different ways with TextInput in mobile and browser and this defeat the code once run everywhere. What do you mean? From the SDK developer standpoint, or from the end-user developer stand point ? From the SDK standpoint, the difference is only on the skin, the 'host' component is the same. From the end-user developer, you must use TextInput in both cases, so where's the difference ? The behavior is different, but that's inherent to mobile vs desktop (eg. you don't have softkeyboard or restricted keyboards on desktop). Please explain Maurice -Message d'origine- De : carlos.rov...@gmail.com [mailto:carlos.rov...@gmail.com] De la part de Carlos Rovira Envoyé : lundi 24 mars 2014 10:54 À : dev@flex.apache.org Objet : Re: RTL support in mobile apps Hi, if there are plans to introduce TLF on mobile TextInput this will change my priorities about change the internals of MaskedTextInput component and will only make it to preserve slot positions. IMO, if now TLF give us a good performance in mobile this days it will be very useful to make it happen since this will be more aligned to the Flex philosophy. Right now we need to deal in different ways with TextInput in mobile and browser and this defeat the code once run everywhere. So +1 to TLF support on mobile is performance is good! :) Please let me know if that's are the plans. Thanks! Carlos 2014-03-23 21:08 GMT+01:00 Maurice Amsellem maurice.amsel...@systar.com: Found a number of tickets on this topic: https://issues.apache.org/jira/browse/FLEX-26365 (closed as later) https://issues.apache.org/jira/browse/FLEX-34145 (closed) https://issues.apache.org/jira/browse/FLEX-34181 (In progress) https://issues.apache.org/jira/browse/FLEX-33750 (open) https://issues.apache.org/jira/browse/FLEX-28107 (later) https://issues.apache.org/jira/browse/FLEX-28103 (later) https://issues.apache.org/jira/browse/FLEX-26169 (later) https://issues.apache.org/jira/browse/FLEX-24502 (later) Maurice -Message d'origine- De : Maurice Amsellem [mailto:maurice.amsel...@systar.com] Envoyé : samedi 22 mars 2014 01:14 À : dev@flex.apache.org Objet : RE: RTL support in mobile apps Yes, that might me the answer: so I need to cancel the
Re: [DISCUSSION] Release Apache Flex FlexUnit 4.2.0 RC2
Hi, I will probably have to find time for that (and the FlexUnit1 code as well). While having FlexUnit1 would removed a dependancy it's not a potential blocker to releasing FlexUnit as this may be. Do we know who even worked on this at Adobe? The issue is that there code is in our repo that might have needed a more official contribution agreement when it went from Adobe to DP to Apache. Do we know how Adobe licensed the code to Digital Primates? And I'd like to close the books on Installer 3.0. Please try to find time to look at the one issue you raised. As far as I can tell but not 100% sure it may happen when you have a mismatch between what AIR_HOME points to and what version of AIR FLEX_HOME is using. But as agreed we won't announce until 4.12.1. I assume you mean release or announce. Thanks, Justin
Re: ApacheCon Slides
On Mon, Mar 24, 2014 at 1:52 PM, Alex Harui aha...@adobe.com wrote: Thanks for the feedback. That diagram is a bit lower-level than I was going to go, but I'm interested to hear why that's an important piece for folks attending a 45 minute session. I do have another diagram from my 360|Stack slides that tries to show the class substitution done by the compiler that I might add. The strands/beads metaphor does a very good job of explaining composition. A few people I discussed this with seemed to have never heard this analogy and feel that it makes a lot of sense. Which is why I think you should probably use that concept and diagram when explaining FlexJS. If you have a better diagram, perhaps that should go on the wiki as well ;-) Thanks, Om -Alex On 3/24/14 1:19 PM, OmPrakash Muppirala bigosma...@gmail.com wrote: For FlexJS, it would be nice if you can use the beads/strands diagrams from here [1] They do a good job of explaining the concepts visually. Thanks, Om [1] https://cwiki.apache.org/confluence/display/FLEX/Creating+Components On Mon, Mar 24, 2014 at 1:12 PM, OmPrakash Muppirala bigosma...@gmail.comwrote: I would like to see mention of all the activities around Flex Mobile: - New components/features (Mobile Grid, Scrollable text) - iOS7 status bar support - Support for low res (120DPI) and hi-res (480, 640 DPI) devices - CSS-Media queries for OS version - CSS-Media queries for device dimensions - New Android 4.x skins (ongoing work) - New iOS7 skins (coming up next) Thanks, Om On Mon, Mar 24, 2014 at 1:03 PM, Alex Harui aha...@adobe.com wrote: ApacheCon required that we upload our slide decks by today. I just posted my slides. If you have time, please take a look and let me know if you think I should make changes. The FlexJS session is essentially the same as the 360|Stack presentation from last year. What's Happening in the Apache Flex Project http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_ Fle x.pdf http://events.linuxfoundation.org/sites/events/files/slides/Apache Con14_Flex.pdf FlexJS: Flex For javaScript: http://events.linuxfoundation.org/sites/events/files/slides/ApacheCon14_ Fle xJS.pdf http://events.linuxfoundation.org/sites/events/files/slides/Apac heCon14_FlexJS.pdf Thanks, -Alex
Re: [DISCUSSION] Release Apache Flex FlexUnit 4.2.0 RC2
Hi, Are we talking about a single file here? https://github.com/apache/flex-flexunit/blob/master/FlexUnit4Test/src/UIRunner.mxml If so I think that can be rewritten easily enough. Justin
Re: ApacheCon Slides
On 3/24/14 3:21 PM, OmPrakash Muppirala bigosma...@gmail.com wrote: On Mon, Mar 24, 2014 at 1:52 PM, Alex Harui aha...@adobe.com wrote: Thanks for the feedback. That diagram is a bit lower-level than I was going to go, but I'm interested to hear why that's an important piece for folks attending a 45 minute session. I do have another diagram from my 360|Stack slides that tries to show the class substitution done by the compiler that I might add. The strands/beads metaphor does a very good job of explaining composition. A few people I discussed this with seemed to have never heard this analogy and feel that it makes a lot of sense. Which is why I think you should probably use that concept and diagram when explaining FlexJS. If you have a better diagram, perhaps that should go on the wiki as well ;-) My diagram is about the compilation process, not composition. I'll see if it composition image fits on a slide. -Alex
Re: ApacheCon Slides
On Mon, Mar 24, 2014 at 2:10 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: No, it isn't about being ashamed. Given the kinds of people I met at the one ApacheCon I went to, I biased this presentation towards folks who don't know Flex, AS, MXML, and Flash. Ok, so it's on purpose. Hmm. I think the whole point of FlexJS is to create a framework that runs without Flash and a key factor is the impact from decisions and statements made by Jobs and Adobe. Are you saying that enough time has passed that these decisions and statements are no longer a factor? I'm not hearing that. Folks still seem to want to make sure their future isn't tied to Flash. But I probably don't talk to as many customers as you might. It's a difficult matter, not easy to discuss by email, even more so writing not in my mother tongue, so sorry in advance if I get misunderstood. You can say that FlexJS will work also on JavaScript, and I think that alone is enough to attract people, you don't need to shoot down on Flash. FlexJS is the power of MXML and AS brought to the javascript world And this has nothing to do with Flash, actually, but rather on the current weaknesses of javascript. Once again, for many non developers around me, they don't make the difference between Flex and Flash, and statements like this one, especially coming from Apache Flex PMC himself, could be harmful if they fall in non-educated ears. Will it decrease the value of FlexJS, if you don't say that Flash is not desirable anymore? What do others think? I agree that there is no need to ding Flash when talking about FlexJS. We are pushing releases of the Flex SDK that work on top of Flash/AIR on one side. We don't want to be saying on the other side that it is 'not a desirable' platform. Flash/AIR has a lot of positives(technically speaking). The only negative it has is that it does not run on iOS and Android browsers. Perhaps we point to that fact that FlexJS solves this problem so that folks can continue to keep one single codebase for all major platforms in the world. Thanks, Om Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : lundi 24 mars 2014 21:42 À : dev@flex.apache.org Objet : Re: ApacheCon Slides On 3/24/14 1:30 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: Flex.pdf Page 6: Eclipse-based IDE = not only (IDEA, FlashDevelop, etc..) Overall, the impression it gave me is that it's not describing what's happening in Apache Flex Project, but rather an advertisement to attract new developers by telling them Hey, there is not only AS in Flex, there are also other technologies, Java, Maven, so don't be afraid to come help us. Should we be ashamed of Flex being Flex, ActionScript and Flash ? No, it isn't about being ashamed. Given the kinds of people I met at the one ApacheCon I went to, I biased this presentation towards folks who don't know Flex, AS, MXML, and Flash. FlexJS.pdf: I think there is a major problem with these slides, starting from page 1: Flex without Flash Flash based solutions no longer desirable. Etc.. For many people people Flex == Flash. So if you say flash is not desirable anymore, they will understand Flex is not desirable anymore. I don't think this is what we are want. Sorry for being harsh, but I think we have had enough with Steve Jobs and Abobe PR. Hmm. I think the whole point of FlexJS is to create a framework that runs without Flash and a key factor is the impact from decisions and statements made by Jobs and Adobe. Are you saying that enough time has passed that these decisions and statements are no longer a factor? I'm not hearing that. Folks still seem to want to make sure their future isn't tied to Flash. But I probably don't talk to as many customers as you might. -Alex
RE: [DISCUSSION] Release Apache Flex FlexUnit 4.2.0 RC2
Are we talking about a single file here? https://github.com/apache/flex-flexunit/blob/master/FlexUnit4Test/src/UIRunner.mxml No, we are talking about the code in the tag below: Adobe:TestRunnerBase / Which resides here: https://github.com/apache/flex-flexunit/tree/master/FlexUnit4UIListener It was originally written (at least primarily) by Alex Uhlmann for FlexUnit .9. Here is a reference to one of the classes in the original form in github: https://github.com/flexunit/flexunit/blob/master/FlexUnit4UIListener/src/org/flexunit/flexui/data/AbstractRowData.as For what it's worth, the UIListener is separate from the FlexUnit 4 core and related tech. It's just a UI for people running FlexUnit outside of Flash Builder/IntelliJ and outside of a CI environment. If needed, we could lose it to move things forward. Mike
RE: ApacheCon Slides
The only negative it has is that it does not run on iOS and Android browsers. That's true, but subtle. And once again, we need to be careful about the possible confusion. When you say Flash/AIR does not run on iOS and Android browsers, some (many?) people understand Flex does not run on iOS and Android. It happened around me many times. WDYT? Maurice -Message d'origine- De : omup...@gmail.com [mailto:omup...@gmail.com] De la part de OmPrakash Muppirala Envoyé : lundi 24 mars 2014 23:28 À : dev@flex.apache.org Objet : Re: ApacheCon Slides On Mon, Mar 24, 2014 at 2:10 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: No, it isn't about being ashamed. Given the kinds of people I met at the one ApacheCon I went to, I biased this presentation towards folks who don't know Flex, AS, MXML, and Flash. Ok, so it's on purpose. Hmm. I think the whole point of FlexJS is to create a framework that runs without Flash and a key factor is the impact from decisions and statements made by Jobs and Adobe. Are you saying that enough time has passed that these decisions and statements are no longer a factor? I'm not hearing that. Folks still seem to want to make sure their future isn't tied to Flash. But I probably don't talk to as many customers as you might. It's a difficult matter, not easy to discuss by email, even more so writing not in my mother tongue, so sorry in advance if I get misunderstood. You can say that FlexJS will work also on JavaScript, and I think that alone is enough to attract people, you don't need to shoot down on Flash. FlexJS is the power of MXML and AS brought to the javascript world And this has nothing to do with Flash, actually, but rather on the current weaknesses of javascript. Once again, for many non developers around me, they don't make the difference between Flex and Flash, and statements like this one, especially coming from Apache Flex PMC himself, could be harmful if they fall in non-educated ears. Will it decrease the value of FlexJS, if you don't say that Flash is not desirable anymore? What do others think? I agree that there is no need to ding Flash when talking about FlexJS. We are pushing releases of the Flex SDK that work on top of Flash/AIR on one side. We don't want to be saying on the other side that it is 'not a desirable' platform. Flash/AIR has a lot of positives(technically speaking). The only negative it has is that it does not run on iOS and Android browsers. Perhaps we point to that fact that FlexJS solves this problem so that folks can continue to keep one single codebase for all major platforms in the world. Thanks, Om Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : lundi 24 mars 2014 21:42 À : dev@flex.apache.org Objet : Re: ApacheCon Slides On 3/24/14 1:30 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: Flex.pdf Page 6: Eclipse-based IDE = not only (IDEA, FlashDevelop, etc..) Overall, the impression it gave me is that it's not describing what's happening in Apache Flex Project, but rather an advertisement to attract new developers by telling them Hey, there is not only AS in Flex, there are also other technologies, Java, Maven, so don't be afraid to come help us. Should we be ashamed of Flex being Flex, ActionScript and Flash ? No, it isn't about being ashamed. Given the kinds of people I met at the one ApacheCon I went to, I biased this presentation towards folks who don't know Flex, AS, MXML, and Flash. FlexJS.pdf: I think there is a major problem with these slides, starting from page 1: Flex without Flash Flash based solutions no longer desirable. Etc.. For many people people Flex == Flash. So if you say flash is not desirable anymore, they will understand Flex is not desirable anymore. I don't think this is what we are want. Sorry for being harsh, but I think we have had enough with Steve Jobs and Abobe PR. Hmm. I think the whole point of FlexJS is to create a framework that runs without Flash and a key factor is the impact from decisions and statements made by Jobs and Adobe. Are you saying that enough time has passed that these decisions and statements are no longer a factor? I'm not hearing that. Folks still seem to want to make sure their future isn't tied to Flash. But I probably don't talk to as many customers as you might. -Alex
Re: [DISCUSSION] Release Apache Flex FlexUnit 4.2.0 RC2
Hi, It was originally written (at least primarily) by Alex Uhlmann for FlexUnit .9. Here is a reference to one of the classes in the original form in github: https://github.com/flexunit/flexunit/blob/master/FlexUnit4UIListener/src/org/flexunit/flexui/data/AbstractRowData.as Which IMO mean it's covered by FlexUnit 0.9 BSD license [1] and there's no issue here? Anyone have a differing opinion? Thanks, Justin 1. http://sourceforge.net/adobe/flexunit/wiki/License/
Re: ApacheCon Slides
Hi, I just uploaded my stuf a few minutes ago: http://events.linuxfoundation.org/sites/events/files/slides/Building%20Enterprise%20Level%20Applications%20with%20Apache%20Flex%20and%20Java-v9-20140313_1547.pdf Lots of info + liking the positive tone you have going there :-) You may want to consider dropping the drop shadow on the text IMO makes it a little hard to read. Thanks, Justin
Re: [DISCUSSION] Release Apache Flex FlexUnit 4.2.0 RC2
I am, of course, not a lawyer, nor an expert on these matter. IMO, the key question is whether any Adobe employee wrote any code that got donated to Apache that is not available on Open@Adobe. A secondary question is whether any code that is on Open@Adobe that has an Adobe copyright can be in the Apache Flex repo. All of the FlexUnit code on Open@Adobe appears to be under BSD. Therefore, two scenarios are ok 1) The code on Open@Adobe is being used without modification 2) The code was modified by DP folks. If (1) the code should really get downloaded from Open@Adobe instead of being in our repo. If (2) the code can be in our repo and the NOTICE and LICENSE need to be updated. It is important to verify that the copyright license text is not MPL. Different rules apply there. And we need to double check on whether the header can be replaced or not. -Alex On 3/24/14 3:38 PM, Michael A. Labriola labri...@digitalprimates.net wrote: Are we talking about a single file here? https://github.com/apache/flex-flexunit/blob/master/FlexUnit4Test/src/UIR unner.mxml No, we are talking about the code in the tag below: Adobe:TestRunnerBase / Which resides here: https://github.com/apache/flex-flexunit/tree/master/FlexUnit4UIListener It was originally written (at least primarily) by Alex Uhlmann for FlexUnit .9. Here is a reference to one of the classes in the original form in github: https://github.com/flexunit/flexunit/blob/master/FlexUnit4UIListener/src/o rg/flexunit/flexui/data/AbstractRowData.as For what it's worth, the UIListener is separate from the FlexUnit 4 core and related tech. It's just a UI for people running FlexUnit outside of Flash Builder/IntelliJ and outside of a CI environment. If needed, we could lose it to move things forward. Mike
Re: ApacheCon Slides
Chris, a couple of points: 1. You seem to switch between the terms 'Flex SDK' and 'FDK' midpoint in the presentation. I think we should all be consistently using the term 'Flex SDK'. 2. The numbers on slide 32 is a outdated. We have more than 40,000 installs of the Flex SDK since we started keeping track. 3. Agree with Justin on the shadow of the text. Otherwise, great job! Thanks, Om On Mon, Mar 24, 2014 at 3:58 PM, Justin Mclean jus...@classsoftware.comwrote: Hi, I just uploaded my stuf a few minutes ago: http://events.linuxfoundation.org/sites/events/files/slides/Building%20Enterprise%20Level%20Applications%20with%20Apache%20Flex%20and%20Java-v9-20140313_1547.pdf Lots of info + liking the positive tone you have going there :-) You may want to consider dropping the drop shadow on the text IMO makes it a little hard to read. Thanks, Justin
Re: ApacheCon Slides
On Mon, Mar 24, 2014 at 3:47 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: The only negative it has is that it does not run on iOS and Android browsers. That's true, but subtle. And once again, we need to be careful about the possible confusion. When you say Flash/AIR does not run on iOS and Android browsers, some (many?) people understand Flex does not run on iOS and Android. It happened around me many times. Happens to me all the time. Which is why we need to make more noise about it. The ApacheCon presentations could definitely be used to spread these messages. Thanks, Om WDYT? Maurice -Message d'origine- De : omup...@gmail.com [mailto:omup...@gmail.com] De la part de OmPrakash Muppirala Envoyé : lundi 24 mars 2014 23:28 À : dev@flex.apache.org Objet : Re: ApacheCon Slides On Mon, Mar 24, 2014 at 2:10 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: No, it isn't about being ashamed. Given the kinds of people I met at the one ApacheCon I went to, I biased this presentation towards folks who don't know Flex, AS, MXML, and Flash. Ok, so it's on purpose. Hmm. I think the whole point of FlexJS is to create a framework that runs without Flash and a key factor is the impact from decisions and statements made by Jobs and Adobe. Are you saying that enough time has passed that these decisions and statements are no longer a factor? I'm not hearing that. Folks still seem to want to make sure their future isn't tied to Flash. But I probably don't talk to as many customers as you might. It's a difficult matter, not easy to discuss by email, even more so writing not in my mother tongue, so sorry in advance if I get misunderstood. You can say that FlexJS will work also on JavaScript, and I think that alone is enough to attract people, you don't need to shoot down on Flash. FlexJS is the power of MXML and AS brought to the javascript world And this has nothing to do with Flash, actually, but rather on the current weaknesses of javascript. Once again, for many non developers around me, they don't make the difference between Flex and Flash, and statements like this one, especially coming from Apache Flex PMC himself, could be harmful if they fall in non-educated ears. Will it decrease the value of FlexJS, if you don't say that Flash is not desirable anymore? What do others think? I agree that there is no need to ding Flash when talking about FlexJS. We are pushing releases of the Flex SDK that work on top of Flash/AIR on one side. We don't want to be saying on the other side that it is 'not a desirable' platform. Flash/AIR has a lot of positives(technically speaking). The only negative it has is that it does not run on iOS and Android browsers. Perhaps we point to that fact that FlexJS solves this problem so that folks can continue to keep one single codebase for all major platforms in the world. Thanks, Om Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : lundi 24 mars 2014 21:42 À : dev@flex.apache.org Objet : Re: ApacheCon Slides On 3/24/14 1:30 PM, Maurice Amsellem maurice.amsel...@systar.com wrote: Flex.pdf Page 6: Eclipse-based IDE = not only (IDEA, FlashDevelop, etc..) Overall, the impression it gave me is that it's not describing what's happening in Apache Flex Project, but rather an advertisement to attract new developers by telling them Hey, there is not only AS in Flex, there are also other technologies, Java, Maven, so don't be afraid to come help us. Should we be ashamed of Flex being Flex, ActionScript and Flash ? No, it isn't about being ashamed. Given the kinds of people I met at the one ApacheCon I went to, I biased this presentation towards folks who don't know Flex, AS, MXML, and Flash. FlexJS.pdf: I think there is a major problem with these slides, starting from page 1: Flex without Flash Flash based solutions no longer desirable. Etc.. For many people people Flex == Flash. So if you say flash is not desirable anymore, they will understand Flex is not desirable anymore. I don't think this is what we are want. Sorry for being harsh, but I think we have had enough with Steve Jobs and Abobe PR. Hmm. I think the whole point of FlexJS is to create a framework that runs without Flash and a key factor is the impact from decisions and statements made by Jobs and Adobe. Are you saying that enough time has passed that these decisions and statements are no longer a factor? I'm not hearing that. Folks still seem to want to make sure their future isn't tied to Flash. But I probably don't talk to as many customers as you might. -Alex
Re: [DISCUSSION] Release Apache Flex FlexUnit 4.2.0 RC2
Hi, All of the FlexUnit code on Open@Adobe appears to be under BSD. Therefore, two scenarios are ok 1) The code on Open@Adobe is being used without modification 2) The code was modified by DP folks. As Mike stated the code was modified from FU 0.9 by Digital Primates. Mike can you reconfirm this is the case? I have also since modified this code to have the correct trademark statement and a link to flex.a.o. It is important to verify that the copyright license text is not MPL. Different rules apply there. It clearly states BSD - why do you think it may be MPL? And we need to double check on whether the header can be replaced or not. Not sure re that. The BSD license say nothing on that mater just that the copyright notice be included we are already doing that. How do we resolve this? Thanks, Justin
RE: [DISCUSSION] Release Apache Flex FlexUnit 4.2.0 RC2
2) The code was modified by DP folks. It was modified significantly as the original code was intended to work with FlexUnit .9 alone and we modified it to work with both projects. It is important to verify that the copyright license text is not MPL. It's a BSD-2 header. Mike
Re: ApacheCon Slides
Hi, Also I think both slide decks need a little more branding (Apache Flex and ASF logos) and of course the trademark statement and a link to flex.a.o wouldn't go astray. Remember more than just attendees of ApacheCon will see these. Thanks, Justin
Re: ApacheCon Slides
Hi, 2. The numbers on slide 32 is a outdated. We have more than 40,000 installs of the Flex SDK since we started keeping track. BTW where do you get the 40,000 number from. I can only see 30,000 in the google stats. That only include 4.9 and up are you including 4.8 in that and if so how did we measure that it got 10,000 downloads? Thanks, Justin
Google stats flex.apache.org
Here is the breakdown till date: /track-installer.html 11,033 /track-installer.html?version=4.11.0os=windows 10,736 /track-installer.html?version=4.10.0os=windows 8,270 /track-installer.html?version=4.11.0os=mac 3,777 /track-installer.html?version=4.10.0os=mac 2,755 /track-installer.html?version=4.12.0os=windows 2,224 /track-installer.html?version=4.9.1os=windows 887 /track-installer.html?version=4.12.0os=mac 794 /track-installer.html?version=4.9.1os=mac 239 /track-installer.html?version=4.11.0os=linux 163 Total: 40,878 Note that we added version/os tracking from 4.9.1 onwards. So, the first number of track-installer.html corresponds to 4.9. What are you seeing? Thanks, Om On Mon, Mar 24, 2014 at 4:27 PM, Justin Mclean jus...@classsoftware.comwrote: Hi, 2. The numbers on slide 32 is a outdated. We have more than 40,000 installs of the Flex SDK since we started keeping track. BTW where do you get the 40,000 number from. I can only see 30,000 in the google stats. That only include 4.9 and up are you including 4.8 in that and if so how did we measure that it got 10,000 downloads? Thanks, Justin
Re: Google stats flex.apache.org
Of course, if we add the pre 4.9 numbers, it will be closer to 50,000. Nick had posted some numbers a while ago. I will see if I can dig that up. Thanks, Om On Mon, Mar 24, 2014 at 4:33 PM, OmPrakash Muppirala bigosma...@gmail.comwrote: Here is the breakdown till date: /track-installer.html 11,033 /track-installer.html?version=4.11.0os=windows 10,736 /track-installer.html?version=4.10.0os=windows 8,270 /track-installer.html?version=4.11.0os=mac 3,777 /track-installer.html?version=4.10.0os=mac 2,755 /track-installer.html?version=4.12.0os=windows 2,224 /track-installer.html?version=4.9.1os=windows 887 /track-installer.html?version=4.12.0os=mac 794 /track-installer.html?version=4.9.1os=mac 239 /track-installer.html?version=4.11.0os=linux 163 Total: 40,878 Note that we added version/os tracking from 4.9.1 onwards. So, the first number of track-installer.html corresponds to 4.9. What are you seeing? Thanks, Om On Mon, Mar 24, 2014 at 4:27 PM, Justin Mclean jus...@classsoftware.comwrote: Hi, 2. The numbers on slide 32 is a outdated. We have more than 40,000 installs of the Flex SDK since we started keeping track. BTW where do you get the 40,000 number from. I can only see 30,000 in the google stats. That only include 4.9 and up are you including 4.8 in that and if so how did we measure that it got 10,000 downloads? Thanks, Justin
Re: [DISCUSSION] Release Apache Flex FlexUnit 4.2.0 RC2
IMO, the Adobe code is a third-party work as defined here: http://www.apache.org/legal/src-headers.html#3party The Adobe code would not be a third-party work if there was an agreement with DP giving DP rights. But once there are significant modifications, it appears we may need to consult legal-discuss for their thoughts. In the meantime, those files should probably retain the Adobe copyright and BSD header. And I will ask Adobe legal for their thoughts. Thanks, -Alex On 3/24/14 4:15 PM, Michael A. Labriola labri...@digitalprimates.net wrote: 2) The code was modified by DP folks. It was modified significantly as the original code was intended to work with FlexUnit .9 alone and we modified it to work with both projects. It is important to verify that the copyright license text is not MPL. It's a BSD-2 header. Mike
Re: [DISCUSSION] Release Apache Flex FlexUnit 4.2.0 RC2
Hi, IMO, the Adobe code is a third-party work as defined here: http://www.apache.org/legal/src-headers.html#3party Note that in this case I believe it would fall under: 5. Major modifications/additions to third-party should be dealt with on a case-by-case basis by the PMC. Not this: 4. Minor modifications/additions to third-party source files should typically be licensed under the same terms as the rest of the rest of the third-party source for convenience. So it's really up to us to decide but if you want to drag legal into it go ahead. In the meantime, those files should probably retain the Adobe copyright and BSD header. See 4 vs 5 above. IMO the headers don't need to change. Thanks, Justin
Re: [DISCUSSION] Release Apache Flex FlexUnit 4.2.0 RC2
On 3/24/14 5:07 PM, Justin Mclean jus...@classsoftware.com wrote: Hi, IMO, the Adobe code is a third-party work as defined here: http://www.apache.org/legal/src-headers.html#3party Note that in this case I believe it would fall under: 5. Major modifications/additions to third-party should be dealt with on a case-by-case basis by the PMC. It wasn't clear that the PMC had total authority. I figured it meant we should get some advice from legal-discuss. Not this: 4. Minor modifications/additions to third-party source files should typically be licensed under the same terms as the rest of the rest of the third-party source for convenience. So it's really up to us to decide but if you want to drag legal into it go ahead. Well, the thing I don't understand is, that this is a collection of files some of which were changed significantly, but some not. So then which rules apply? That's what I want to find out. In the meantime, those files should probably retain the Adobe copyright and BSD header. See 4 vs 5 above. IMO the headers don't need to change. Thanks, Justin