Installer and locales
HI, I've just refactoring the locale/language code in the installer and think I found a bug, it's been reported but not 100% sure if it was fixed. Basically there are there are two possibilities: 1. User locale is set to one of the supported locales so that should be default locale. 2. User locale is not set to one of the supported locales so it should be set to the default locale (currently en_US). Look like that n the second case it wasn't being set back to en_US as it was setting to en_US first then setting to user locales. Anyone not using one of these locales like to give the current installer a try to confirm. de_DE, en_US, eu_AU, en_GB, es_ES, fr_FR, el_GR, nl_NL, pt_BR And as always the installer could do with more locales if anyone is willing to do some translation. Thanks, Justin
Re: Air Stage Text Issue
This may work for you: http://blogs.adobe.com/cantrell/archives/2011/09/native-text-input-with-stagetext.html On 30/09/2013 03:28, Flexicious.com wrote: This may not be a Flex/Apache issue, but just wanted to see if anyone has been successful with text on Air mobile. The issue basically revolves around the native text input. It is rendered on top of the display list, and it does not scroll within its parent scroller. So dropdowns and such that render on top of it do not hide it, and if the user scrolls the text stay in place. There are some more details here: http://stackoverflow.com/questions/15840504/what-options-are-available-for-handling-text-input-on-android-using-adobe-air https://bugbase.adobe.com/index.cfm?event=bug&id=3302441 Some have suggested using the older spark.mobile.TextInputSkin but with Air 3.8 and iOS7 on iPad , the workaround suggested is not an option. It seems to be even more problematic, as you type it selects what you have typed so you cannot effectively type anything without erasing what was there. A comment from an Adobe person here suggests the issue with the older TextinpustSkin is a Flex SDK issue. https://bugbase.adobe.com/index.cfm?event=bug&id=3292370 Has anyone run into this, if so, been able to handle it gracefully? -- Lee Burrows ActionScripter
Spark DateFieldSpinner
Hi, I wonder if anyone has already considered to combine the DateField with the Mobile DateSpinner? I have created a small example app with view source enabled [1] ... [1] https://dl.dropboxusercontent.com/u/352808/apacheflex/datefieldspinner/DateFieldSpinnerTest.html To make this work in Flash Builder 4.6 you need to the Mobile-Theme under Properties > Flex Theme. -- Sebastian (PPMC) Interaction Designer Looking for a Login Example with Apache Flex? Please check out this code: http://code.google.com/p/masuland/wiki/LoginExample
RE: Falcon progress
I pulled the latest code, including your change, and ran SDKSWCTests again (3 times) inside Eclipse by right clicking on SDKSWCTests.java in Package Explorer and choosing Debug As > JUnit Test. Only one test is still failing for me, namely textLayoutSWC(). The JUnit view says java.lang.AssertionError: Expected: is <0> got: <1> at f.SDKSWCTests.compileSWC(SDKSWCTests.java:110) at f.SDKSWCTests.textLayoutSWC(SDKSWCTests.java:357) When I run just that one test, the Console view says environment property - FLEX_HOME = D:\asf\flex-sdk environment property - PLAYERGLOBAL_HOME = D:\asf\prerequisites\player environment property - TLF_HOME = null environment property - AIR_HOME = D:\asf\prerequisites\AdobeAIRSDK environment property - FLASHPLAYER_DEBUGGER = D:\asf\prerequisites\fp_11.1.102.63_archive\11_1r102_63_32bit_debug\flashplayer11_1r102_63_win_sa_debug_32bit.exe Loading configuration: D:\asf\flex-sdk\frameworks\flex-config.xml command line Error: unable to open 'D:\asf\flex-falcon\compiler.tests\null\compile-config.xml'. - Gordon -Original Message- From: Erik de Bruin [mailto:e...@ixsoftware.nl] Sent: Saturday, September 28, 2013 5:25 AM To: dev@flex.apache.org Subject: Re: Falcon progress Interesting... On my end only 'sparkSWC' fails. The other tests all pass as configured (i.e. with -ignore-problems set to cause the tests not to fail on warnings). I've added @Ignore on 'sparkSWC' as I agree all committed tests should either pass or be ignored. What failures are you seeing on the 5 other tests that fail on your end? EdB On Tue, Sep 24, 2013 at 11:37 PM, Gordon Smith wrote: > Congratulations on getting more SWCs to compile with Falcon! It looks like > only the following 6 tests in SDKSWCTests still have problems: > > authoringsupportSWC() > automation_flashflexkitSWC() > chartsSWC() > mxSWC() > sparkSWC() > textLayoutSWC() > > Is there a reason why the @Ignore was removed from these failing tests? I > recommend keepng the Falcon test suite passing at all times, so that > developers know when they've introduced a new failure. > > Most of the changes seem to have been minor changes to SDK code to placate > Falcon since it is stricter than the old compiler, or changes to the > compilation options for the test to suppress errors or warnings that we can > fix later. However, I'm concerned about changes to Falcon itself like this > that are workarounds rather than fixes: > > FLEX-33713: An NPE is thrown by this code when compiling 'spark.swc' from > the SDK. The null check bypasses the issue, it doesn't do anything to fix the > root cause... > > I think this kind of thing is OK as long as these changes are trackable in > some way that makes it possible to go back and make the actual fix later. Do > we have to just rely on 'git log' for to do this? If so, we should put > something we can search for like "WORKAROUND" in the Git comment. > > Has anybody gotten a complete app to compile with Falcon yet? Does it run > correctly? > > - Gordon > -- Ix Multimedia Software Jan Luykenstraat 27 3521 VB Utrecht T. 06-51952295 I. www.ixsoftware.nl
[FALCON] using coerce_a
Gordon, Darrell (mostly) My latest problem appears to be in assigning to a Dictionary. For the following code: var dict:Dictionary = new Dictionary(); Var i:int; Dict["foo"] = (i == 0 ? new SomeClass() : new SomeOtherClass()); MXMLC generates a coerce_a after the constructProp for both SomeClass and SomeOtherClass. For some reason, FlashPlayer doesn't care if the coerce_a is missing, but AIR seems to. The question is, should a complex assignment statement like the above know about the destination type as the clauses of the ternary statement are being reduced? And if so, how would we get that knowledge into the reducer. Alternatively, is it safe to add a coerce_a before an assignment to a Dictionary? If so, what would be the recommended way to determine the assignment is to a Dictionary? Thanks, -Alex
RE: [FALCON] using coerce_a
Can you remind me what the "a" in coerce_a means? - Gordon -Original Message- From: Alex Harui [mailto:aha...@adobe.com] Sent: Monday, September 30, 2013 2:43 PM To: dev@flex.apache.org Subject: [FALCON] using coerce_a Gordon, Darrell (mostly) My latest problem appears to be in assigning to a Dictionary. For the following code: var dict:Dictionary = new Dictionary(); Var i:int; Dict["foo"] = (i == 0 ? new SomeClass() : new SomeOtherClass()); MXMLC generates a coerce_a after the constructProp for both SomeClass and SomeOtherClass. For some reason, FlashPlayer doesn't care if the coerce_a is missing, but AIR seems to. The question is, should a complex assignment statement like the above know about the destination type as the clauses of the ternary statement are being reduced? And if so, how would we get that knowledge into the reducer. Alternatively, is it safe to add a coerce_a before an assignment to a Dictionary? If so, what would be the recommended way to determine the assignment is to a Dictionary? Thanks, -Alex
Installer stats - now at 7,500
Hi, It's been just under 2 months since the 4.10 release and we have just passed the 7,500 successful installs mark. We're also approaching the 20,000 mark since we started measuring installs. (This only counts installs that complete via the installer not via downloads off our site or other mirrors.) Thanks, Justin
Re: [FALCON] using coerce_a
coerce_a Operation Coerce a value to the any type. Format coerce_a Forms Stack ..., value => ..., value Description Indicates to the verifier that the value on the stack is of the any type (*). Does nothing to value. On Mon, Sep 30, 2013 at 6:03 PM, Gordon Smith wrote: > Can you remind me what the "a" in coerce_a means? > > - Gordon > > -Original Message- > From: Alex Harui [mailto:aha...@adobe.com] > Sent: Monday, September 30, 2013 2:43 PM > To: dev@flex.apache.org > Subject: [FALCON] using coerce_a > > Gordon, Darrell (mostly) > > My latest problem appears to be in assigning to a Dictionary. For the > following code: > > var dict:Dictionary = new Dictionary(); > Var i:int; > Dict["foo"] = (i == 0 ? new SomeClass() : new SomeOtherClass()); > > MXMLC generates a coerce_a after the constructProp for both SomeClass and > SomeOtherClass. > For some reason, FlashPlayer doesn't care if the coerce_a is missing, but > AIR seems to. > > The question is, should a complex assignment statement like the above know > about the destination type as the clauses of the ternary statement are > being reduced? And if so, how would we get that knowledge into the reducer. > > Alternatively, is it safe to add a coerce_a before an assignment to a > Dictionary? If so, what would be the recommended way to determine the > assignment is to a Dictionary? > > Thanks, > -Alex > >
[DISCUSS] Apache Flex Bylaws
Hi, Looks like as part of the graduation process that we didn't get together and come up with a list of bylaws for Apache Flex. Some example Apache project bylaws: http://hadoop.apache.org/bylaws.html http://cloudstack.apache.org/bylaws.html https://cwiki.apache.org/confluence/display/Hive/Bylaws https://cwiki.apache.org/confluence/display/KAFKA/Bylaws http://ant.apache.org/bylaws.html As well as listing out the roles and responsibilities on the project it needs to spell out the voting rules. Here's some of the things we would need to reach consensus on. If PMC members (and others) could indicate their preferences to there questions below we'll start getting a document together based on this what other Apache projects have done. If you have no strong preference indicate that and this is not a vote, just more of what you think bests fits in our community. 1. Voting procedure for committers a) lazy-majority (more +1s than -1s) b) lazy-consensus (no -1's) c) consensus (3 +1's no -1's) 2. Voting procedure for PMC members a) lazy-majority (more +1s than -1s) b) lazy-consensus (no -1's) c) consensus (3 +1's no -1's) d) 2/3 majority (3 +1's and twice an many +1 as -1) 3. Voting procedure for Chair a) lazy-majority (more +1s than -1s) b) lazy-consensus (no -1's) c) consensus (3 +1's no -1's) d) 2/3 majority (3 +1's and twice an many +1 as -1) 4. Length of term of Chair a) until they resign b) 1 year d) reviewed every year with option to continue c) other 5. Voting procedure on changing bylaws a) lazy-majority (more +1s than -1s) b) lazy-consensus (no -1's) c) consensus (3 +1's no -1's) d) 2/3 majority (3 +1's and twice an many +1 as -1) I notice other Bylaws use the word active a lot eg "active PMC", "active committers", but don't seem to define it. Should we try and define or at least give some guidelines what an active person on the project is? There seems to be a vague "not contributing to the project for 6 months" in one or two of the existing bylaws. There are several other thing in the bylaws eg removal of PMC or committer but as they are hopefully unlikely to happen so I can't see consensus being hard to reach there. I assume once we have a complete document put together we would need to take a vote and inform the board. Thanks, Justin
Re: Spark DateFieldSpinner
Does anyone know how to use the Spark DateFieldSpinner without the need to select the Flex Mobile Theme in FlashBuilder? Thanks :) -- Sebastian (PPMC) Interaction Designer Looking for a Login Example with Apache Flex? Please check out this code: http://code.google.com/p/masuland/wiki/LoginExample On Mon, Sep 30, 2013 at 9:43 PM, Sebastian Mohr wrote: > Hi, > > I wonder if anyone has already considered to > combine the DateField with the Mobile DateSpinner? > I have created a small example app with view source > enabled [1] ... > > [1] > https://dl.dropboxusercontent.com/u/352808/apacheflex/datefieldspinner/DateFieldSpinnerTest.html > > To make this work in Flash Builder 4.6 you need to > the Mobile-Theme under Properties > Flex Theme. > > > -- > Sebastian (PPMC) > Interaction Designer > > Looking for a Login Example with Apache Flex? Please check out this code: > http://code.google.com/p/masuland/wiki/LoginExample >
RE: [FALCON] using coerce_a
Thanks Darrell. Here are my answers to Alex's questions: > Should a complex assignment statement like the above know about the > destination type as the clauses of the ternary statement are being reduced? No. > Is it safe to add a coerce_a before an assignment to a Dictionary? Yes. > If so, what would be the recommended way to determine the assignment is to a > Dictionary? The compile-time type of foo[bar], where foo has any type (not just Dictionary), is type *. So I think it is OK to codegen coerce_a before assignment to any foo[bar] expression. Darrell, do you agree or disagree? - Gordon -Original Message- From: Darrell Loverin [mailto:darrell.love...@gmail.com] Sent: Monday, September 30, 2013 3:41 PM To: dev@flex.apache.org Subject: Re: [FALCON] using coerce_a coerce_a Operation Coerce a value to the any type. Format coerce_a Forms Stack ..., value => ..., value Description Indicates to the verifier that the value on the stack is of the any type (*). Does nothing to value. On Mon, Sep 30, 2013 at 6:03 PM, Gordon Smith wrote: > Can you remind me what the "a" in coerce_a means? > > - Gordon > > -Original Message- > From: Alex Harui [mailto:aha...@adobe.com] > Sent: Monday, September 30, 2013 2:43 PM > To: dev@flex.apache.org > Subject: [FALCON] using coerce_a > > Gordon, Darrell (mostly) > > My latest problem appears to be in assigning to a Dictionary. For the > following code: > > var dict:Dictionary = new Dictionary(); Var i:int; Dict["foo"] = (i == > 0 ? new SomeClass() : new SomeOtherClass()); > > MXMLC generates a coerce_a after the constructProp for both SomeClass > and SomeOtherClass. > For some reason, FlashPlayer doesn't care if the coerce_a is missing, > but AIR seems to. > > The question is, should a complex assignment statement like the above > know about the destination type as the clauses of the ternary > statement are being reduced? And if so, how would we get that knowledge into > the reducer. > > Alternatively, is it safe to add a coerce_a before an assignment to a > Dictionary? If so, what would be the recommended way to determine the > assignment is to a Dictionary? > > Thanks, > -Alex > >
Re: [DISCUSS] Apache Flex Bylaws
On 9/30/2013 7:19 PM, Justin Mclean wrote: Hi, Looks like as part of the graduation process that we didn't get together and come up with a list of bylaws for Apache Flex. Some example Apache project bylaws: http://hadoop.apache.org/bylaws.html http://cloudstack.apache.org/bylaws.html https://cwiki.apache.org/confluence/display/Hive/Bylaws https://cwiki.apache.org/confluence/display/KAFKA/Bylaws http://ant.apache.org/bylaws.html As well as listing out the roles and responsibilities on the project it needs to spell out the voting rules. Here's some of the things we would need to reach consensus on. If PMC members (and others) could indicate their preferences to there questions below we'll start getting a document together based on this what other Apache projects have done. If you have no strong preference indicate that and this is not a vote, just more of what you think bests fits in our community. 1. Voting procedure for committers a) lazy-majority (more +1s than -1s) b) lazy-consensus (no -1's) c) consensus (3 +1's no -1's) b 2. Voting procedure for PMC members a) lazy-majority (more +1s than -1s) b) lazy-consensus (no -1's) c) consensus (3 +1's no -1's) d) 2/3 majority (3 +1's and twice an many +1 as -1) c 3. Voting procedure for Chair a) lazy-majority (more +1s than -1s) b) lazy-consensus (no -1's) c) consensus (3 +1's no -1's) d) 2/3 majority (3 +1's and twice an many +1 as -1) c 4. Length of term of Chair a) until they resign b) 1 year d) reviewed every year with option to continue c) other d 5. Voting procedure on changing bylaws a) lazy-majority (more +1s than -1s) b) lazy-consensus (no -1's) c) consensus (3 +1's no -1's) d) 2/3 majority (3 +1's and twice an many +1 as -1) d I notice other Bylaws use the word active a lot eg "active PMC", "active committers", but don't seem to define it. Should we try and define or at least give some guidelines what an active person on the project is? There seems to be a vague "not contributing to the project for 6 months" in one or two of the existing bylaws. I guess we'd have to define formally what a contribution is. Is it activity on a mailing list? Actual commits? Jira activity? Mailing list moderation? It sounds like it is a hard thing to concretely define. -- Jeffry Houser Technical Entrepreneur http://www.jeffryhouser.com 203-379-0773
Re: [FALCON] using coerce_a
I generally agree. I don't think I'd try to do anything inside the ternary expression reduce. I'd look at the coerce() method in ABCGeneratingReducer to try to figure out why it is not being called. Question for Alex. What error are you seeing on AIR? A TypeError, a ReferenceError? -Darrell On Mon, Sep 30, 2013 at 8:00 PM, Gordon Smith wrote: > Thanks Darrell. > > Here are my answers to Alex's questions: > > > Should a complex assignment statement like the above know about the > destination type as the clauses of the ternary statement are being reduced? > > No. > > > Is it safe to add a coerce_a before an assignment to a Dictionary? > > Yes. > > > If so, what would be the recommended way to determine the assignment is > to a Dictionary? > > The compile-time type of foo[bar], where foo has any type (not just > Dictionary), is type *. So I think it is OK to codegen coerce_a before > assignment to any foo[bar] expression. > > Darrell, do you agree or disagree? > > - Gordon > > > -Original Message- > From: Darrell Loverin [mailto:darrell.love...@gmail.com] > Sent: Monday, September 30, 2013 3:41 PM > To: dev@flex.apache.org > Subject: Re: [FALCON] using coerce_a > > coerce_a Operation > > Coerce a value to the any type. > > Format > > coerce_a > > Forms > > Stack > > ..., value => ..., value Description > > Indicates to the verifier that the value on the stack is of the any type > (*). Does nothing to value. > > > On Mon, Sep 30, 2013 at 6:03 PM, Gordon Smith wrote: > > > Can you remind me what the "a" in coerce_a means? > > > > - Gordon > > > > -Original Message- > > From: Alex Harui [mailto:aha...@adobe.com] > > Sent: Monday, September 30, 2013 2:43 PM > > To: dev@flex.apache.org > > Subject: [FALCON] using coerce_a > > > > Gordon, Darrell (mostly) > > > > My latest problem appears to be in assigning to a Dictionary. For the > > following code: > > > > var dict:Dictionary = new Dictionary(); Var i:int; Dict["foo"] = (i == > > 0 ? new SomeClass() : new SomeOtherClass()); > > > > MXMLC generates a coerce_a after the constructProp for both SomeClass > > and SomeOtherClass. > > For some reason, FlashPlayer doesn't care if the coerce_a is missing, > > but AIR seems to. > > > > The question is, should a complex assignment statement like the above > > know about the destination type as the clauses of the ternary > > statement are being reduced? And if so, how would we get that knowledge > into the reducer. > > > > Alternatively, is it safe to add a coerce_a before an assignment to a > > Dictionary? If so, what would be the recommended way to determine the > > assignment is to a Dictionary? > > > > Thanks, > > -Alex > > > > >
Re: Migration to Apachle Flex 4.10
Can Flex 4.10 work with Flash Builder 4.5.1 ? TIA, Oleg. On Sat, Sep 28, 2013 at 7:56 PM, piotr.zarzycki wrote: > I'm sorry Amit for wrong link. This is correct -> http://bit.ly/1azHNgM > > > > - > Flex/Air Developer > -- > View this message in context: > http://apache-flex-development.247.n4.nabble.com/Migration-to-Apachle-Flex-4-10-tp30169p30180.html > Sent from the Apache Flex Development mailing list archive at Nabble.com. > -- Thank you, Oleg.
Re: [DISCUSS] Apache Flex Bylaws
Hi, Thanks for your response. > I guess we'd have to define formally what a contribution is. Is it activity > on a mailing list? Actual commits? Jira activity? Mailing list moderation? > It sounds like it is a hard thing to concretely define. Yep. I'm included to go with with "any activity in the past 6 months" that includes any JIRA, mailing list, Git etc no matter how minor and they can "rejoin" at any time. I see no reason to punish people who want to contribute who haven't for a while and otherwise it would become too subjective. Thanks, Justin
Re: [DISCUSS] Apache Flex Bylaws
Hi, I've put up some (very) draft bylaws, mostly a modified form of other Apache projects. Please feel free to edit or suggest changes or tell me if you disagree with anything. https://cwiki.apache.org/confluence/display/FLEX/Draft+Bylaws Thanks, Justin
Re: [DISCUSS] Apache Flex Bylaws
Hi, Guess I answer my own questions > 1. Voting procedure for committers a or b, but more towards a. > 2. Voting procedure for PMC members b -IMO needs a higher bar than committers. > 3. Voting procedure for Chair c or d. > 4. Length of term of Chair b or d. Should be shared about, but does anyone actually want to take it on? > 5. Voting procedure on changing bylaws c or d. Thanks, Justin
Re: Migration to Apachle Flex 4.10
Thank you all , Let me try to migrate the application and will post any thing if come with issues. Amit On Tue, Oct 1, 2013 at 6:52 AM, Oleg Konovalov wrote: > Can Flex 4.10 work with Flash Builder 4.5.1 ? > > TIA, > Oleg. > > > On Sat, Sep 28, 2013 at 7:56 PM, piotr.zarzycki > wrote: > > > I'm sorry Amit for wrong link. This is correct -> http://bit.ly/1azHNgM > > > > > > > > - > > Flex/Air Developer > > -- > > View this message in context: > > > http://apache-flex-development.247.n4.nabble.com/Migration-to-Apachle-Flex-4-10-tp30169p30180.html > > Sent from the Apache Flex Development mailing list archive at Nabble.com. > > > > > > -- > Thank you, > Oleg. >
Re: [DISCUSS] Apache Flex Bylaws
Hi, And it been pointed out to me that coming up with our own bylaws was part of the board resolution when we first become a TLP. [1] Oppps, oh well, better late than never :-) Thanks, Justin 1. https://cwiki.apache.org/confluence/display/FLEX/Graduation+Resolution
Re: [DISCUSS] Apache Flex Bylaws
On Sep 30, 2013 4:19 PM, "Justin Mclean" wrote: > > Hi, > > Looks like as part of the graduation process that we didn't get together and come up with a list of bylaws for Apache Flex. > > Some example Apache project bylaws: > http://hadoop.apache.org/bylaws.html > http://cloudstack.apache.org/bylaws.html > https://cwiki.apache.org/confluence/display/Hive/Bylaws > https://cwiki.apache.org/confluence/display/KAFKA/Bylaws > http://ant.apache.org/bylaws.html > > As well as listing out the roles and responsibilities on the project it needs to spell out the voting rules. > > Here's some of the things we would need to reach consensus on. If PMC members (and others) could indicate > their preferences to there questions below we'll start getting a document together based on this what other > Apache projects have done. > Here are my preferences: > If you have no strong preference indicate that and this is not a vote, just more of what you think bests fits in our > community. > > 1. Voting procedure for committers > a) lazy-majority (more +1s than -1s) > b) lazy-consensus (no -1's) > c) consensus (3 +1's no -1's) > c. > 2. Voting procedure for PMC members > a) lazy-majority (more +1s than -1s) > b) lazy-consensus (no -1's) > c) consensus (3 +1's no -1's) > d) 2/3 majority (3 +1's and twice an many +1 as -1) c. > > 3. Voting procedure for Chair > a) lazy-majority (more +1s than -1s) > b) lazy-consensus (no -1's) > c) consensus (3 +1's no -1's) > d) 2/3 majority (3 +1's and twice an many +1 as -1) > c. > 4. Length of term of Chair > a) until they resign > b) 1 year > d) reviewed every year with option to continue > c) other a or d > > 5. Voting procedure on changing bylaws > a) lazy-majority (more +1s than -1s) > b) lazy-consensus (no -1's) > c) consensus (3 +1's no -1's) > d) 2/3 majority (3 +1's and twice an many +1 as -1) c. > > I notice other Bylaws use the word active a lot eg "active PMC", "active committers", but don't seem to define it. > Should we try and define or at least give some guidelines what an active person on the project is? I think we should. Although I am not sure what the distinction means. Are you thinking of restricting voting privileges to non active PMC? > There seems to be a vague "not contributing to the project for 6 months" in one or two of the existing bylaws. > > There are several other thing in the bylaws eg removal of PMC or committer but as they are hopefully unlikely to > happen so I can't see consensus being hard to reach there. IMHO, there are way too many committers/PMC members who have not contributed anything since the project started. Maybe, we should check once in a while to see if they are still around. Thanks, Om > > I assume once we have a complete document put together we would need to take a vote and inform the board. > > Thanks, > Justin
Re: [DISCUSS] Apache Flex Bylaws
On Sep 30, 2013 5:59 PM, "Jeffry Houser" wrote: > > On 9/30/2013 7:19 PM, Justin Mclean wrote: >> >> Hi, >> >> Looks like as part of the graduation process that we didn't get together and come up with a list of bylaws for Apache Flex. >> >> Some example Apache project bylaws: >> http://hadoop.apache.org/bylaws.html >> http://cloudstack.apache.org/bylaws.html >> https://cwiki.apache.org/confluence/display/Hive/Bylaws >> https://cwiki.apache.org/confluence/display/KAFKA/Bylaws >> http://ant.apache.org/bylaws.html >> >> As well as listing out the roles and responsibilities on the project it needs to spell out the voting rules. >> >> Here's some of the things we would need to reach consensus on. If PMC members (and others) could indicate >> their preferences to there questions below we'll start getting a document together based on this what other >> Apache projects have done. >> >> If you have no strong preference indicate that and this is not a vote, just more of what you think bests fits in our >> community. >> >> 1. Voting procedure for committers >> a) lazy-majority (more +1s than -1s) >> b) lazy-consensus (no -1's) >> c) consensus (3 +1's no -1's) > > b > >> 2. Voting procedure for PMC members >> a) lazy-majority (more +1s than -1s) >> b) lazy-consensus (no -1's) >> c) consensus (3 +1's no -1's) >> d) 2/3 majority (3 +1's and twice an many +1 as -1) > > c >> >> 3. Voting procedure for Chair >> a) lazy-majority (more +1s than -1s) >> b) lazy-consensus (no -1's) >> c) consensus (3 +1's no -1's) >> d) 2/3 majority (3 +1's and twice an many +1 as -1) > > c >> >> 4. Length of term of Chair >> a) until they resign >> b) 1 year >> d) reviewed every year with option to continue >> c) other > > d >> >> 5. Voting procedure on changing bylaws >> a) lazy-majority (more +1s than -1s) >> b) lazy-consensus (no -1's) >> c) consensus (3 +1's no -1's) >> d) 2/3 majority (3 +1's and twice an many +1 as -1) > > d >> >> I notice other Bylaws use the word active a lot eg "active PMC", "active committers", but don't seem to define it. >> Should we try and define or at least give some guidelines what an active person on the project is? >> There seems to be a vague "not contributing to the project for 6 months" in one or two of the existing bylaws. > > > I guess we'd have to define formally what a contribution is. Is it activity on a mailing list? Actual commits? Jira activity? Mailing list moderation? > It sounds like it is a hard thing to concretely define. Do you want to take a stab at defining it? I think the bar should be pretty low. Something like at least one email on dev, users or private in 6 months. Or a commit in 6 months. > > -- > Jeffry Houser > Technical Entrepreneur > http://www.jeffryhouser.com > 203-379-0773 >
RE: [FALCON] using coerce_a
There's no error, the Dictionary value appears be a Number instead of an instance. Are you sure that Vector access doesn't come through this code? I think I'll start by testing only for ANY_TYPE and adding a coerce_a instead of calling coerce() which would also coerce to any destination type. Thanks for the advice, -Alex From: Darrell Loverin [darrell.love...@gmail.com] Sent: Monday, September 30, 2013 6:11 PM To: dev@flex.apache.org Subject: Re: [FALCON] using coerce_a I generally agree. I don't think I'd try to do anything inside the ternary expression reduce. I'd look at the coerce() method in ABCGeneratingReducer to try to figure out why it is not being called. Question for Alex. What error are you seeing on AIR? A TypeError, a ReferenceError? -Darrell On Mon, Sep 30, 2013 at 8:00 PM, Gordon Smith wrote: > Thanks Darrell. > > Here are my answers to Alex's questions: > > > Should a complex assignment statement like the above know about the > destination type as the clauses of the ternary statement are being reduced? > > No. > > > Is it safe to add a coerce_a before an assignment to a Dictionary? > > Yes. > > > If so, what would be the recommended way to determine the assignment is > to a Dictionary? > > The compile-time type of foo[bar], where foo has any type (not just > Dictionary), is type *. So I think it is OK to codegen coerce_a before > assignment to any foo[bar] expression. > > Darrell, do you agree or disagree? > > - Gordon > > > -Original Message- > From: Darrell Loverin [mailto:darrell.love...@gmail.com] > Sent: Monday, September 30, 2013 3:41 PM > To: dev@flex.apache.org > Subject: Re: [FALCON] using coerce_a > > coerce_a Operation > > Coerce a value to the any type. > > Format > > coerce_a > > Forms > > Stack > > ..., value => ..., value Description > > Indicates to the verifier that the value on the stack is of the any type > (*). Does nothing to value. > > > On Mon, Sep 30, 2013 at 6:03 PM, Gordon Smith wrote: > > > Can you remind me what the "a" in coerce_a means? > > > > - Gordon > > > > -Original Message- > > From: Alex Harui [mailto:aha...@adobe.com] > > Sent: Monday, September 30, 2013 2:43 PM > > To: dev@flex.apache.org > > Subject: [FALCON] using coerce_a > > > > Gordon, Darrell (mostly) > > > > My latest problem appears to be in assigning to a Dictionary. For the > > following code: > > > > var dict:Dictionary = new Dictionary(); Var i:int; Dict["foo"] = (i == > > 0 ? new SomeClass() : new SomeOtherClass()); > > > > MXMLC generates a coerce_a after the constructProp for both SomeClass > > and SomeOtherClass. > > For some reason, FlashPlayer doesn't care if the coerce_a is missing, > > but AIR seems to. > > > > The question is, should a complex assignment statement like the above > > know about the destination type as the clauses of the ternary > > statement are being reduced? And if so, how would we get that knowledge > into the reducer. > > > > Alternatively, is it safe to add a coerce_a before an assignment to a > > Dictionary? If so, what would be the recommended way to determine the > > assignment is to a Dictionary? > > > > Thanks, > > -Alex > > > > >
Re: Falcon progress
I added an assert that checks if the TLF_HOME variable is set (either system wide or in the unittest-properties). My guess is that if you set this variable to point to the flex-tlf working copy, that test will pass for you as well. EdB On Mon, Sep 30, 2013 at 10:25 PM, Gordon Smith wrote: > I pulled the latest code, including your change, and ran SDKSWCTests again (3 > times) inside Eclipse by right clicking on SDKSWCTests.java in Package > Explorer and choosing Debug As > JUnit Test. Only one test is still failing > for me, namely textLayoutSWC(). The JUnit view says > > java.lang.AssertionError: > Expected: is <0> > got: <1> > > at f.SDKSWCTests.compileSWC(SDKSWCTests.java:110) > at f.SDKSWCTests.textLayoutSWC(SDKSWCTests.java:357) > > When I run just that one test, the Console view says > > environment property - FLEX_HOME = D:\asf\flex-sdk > environment property - PLAYERGLOBAL_HOME = D:\asf\prerequisites\player > environment property - TLF_HOME = null > environment property - AIR_HOME = D:\asf\prerequisites\AdobeAIRSDK > environment property - FLASHPLAYER_DEBUGGER = > D:\asf\prerequisites\fp_11.1.102.63_archive\11_1r102_63_32bit_debug\flashplayer11_1r102_63_win_sa_debug_32bit.exe > Loading configuration: D:\asf\flex-sdk\frameworks\flex-config.xml > > command line > Error: unable to open > 'D:\asf\flex-falcon\compiler.tests\null\compile-config.xml'. > > - Gordon > > > -Original Message- > From: Erik de Bruin [mailto:e...@ixsoftware.nl] > Sent: Saturday, September 28, 2013 5:25 AM > To: dev@flex.apache.org > Subject: Re: Falcon progress > > Interesting... On my end only 'sparkSWC' fails. The other tests all pass as > configured (i.e. with -ignore-problems set to cause the tests not to fail on > warnings). I've added @Ignore on 'sparkSWC' as I agree all committed tests > should either pass or be ignored. > > What failures are you seeing on the 5 other tests that fail on your end? > > EdB > > > > On Tue, Sep 24, 2013 at 11:37 PM, Gordon Smith wrote: >> Congratulations on getting more SWCs to compile with Falcon! It looks like >> only the following 6 tests in SDKSWCTests still have problems: >> >> authoringsupportSWC() >> automation_flashflexkitSWC() >> chartsSWC() >> mxSWC() >> sparkSWC() >> textLayoutSWC() >> >> Is there a reason why the @Ignore was removed from these failing tests? I >> recommend keepng the Falcon test suite passing at all times, so that >> developers know when they've introduced a new failure. >> >> Most of the changes seem to have been minor changes to SDK code to placate >> Falcon since it is stricter than the old compiler, or changes to the >> compilation options for the test to suppress errors or warnings that we can >> fix later. However, I'm concerned about changes to Falcon itself like this >> that are workarounds rather than fixes: >> >> FLEX-33713: An NPE is thrown by this code when compiling 'spark.swc' >> from the SDK. The null check bypasses the issue, it doesn't do anything to >> fix the root cause... >> >> I think this kind of thing is OK as long as these changes are trackable in >> some way that makes it possible to go back and make the actual fix later. Do >> we have to just rely on 'git log' for to do this? If so, we should put >> something we can search for like "WORKAROUND" in the Git comment. >> >> Has anybody gotten a complete app to compile with Falcon yet? Does it run >> correctly? >> >> - Gordon >> > > > > -- > Ix Multimedia Software > > Jan Luykenstraat 27 > 3521 VB Utrecht > > T. 06-51952295 > I. www.ixsoftware.nl -- Ix Multimedia Software Jan Luykenstraat 27 3521 VB Utrecht T. 06-51952295 I. www.ixsoftware.nl
Re: [DISCUSS] Apache Flex Bylaws
Hi, > I think we should. Although I am not sure what the distinction means. Are > you thinking of restricting voting privileges to non active PMC? Yes and that's in line with most of the Apache project bylaws (that I've seen). They say you need to be an active committer or PMC member to vote. We generally get enough people voting that it's not an issue, except with release candidates and there you have to actually do something ie test/check the RC. > IMHO, there are way too many committers/PMC members who have not > contributed anything since the project started. No real harm is keeping then as it's a possibility are they might contribute again in the future. But perhaps have an email once every 6 months asking since you're not been active do you still want to be a committer/PMC member? ie give them the option to retire. Thanks, Justin
RE: [DISCUSS] Apache Flex Bylaws
I would prefer we not set a bar for active. Our former mentors are rarely active on the public lists, but I like having them involved in the vote and discussion on the private list. -Alex From: omup...@gmail.com [omup...@gmail.com] On Behalf Of OmPrakash Muppirala [bigosma...@gmail.com] Sent: Monday, September 30, 2013 10:48 PM To: dev@flex.apache.org Cc: Justin Mclean Subject: Re: [DISCUSS] Apache Flex Bylaws On Sep 30, 2013 5:59 PM, "Jeffry Houser" wrote: > > On 9/30/2013 7:19 PM, Justin Mclean wrote: >> >> Hi, >> >> Looks like as part of the graduation process that we didn't get together and come up with a list of bylaws for Apache Flex. >> >> Some example Apache project bylaws: >> http://hadoop.apache.org/bylaws.html >> http://cloudstack.apache.org/bylaws.html >> https://cwiki.apache.org/confluence/display/Hive/Bylaws >> https://cwiki.apache.org/confluence/display/KAFKA/Bylaws >> http://ant.apache.org/bylaws.html >> >> As well as listing out the roles and responsibilities on the project it needs to spell out the voting rules. >> >> Here's some of the things we would need to reach consensus on. If PMC members (and others) could indicate >> their preferences to there questions below we'll start getting a document together based on this what other >> Apache projects have done. >> >> If you have no strong preference indicate that and this is not a vote, just more of what you think bests fits in our >> community. >> >> 1. Voting procedure for committers >> a) lazy-majority (more +1s than -1s) >> b) lazy-consensus (no -1's) >> c) consensus (3 +1's no -1's) > > b > >> 2. Voting procedure for PMC members >> a) lazy-majority (more +1s than -1s) >> b) lazy-consensus (no -1's) >> c) consensus (3 +1's no -1's) >> d) 2/3 majority (3 +1's and twice an many +1 as -1) > > c >> >> 3. Voting procedure for Chair >> a) lazy-majority (more +1s than -1s) >> b) lazy-consensus (no -1's) >> c) consensus (3 +1's no -1's) >> d) 2/3 majority (3 +1's and twice an many +1 as -1) > > c >> >> 4. Length of term of Chair >> a) until they resign >> b) 1 year >> d) reviewed every year with option to continue >> c) other > > d >> >> 5. Voting procedure on changing bylaws >> a) lazy-majority (more +1s than -1s) >> b) lazy-consensus (no -1's) >> c) consensus (3 +1's no -1's) >> d) 2/3 majority (3 +1's and twice an many +1 as -1) > > d >> >> I notice other Bylaws use the word active a lot eg "active PMC", "active committers", but don't seem to define it. >> Should we try and define or at least give some guidelines what an active person on the project is? >> There seems to be a vague "not contributing to the project for 6 months" in one or two of the existing bylaws. > > > I guess we'd have to define formally what a contribution is. Is it activity on a mailing list? Actual commits? Jira activity? Mailing list moderation? > It sounds like it is a hard thing to concretely define. Do you want to take a stab at defining it? I think the bar should be pretty low. Something like at least one email on dev, users or private in 6 months. Or a commit in 6 months. > > -- > Jeffry Houser > Technical Entrepreneur > http://www.jeffryhouser.com > 203-379-0773 >
Re: [DISCUSS] Apache Flex Bylaws
HI, > I think the bar should be pretty low. Something like at least one email on > dev, users or private in 6 months. Or a commit in 6 months. > How about this (a little long I know but covers all bases I think): In the last 3 months: An email to the dev, private or user list, or a JIRA issue commented on or raised, or the website or wiki edited. In the last 6 months: Voted on something, or committed a patch or tested a release candidate? Or person is a pervious chair or mentor. That should catch even the most minimally active people. I can't think of a single case where an "inactive" person has voted. Thanks, Justin
