Re: [Vote] release for myfaces 1.2.1
Is this still a vote now? Can we start a separate thread? regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: the plugin includes the notice/license if not present, it generates them -M On Dec 20, 2007 8:06 AM, simon [EMAIL PROTECTED] wrote: On Thu, 2007-12-20 at 07:52 +0100, Matthias Wessendorf wrote: As it happens, there is quite a vigorous discussion going on right now in both apache commons and legal-discuss lists regarding NOTICE and LICENSE files. Some people would like to use a maven feature to auto-generate LICENSE and NOTICE files. This is quite controversial, though, and a number of people prefer the existing apache-commons way (including myself). I like having them in the root folder, like done in Trinidad https://svn.apache.org/repos/asf/myfaces/trinidad/trunk_1.2.x/ but, the auto-generation plugins is fine for including them in every JAR. Trinidad uses that plugin as well. ?? Do you mean there is a license and notice checked in, but the built jar ignores these and uses something different instead? That sounds like the worst of both worlds to me. It would: (a) double the work, as *both* need to be checked for correctness, and (b) is really confusing for users if they differ. Regards, Simon -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Re: [Vote] release for Trinidad Plugins 1.2.6
+1 regards, Martin On 12/20/07, Bernd Bohmann [EMAIL PROTECTED] wrote: +1 Matthias Wessendorf schrieb: yes, that is correct. -M On Dec 20, 2007 8:27 AM, Bernd Bohmann [EMAIL PROTECTED] wrote: Hello, are the release notes correct? http://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12310661styleName=Htmlversion=12312879 Regards Bernd Matthias Wessendorf schrieb: +1 On Dec 20, 2007 7:39 AM, Leonardo Uribe [EMAIL PROTECTED] wrote: Hi, I was running the needed tasks to get the 1.2.6 release of Trinidad Plugins out. This task is necessary for release Myfaces 1.2.1, because it uses faces plugin. Please note that this vote concerns all of the following parts: 1. Maven artifact group org.apache.myfaces.trinidadbuild v 1.2.6 [1] The artifacts are deployed to my private Apache account ([1]). Please take a look at the 1.2.6 artifacts and vote! Please note: This vote is majority approval with a minimum of three +1 votes (see [3]). -- -- [ ] +1 for community members who have reviewed the bits [ ] +0 [ ] -1 for fatal flaws that should cause these bits not to be released, and why.. Thanks, Leonardo Uribe [1] http://people.apache.org/~lu4242/trinidad-build126 [2] http://www.apache.org/foundation/voting.html#ReleaseVotes -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Re: Server state saving and multiple frames
Hi Nicu, we should include Mario in this discussion - he implemented a solution for this in Orchestra. Also, how about Trinidad, in Trinidad there is dialog handling as well, how is this done? regards, Martin On 12/19/07, simon [EMAIL PROTECTED] wrote: Hi Nicu, I haven't got time to look at this closely, but IMO siomething like this is definitely needed in MyFaces. A user with multiple windows is certainly going to have trouble at the moment. I think a modification to the view pool to include a window id (or frame id) is definitely a good idea. The second part of the problem still remains: how to associate a different id with each window/frame. Checking CommandLink components for a target attribute is clever; it doesn't solve all the cases but does solve some. Regards, Simon On Tue, 2007-12-18 at 19:07 +0200, Nicu Mercioiu wrote: Hi, There is a problem in JSF when more than one window are opened in an application. There are only a maximum number of NUMBER_OF_VIEWS_IN_SESSION view states saved at one moment (when server side state saving is enabled). If you have 2 windows opened and you navigate on one of them for NUMBER_OF_VIEWS_IN_SESSION times, you will lose the other window's state. I've been facing this problem while developing a project so I've implemented a solution for it. The solution is having a number of view states saved for each opened window at one moment. For determining when a new window (frame) is opened, the target of the submitting component (or its enclosing form) is used. This is obtained in the HtmlLinkRendererBase's and HtmlButtonRendererBase's decode methods and it is set in the RequestMap. Using the submitted target, the JspStateManagerImpl figures out whether a new frame was opened. If so, a new frame id is generated. In the renderResponse phase, the frameId is encoded in the javax.faces.ViewState field and is used along with the viewId to save the state in a SerializedViewCollection. In the restore view phase the frameId is decoded from the javax.faces.ViewState field and is used along with the viewId to restore the corresponding state from the SerializedViewCollection. In SerializedViewCollection instead of a list of recently used views, now a list is kept for each frameId. The following context params are defined for configuring this. NUMBER_OF_FRAMES_IN_SESSION (max frames stored) NUMBER_OF_VIEWS_IN_FRAME (max views stored per frame) These replace the old: NUMBER_OF_VIEWS_IN_SESSION context-param. What is your opinion on this solution? Of course this solution only works with MyFaces Tomahawk's commandLink and commandButton. Ohter component sets that do not use a custom stateManager might use this feature if they will just modify the renderers of command components to set the target attribute in the requestMap. An extra feature would be to enable this for outputLinks (plain old links) and for JS (openWindow). The solution for this is quite simple, just add a GET parameter named 'target' and set the value the same as the target attribute. In the JspStateManagerImpl this value is obtained from the requestParameterMap and used the same as in the other case. Do you think this would be useful too? Regards, Nicu -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
[jira] Commented: (TRINIDAD-845) choice between IFrame-PPR or XHR should be possible
[ https://issues.apache.org/jira/browse/TRINIDAD-845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12553610 ] Michael Kutsch commented on TRINIDAD-845: - We have the same problem. I am working for an insurance company and our new calculator for insurances will run on trinidad and use PPR. Since internally we use IE6 we cannot switch to XHR based newer versions. The calculator will be used by resellers too. I don't see that we can get them all to use IE7 or Firefox. The choice of the communication mechanism PPR or XHR is very important. choice between IFrame-PPR or XHR should be possible Key: TRINIDAD-845 URL: https://issues.apache.org/jira/browse/TRINIDAD-845 Project: MyFaces Trinidad Issue Type: New Feature Affects Versions: 1.2.2-core Reporter: Jan Görß It should be possible to change the interactive server communication between IFrame-PPR and XHR. Our customers are banking houses and they define the security constrains (IE5.5+ with deactived ActiveX). IE 5.5 - 6 are the problem. IE 7 has a possibility to run XHR without ActiveX. At the beginning of our projects we decide to take Trinidad because of it's IFrame-PPR solution and our WebApps must run without AJAX. I think other users of trinidad have the same problem also. Taken from [EMAIL PROTECTED] Hi, I know all the issues with not able to use IE w/ activated ActiveX. Sure, a solution is possible, but it is (as always) a matter of time. Are you willing to help out on this? I'll also check our internal requirements, since we have also customers, that may upgrade to Trinidad, coming from ADF Faces 10.x. Can you file an issue in our jira w/ some details, that we don't forget it? Thx, Matthias On Nov 28, 2007 1:38 PM, Goerss, Jan [EMAIL PROTECTED] wrote: The problem is that our customers are banking houses. They define the security constrains (IE5.5+ with deactivated ActiveX). We use Trinidad 1.2.1 and we can't switch to latest version of Trinidad with AJAX. I hope, a fallback option is possible, otherwise we can't use the latest version of Trinidad and our project can work with Trinidad version 1.2.1 only. The problems are with this constrain: - we get no bugfixing more - we can't use new components and features (new releases) - we can't use new trinidad development I hope a solution is possible. Thanks, Jan -Ursprüngliche Nachricht- Von: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Im Auftrag von Matthias Wessendorf Gesendet: Mittwoch, 28. November 2007 11:04 An: MyFaces Discussion Betreff: Re: Trinidad Ajax Hi Jan, yes, we switched to XHR with the release of 1.0.2 (and 1.2.2). There is no way to use IFrame instead of XHR. This will remain, because Ajax is pretty common these days and other JSF-libs have that as well. Providing an option to fallback might be possible, but not sure if that will be done. Is IE 5.5 really still supported by M$? If not, I'd strongly recommend to kick that guy out. Sure... only the IE7 doesn't require ActiveX for XHR, but IE does. I am not aware of a solution for that. Do you have any ideas ? -Matthias On Nov 28, 2007 10:48 AM, Goerss, Jan [EMAIL PROTECTED] wrote: Hello all, we noticed that MyFaces Trinidad is changing the PPR mechanism towards real AJAX and the XmlHttpRequest object. We are using Trinidad is the finance sector. There it is normal that the clients use the Internet Explorer with deactivated ActiveX. Furthermore they use not the newest browsers, so we although need to support at least IE 5.5+. At the beginning of our project we decide to take Trinidad because of it's IFrame-PPR solution. It seems as if it is not possible to choose between the two modes: PPR or AJAX. Will this remain? If so why? Otherwise, we think that a lot of Trinidad library users will encounter problems with their clientsJ Thanks in advance, Jan -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
Re: Server state saving and multiple frames
Hi, anyone is invited to contribute with ideas of course, but I still don't understand what this (multiple-frames state saving) has to do with dialog handling and conversations!? On Dec 20, 2007 9:13 AM, Martin Marinschek [EMAIL PROTECTED] wrote: Hi Nicu, we should include Mario in this discussion - he implemented a solution for this in Orchestra. Also, how about Trinidad, in Trinidad there is dialog handling as well, how is this done? regards, Martin On 12/19/07, simon [EMAIL PROTECTED] wrote: Hi Nicu, I haven't got time to look at this closely, but IMO siomething like this is definitely needed in MyFaces. A user with multiple windows is certainly going to have trouble at the moment. I think a modification to the view pool to include a window id (or frame id) is definitely a good idea. The second part of the problem still remains: how to associate a different id with each window/frame. Checking CommandLink components for a target attribute is clever; it doesn't solve all the cases but does solve some. Regards, Simon On Tue, 2007-12-18 at 19:07 +0200, Nicu Mercioiu wrote: Hi, There is a problem in JSF when more than one window are opened in an application. There are only a maximum number of NUMBER_OF_VIEWS_IN_SESSION view states saved at one moment (when server side state saving is enabled). If you have 2 windows opened and you navigate on one of them for NUMBER_OF_VIEWS_IN_SESSION times, you will lose the other window's state. I've been facing this problem while developing a project so I've implemented a solution for it. The solution is having a number of view states saved for each opened window at one moment. For determining when a new window (frame) is opened, the target of the submitting component (or its enclosing form) is used. This is obtained in the HtmlLinkRendererBase's and HtmlButtonRendererBase's decode methods and it is set in the RequestMap. Using the submitted target, the JspStateManagerImpl figures out whether a new frame was opened. If so, a new frame id is generated. In the renderResponse phase, the frameId is encoded in the javax.faces.ViewState field and is used along with the viewId to save the state in a SerializedViewCollection. In the restore view phase the frameId is decoded from the javax.faces.ViewState field and is used along with the viewId to restore the corresponding state from the SerializedViewCollection. In SerializedViewCollection instead of a list of recently used views, now a list is kept for each frameId. The following context params are defined for configuring this. NUMBER_OF_FRAMES_IN_SESSION (max frames stored) NUMBER_OF_VIEWS_IN_FRAME (max views stored per frame) These replace the old: NUMBER_OF_VIEWS_IN_SESSION context-param. What is your opinion on this solution? Of course this solution only works with MyFaces Tomahawk's commandLink and commandButton. Ohter component sets that do not use a custom stateManager might use this feature if they will just modify the renderers of command components to set the target attribute in the requestMap. An extra feature would be to enable this for outputLinks (plain old links) and for JS (openWindow). The solution for this is quite simple, just add a GET parameter named 'target' and set the value the same as the target attribute. In the JspStateManagerImpl this value is obtained from the requestParameterMap and used the same as in the other case. Do you think this would be useful too? Regards, Nicu -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- Cristi Toth - Codebeat www.codebeat.ro
[jira] Commented: (TRINIDAD-845) choice between IFrame-PPR or XHR should be possible
[ https://issues.apache.org/jira/browse/TRINIDAD-845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12553613 ] Matthias Weßendorf commented on TRINIDAD-845: - IFrame or XHR you mean. PPR is the Trinidad feature. The impl-detail was IFrame-based; now it is XHR-based; choice between IFrame-PPR or XHR should be possible Key: TRINIDAD-845 URL: https://issues.apache.org/jira/browse/TRINIDAD-845 Project: MyFaces Trinidad Issue Type: New Feature Affects Versions: 1.2.2-core Reporter: Jan Görß It should be possible to change the interactive server communication between IFrame-PPR and XHR. Our customers are banking houses and they define the security constrains (IE5.5+ with deactived ActiveX). IE 5.5 - 6 are the problem. IE 7 has a possibility to run XHR without ActiveX. At the beginning of our projects we decide to take Trinidad because of it's IFrame-PPR solution and our WebApps must run without AJAX. I think other users of trinidad have the same problem also. Taken from [EMAIL PROTECTED] Hi, I know all the issues with not able to use IE w/ activated ActiveX. Sure, a solution is possible, but it is (as always) a matter of time. Are you willing to help out on this? I'll also check our internal requirements, since we have also customers, that may upgrade to Trinidad, coming from ADF Faces 10.x. Can you file an issue in our jira w/ some details, that we don't forget it? Thx, Matthias On Nov 28, 2007 1:38 PM, Goerss, Jan [EMAIL PROTECTED] wrote: The problem is that our customers are banking houses. They define the security constrains (IE5.5+ with deactivated ActiveX). We use Trinidad 1.2.1 and we can't switch to latest version of Trinidad with AJAX. I hope, a fallback option is possible, otherwise we can't use the latest version of Trinidad and our project can work with Trinidad version 1.2.1 only. The problems are with this constrain: - we get no bugfixing more - we can't use new components and features (new releases) - we can't use new trinidad development I hope a solution is possible. Thanks, Jan -Ursprüngliche Nachricht- Von: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Im Auftrag von Matthias Wessendorf Gesendet: Mittwoch, 28. November 2007 11:04 An: MyFaces Discussion Betreff: Re: Trinidad Ajax Hi Jan, yes, we switched to XHR with the release of 1.0.2 (and 1.2.2). There is no way to use IFrame instead of XHR. This will remain, because Ajax is pretty common these days and other JSF-libs have that as well. Providing an option to fallback might be possible, but not sure if that will be done. Is IE 5.5 really still supported by M$? If not, I'd strongly recommend to kick that guy out. Sure... only the IE7 doesn't require ActiveX for XHR, but IE does. I am not aware of a solution for that. Do you have any ideas ? -Matthias On Nov 28, 2007 10:48 AM, Goerss, Jan [EMAIL PROTECTED] wrote: Hello all, we noticed that MyFaces Trinidad is changing the PPR mechanism towards real AJAX and the XmlHttpRequest object. We are using Trinidad is the finance sector. There it is normal that the clients use the Internet Explorer with deactivated ActiveX. Furthermore they use not the newest browsers, so we although need to support at least IE 5.5+. At the beginning of our project we decide to take Trinidad because of it's IFrame-PPR solution. It seems as if it is not possible to choose between the two modes: PPR or AJAX. Will this remain? If so why? Otherwise, we think that a lot of Trinidad library users will encounter problems with their clientsJ Thanks in advance, Jan -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
Re: Server state saving and multiple frames
Orchestra provides a tag that can be explicitly wrapped around links (eg h:outputLink). It then encodes a window id into the url params. So it also partially addresses the second part of the problem (identifying windows/frames). It covers more cases than Nicu's suggestion, but it does require modifications to pages. Orchestra then uses this window-id to provide a new set of conversation-scoped beans per window. But nothing in orchestra addresses the problem of the view-state pool, so unless I misunderstand something this Orchestra feature is not really usable; the backing beans will work right but when the viewstate gets screwed up after 4 page-views that is no consolation. I think myfaces core needs to provide a per-window viewstate cache. It can then just allow something else to figure out how to identify the windows, allowing the Orchestra approach or Nicu's approach, or whatever people come up with. Yes, it would be interesting to hear how Trinidad dialogs handle this.. Regards, Simon Martin Marinschek [EMAIL PROTECTED] schrieb: Hi Nicu, we should include Mario in this discussion - he implemented a solution for this in Orchestra. Also, how about Trinidad, in Trinidad there is dialog handling as well, how is this done? regards, Martin On 12/19/07, simon [EMAIL PROTECTED] wrote: Hi Nicu, I haven't got time to look at this closely, but IMO siomething like this is definitely needed in MyFaces. A user with multiple windows is certainly going to have trouble at the moment. I think a modification to the view pool to include a window id (or frame id) is definitely a good idea. The second part of the problem still remains: how to associate a different id with each window/frame. Checking CommandLink components for a target attribute is clever; it doesn't solve all the cases but does solve some. Regards, Simon On Tue, 2007-12-18 at 19:07 +0200, Nicu Mercioiu wrote: Hi, There is a problem in JSF when more than one window are opened in an application. There are only a maximum number of NUMBER_OF_VIEWS_IN_SESSION view states saved at one moment (when server side state saving is enabled). If you have 2 windows opened and you navigate on one of them for NUMBER_OF_VIEWS_IN_SESSION times, you will lose the other window's state. I've been facing this problem while developing a project so I've implemented a solution for it. The solution is having a number of view states saved for each opened window at one moment. For determining when a new window (frame) is opened, the target of the submitting component (or its enclosing form) is used. This is obtained in the HtmlLinkRendererBase's and HtmlButtonRendererBase's decode methods and it is set in the RequestMap. Using the submitted target, the JspStateManagerImpl figures out whether a new frame was opened. If so, a new frame id is generated. In the renderResponse phase, the frameId is encoded in the javax.faces.ViewState field and is used along with the viewId to save the state in a SerializedViewCollection. In the restore view phase the frameId is decoded from the javax.faces.ViewState field and is used along with the viewId to restore the corresponding state from the SerializedViewCollection. In SerializedViewCollection instead of a list of recently used views, now a list is kept for each frameId. The following context params are defined for configuring this. NUMBER_OF_FRAMES_IN_SESSION (max frames stored) NUMBER_OF_VIEWS_IN_FRAME (max views stored per frame) These replace the old: NUMBER_OF_VIEWS_IN_SESSION context-param. What is your opinion on this solution? Of course this solution only works with MyFaces Tomahawk's commandLink and commandButton. Ohter component sets that do not use a custom stateManager might use this feature if they will just modify the renderers of command components to set the target attribute in the requestMap. An extra feature would be to enable this for outputLinks (plain old links) and for JS (openWindow). The solution for this is quite simple, just add a GET parameter named 'target' and set the value the same as the target attribute. In the JspStateManagerImpl this value is obtained from the requestParameterMap and used the same as in the other case. Do you think this would be useful too? Regards, Nicu -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Re: [Vote] release for Trinidad Plugins 1.2.6
+1 Thank you Leonardo! Bruno On 20/12/2007, Martin Marinschek [EMAIL PROTECTED] wrote: +1 regards, Martin On 12/20/07, Bernd Bohmann [EMAIL PROTECTED] wrote: +1 Matthias Wessendorf schrieb: yes, that is correct. -M On Dec 20, 2007 8:27 AM, Bernd Bohmann [EMAIL PROTECTED] wrote: Hello, are the release notes correct? http://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12310661styleName=Htmlversion=12312879 Regards Bernd Matthias Wessendorf schrieb: +1 On Dec 20, 2007 7:39 AM, Leonardo Uribe [EMAIL PROTECTED] wrote: Hi, I was running the needed tasks to get the 1.2.6 release of Trinidad Plugins out. This task is necessary for release Myfaces 1.2.1, because it uses faces plugin. Please note that this vote concerns all of the following parts: 1. Maven artifact group org.apache.myfaces.trinidadbuild v 1.2.6 [1] The artifacts are deployed to my private Apache account ([1]). Please take a look at the 1.2.6 artifacts and vote! Please note: This vote is majority approval with a minimum of three +1 votes (see [3]). -- -- [ ] +1 for community members who have reviewed the bits [ ] +0 [ ] -1 for fatal flaws that should cause these bits not to be released, and why.. Thanks, Leonardo Uribe [1] http://people.apache.org/~lu4242/trinidad-build126 [2] http://www.apache.org/foundation/voting.html#ReleaseVotes -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
[jira] Commented: (TRINIDAD-845) choice between IFrame-PPR or XHR should be possible
[ https://issues.apache.org/jira/browse/TRINIDAD-845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12553615 ] Matthias Weßendorf commented on TRINIDAD-845: - Michael, does IE6 supports XHR, but only via ActiveX. Does that mean, you have ActiveX disabled as well ? choice between IFrame-PPR or XHR should be possible Key: TRINIDAD-845 URL: https://issues.apache.org/jira/browse/TRINIDAD-845 Project: MyFaces Trinidad Issue Type: New Feature Affects Versions: 1.2.2-core Reporter: Jan Görß It should be possible to change the interactive server communication between IFrame-PPR and XHR. Our customers are banking houses and they define the security constrains (IE5.5+ with deactived ActiveX). IE 5.5 - 6 are the problem. IE 7 has a possibility to run XHR without ActiveX. At the beginning of our projects we decide to take Trinidad because of it's IFrame-PPR solution and our WebApps must run without AJAX. I think other users of trinidad have the same problem also. Taken from [EMAIL PROTECTED] Hi, I know all the issues with not able to use IE w/ activated ActiveX. Sure, a solution is possible, but it is (as always) a matter of time. Are you willing to help out on this? I'll also check our internal requirements, since we have also customers, that may upgrade to Trinidad, coming from ADF Faces 10.x. Can you file an issue in our jira w/ some details, that we don't forget it? Thx, Matthias On Nov 28, 2007 1:38 PM, Goerss, Jan [EMAIL PROTECTED] wrote: The problem is that our customers are banking houses. They define the security constrains (IE5.5+ with deactivated ActiveX). We use Trinidad 1.2.1 and we can't switch to latest version of Trinidad with AJAX. I hope, a fallback option is possible, otherwise we can't use the latest version of Trinidad and our project can work with Trinidad version 1.2.1 only. The problems are with this constrain: - we get no bugfixing more - we can't use new components and features (new releases) - we can't use new trinidad development I hope a solution is possible. Thanks, Jan -Ursprüngliche Nachricht- Von: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Im Auftrag von Matthias Wessendorf Gesendet: Mittwoch, 28. November 2007 11:04 An: MyFaces Discussion Betreff: Re: Trinidad Ajax Hi Jan, yes, we switched to XHR with the release of 1.0.2 (and 1.2.2). There is no way to use IFrame instead of XHR. This will remain, because Ajax is pretty common these days and other JSF-libs have that as well. Providing an option to fallback might be possible, but not sure if that will be done. Is IE 5.5 really still supported by M$? If not, I'd strongly recommend to kick that guy out. Sure... only the IE7 doesn't require ActiveX for XHR, but IE does. I am not aware of a solution for that. Do you have any ideas ? -Matthias On Nov 28, 2007 10:48 AM, Goerss, Jan [EMAIL PROTECTED] wrote: Hello all, we noticed that MyFaces Trinidad is changing the PPR mechanism towards real AJAX and the XmlHttpRequest object. We are using Trinidad is the finance sector. There it is normal that the clients use the Internet Explorer with deactivated ActiveX. Furthermore they use not the newest browsers, so we although need to support at least IE 5.5+. At the beginning of our project we decide to take Trinidad because of it's IFrame-PPR solution. It seems as if it is not possible to choose between the two modes: PPR or AJAX. Will this remain? If so why? Otherwise, we think that a lot of Trinidad library users will encounter problems with their clientsJ Thanks in advance, Jan -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
Re: Server state saving and multiple frames
Nicu's solution works fine for MyFaces commandLinks using the MyFaces JspStateManager. Someone using MyFaces Tomahawk can take advantage out of this new view pool approach. The big problem here: It does not work using a component library that decorates the statemanager (eg. trinidad). So far I did not take a look to trinidad's statesaving mechanism. It'll be worth to think about a more common solution for having a state view pool. -Thomas On Dec 20, 2007 9:41 AM, Simon Kitching [EMAIL PROTECTED] wrote: Orchestra provides a tag that can be explicitly wrapped around links (eg h:outputLink). It then encodes a window id into the url params. So it also partially addresses the second part of the problem (identifying windows/frames). It covers more cases than Nicu's suggestion, but it does require modifications to pages. Orchestra then uses this window-id to provide a new set of conversation-scoped beans per window. But nothing in orchestra addresses the problem of the view-state pool, so unless I misunderstand something this Orchestra feature is not really usable; the backing beans will work right but when the viewstate gets screwed up after 4 page-views that is no consolation. I think myfaces core needs to provide a per-window viewstate cache. It can then just allow something else to figure out how to identify the windows, allowing the Orchestra approach or Nicu's approach, or whatever people come up with. Yes, it would be interesting to hear how Trinidad dialogs handle this.. Regards, Simon Martin Marinschek [EMAIL PROTECTED] schrieb: Hi Nicu, we should include Mario in this discussion - he implemented a solution for this in Orchestra. Also, how about Trinidad, in Trinidad there is dialog handling as well, how is this done? regards, Martin On 12/19/07, simon [EMAIL PROTECTED] wrote: Hi Nicu, I haven't got time to look at this closely, but IMO siomething like this is definitely needed in MyFaces. A user with multiple windows is certainly going to have trouble at the moment. I think a modification to the view pool to include a window id (or frame id) is definitely a good idea. The second part of the problem still remains: how to associate a different id with each window/frame. Checking CommandLink components for a target attribute is clever; it doesn't solve all the cases but does solve some. Regards, Simon On Tue, 2007-12-18 at 19:07 +0200, Nicu Mercioiu wrote: Hi, There is a problem in JSF when more than one window are opened in an application. There are only a maximum number of NUMBER_OF_VIEWS_IN_SESSION view states saved at one moment (when server side state saving is enabled). If you have 2 windows opened and you navigate on one of them for NUMBER_OF_VIEWS_IN_SESSION times, you will lose the other window's state. I've been facing this problem while developing a project so I've implemented a solution for it. The solution is having a number of view states saved for each opened window at one moment. For determining when a new window (frame) is opened, the target of the submitting component (or its enclosing form) is used. This is obtained in the HtmlLinkRendererBase's and HtmlButtonRendererBase's decode methods and it is set in the RequestMap. Using the submitted target, the JspStateManagerImpl figures out whether a new frame was opened. If so, a new frame id is generated. In the renderResponse phase, the frameId is encoded in the javax.faces.ViewState field and is used along with the viewId to save the state in a SerializedViewCollection. In the restore view phase the frameId is decoded from the javax.faces.ViewState field and is used along with the viewId to restore the corresponding state from the SerializedViewCollection. In SerializedViewCollection instead of a list of recently used views, now a list is kept for each frameId. The following context params are defined for configuring this. NUMBER_OF_FRAMES_IN_SESSION (max frames stored) NUMBER_OF_VIEWS_IN_FRAME (max views stored per frame) These replace the old: NUMBER_OF_VIEWS_IN_SESSION context-param. What is your opinion on this solution? Of course this solution only works with MyFaces Tomahawk's commandLink and commandButton. Ohter component sets that do not use a custom stateManager might use this feature if they will just modify the renderers of command components to set the target attribute in the requestMap. An extra feature would be to enable this for outputLinks (plain old links) and for JS (openWindow). The solution for this is quite simple, just add a GET parameter named 'target' and set the value the same as the target attribute. In the JspStateManagerImpl this value is obtained
[jira] Created: (MYFACES-1791) state management and multiple frames
state management and multiple frames - Key: MYFACES-1791 URL: https://issues.apache.org/jira/browse/MYFACES-1791 Project: MyFaces Core Issue Type: Improvement Components: General Affects Versions: 1.2.0, 1.1.5, 1.1.6-SNAPSHOT, 1.2.2-SNAPSHOT Reporter: Nicu Mercioiu There is a problem in JSF when more than one window are opened in an application. There are only a maximum number of NUMBER_OF_VIEWS_IN_SESSION view states saved at one moment (when server side state saving is enabled). If 2 windows are opened and you navigate on one of them for NUMBER_OF_VIEWS_IN_SESSION times, you will lose the other window's state. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
Re: [Vote] release for myfaces 1.2.1
tested Trinidad 12x trunk against it. Fine so far! On Dec 20, 2007 7:53 AM, Matthias Wessendorf [EMAIL PROTECTED] wrote: Thanks for replacing the bits. +1 -Matthias On Dec 19, 2007 9:13 PM, simon [EMAIL PROTECTED] wrote: On Wed, 2007-12-19 at 11:54 -0800, Grant Smith wrote: +1 That was very quick, Grant! Have you really inspected all the artifacts Leonardo created to see if they are right? Licenses need to be correct, MANIFEST files should be double-checked, checksums verified, etc. Regards, Simon -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
[Trinidad] 1.0.x trunk
Hi, just moved the 1.0.x (for JSF 1.1) to this location: https://svn.apache.org/repos/asf/myfaces/trinidad/trunk/ so, no more trinidad/trinidad -M -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: Server state saving and multiple frames
My solution doesn't work only with commandLinks but works as well with outputLink or JavaScript by adding a GET parameter like in the following example: h:outputLink value=multipleFrames.jsf target=_blank f:param name=target value=_blank/ /h:outputLink It also works for any submitted form with a target attribute, for example a form submitted by commandButton. Regards, Nicu On Dec 20, 2007 10:41 AM, Simon Kitching [EMAIL PROTECTED] wrote: Orchestra provides a tag that can be explicitly wrapped around links (eg h:outputLink). It then encodes a window id into the url params. So it also partially addresses the second part of the problem (identifying windows/frames). It covers more cases than Nicu's suggestion, but it does require modifications to pages. Orchestra then uses this window-id to provide a new set of conversation-scoped beans per window. But nothing in orchestra addresses the problem of the view-state pool, so unless I misunderstand something this Orchestra feature is not really usable; the backing beans will work right but when the viewstate gets screwed up after 4 page-views that is no consolation. I think myfaces core needs to provide a per-window viewstate cache. It can then just allow something else to figure out how to identify the windows, allowing the Orchestra approach or Nicu's approach, or whatever people come up with. Yes, it would be interesting to hear how Trinidad dialogs handle this.. Regards, Simon Martin Marinschek [EMAIL PROTECTED] schrieb: Hi Nicu, we should include Mario in this discussion - he implemented a solution for this in Orchestra. Also, how about Trinidad, in Trinidad there is dialog handling as well, how is this done? regards, Martin On 12/19/07, simon [EMAIL PROTECTED] wrote: Hi Nicu, I haven't got time to look at this closely, but IMO siomething like this is definitely needed in MyFaces. A user with multiple windows is certainly going to have trouble at the moment. I think a modification to the view pool to include a window id (or frame id) is definitely a good idea. The second part of the problem still remains: how to associate a different id with each window/frame. Checking CommandLink components for a target attribute is clever; it doesn't solve all the cases but does solve some. Regards, Simon On Tue, 2007-12-18 at 19:07 +0200, Nicu Mercioiu wrote: Hi, There is a problem in JSF when more than one window are opened in an application. There are only a maximum number of NUMBER_OF_VIEWS_IN_SESSION view states saved at one moment (when server side state saving is enabled). If you have 2 windows opened and you navigate on one of them for NUMBER_OF_VIEWS_IN_SESSION times, you will lose the other window's state. I've been facing this problem while developing a project so I've implemented a solution for it. The solution is having a number of view states saved for each opened window at one moment. For determining when a new window (frame) is opened, the target of the submitting component (or its enclosing form) is used. This is obtained in the HtmlLinkRendererBase's and HtmlButtonRendererBase's decode methods and it is set in the RequestMap. Using the submitted target, the JspStateManagerImpl figures out whether a new frame was opened. If so, a new frame id is generated. In the renderResponse phase, the frameId is encoded in the javax.faces.ViewState field and is used along with the viewId to save the state in a SerializedViewCollection. In the restore view phase the frameId is decoded from the javax.faces.ViewState field and is used along with the viewId to restore the corresponding state from the SerializedViewCollection. In SerializedViewCollection instead of a list of recently used views, now a list is kept for each frameId. The following context params are defined for configuring this. NUMBER_OF_FRAMES_IN_SESSION (max frames stored) NUMBER_OF_VIEWS_IN_FRAME (max views stored per frame) These replace the old: NUMBER_OF_VIEWS_IN_SESSION context-param. What is your opinion on this solution? Of course this solution only works with MyFaces Tomahawk's commandLink and commandButton. Ohter component sets that do not use a custom stateManager might use this feature if they will just modify the renderers of command components to set the target attribute in the requestMap. An extra feature would be to enable this for outputLinks (plain old links) and for JS (openWindow). The solution for this is quite simple, just add a GET parameter named 'target' and set the value the same as the target attribute. In the JspStateManagerImpl this value is obtained from the requestParameterMap and used the same as
Re: Server state saving and multiple frames
What Thomas said about Trinidad and any other component library with their own state manager is very true. It would be nice to have the SerializedViewCollection (the view pool from JspStateManagerImpl) to be delegated to handle the actual storing of the views. But that still implies modifying those state managers On Dec 20, 2007 11:01 AM, Nicu Mercioiu [EMAIL PROTECTED] wrote: My solution doesn't work only with commandLinks but works as well with outputLink or JavaScript by adding a GET parameter like in the following example: h:outputLink value=multipleFrames.jsf target=_blank f:param name=target value=_blank/ /h:outputLink It also works for any submitted form with a target attribute, for example a form submitted by commandButton. Regards, Nicu On Dec 20, 2007 10:41 AM, Simon Kitching [EMAIL PROTECTED] wrote: Orchestra provides a tag that can be explicitly wrapped around links (eg h:outputLink). It then encodes a window id into the url params. So it also partially addresses the second part of the problem (identifying windows/frames). It covers more cases than Nicu's suggestion, but it does require modifications to pages. Orchestra then uses this window-id to provide a new set of conversation-scoped beans per window. But nothing in orchestra addresses the problem of the view-state pool, so unless I misunderstand something this Orchestra feature is not really usable; the backing beans will work right but when the viewstate gets screwed up after 4 page-views that is no consolation. I think myfaces core needs to provide a per-window viewstate cache. It can then just allow something else to figure out how to identify the windows, allowing the Orchestra approach or Nicu's approach, or whatever people come up with. Yes, it would be interesting to hear how Trinidad dialogs handle this.. Regards, Simon Martin Marinschek [EMAIL PROTECTED] schrieb: Hi Nicu, we should include Mario in this discussion - he implemented a solution for this in Orchestra. Also, how about Trinidad, in Trinidad there is dialog handling as well, how is this done? regards, Martin On 12/19/07, simon [EMAIL PROTECTED] wrote: Hi Nicu, I haven't got time to look at this closely, but IMO siomething like this is definitely needed in MyFaces. A user with multiple windows is certainly going to have trouble at the moment. I think a modification to the view pool to include a window id (or frame id) is definitely a good idea. The second part of the problem still remains: how to associate a different id with each window/frame. Checking CommandLink components for a target attribute is clever; it doesn't solve all the cases but does solve some. Regards, Simon On Tue, 2007-12-18 at 19:07 +0200, Nicu Mercioiu wrote: Hi, There is a problem in JSF when more than one window are opened in an application. There are only a maximum number of NUMBER_OF_VIEWS_IN_SESSION view states saved at one moment (when server side state saving is enabled). If you have 2 windows opened and you navigate on one of them for NUMBER_OF_VIEWS_IN_SESSION times, you will lose the other window's state. I've been facing this problem while developing a project so I've implemented a solution for it. The solution is having a number of view states saved for each opened window at one moment. For determining when a new window (frame) is opened, the target of the submitting component (or its enclosing form) is used. This is obtained in the HtmlLinkRendererBase's and HtmlButtonRendererBase's decode methods and it is set in the RequestMap. Using the submitted target, the JspStateManagerImpl figures out whether a new frame was opened. If so, a new frame id is generated. In the renderResponse phase, the frameId is encoded in the javax.faces.ViewState field and is used along with the viewId to save the state in a SerializedViewCollection. In the restore view phase the frameId is decoded from the javax.faces.ViewState field and is used along with the viewId to restore the corresponding state from the SerializedViewCollection. In SerializedViewCollection instead of a list of recently used views, now a list is kept for each frameId. The following context params are defined for configuring this. NUMBER_OF_FRAMES_IN_SESSION (max frames stored) NUMBER_OF_VIEWS_IN_FRAME (max views stored per frame) These replace the old: NUMBER_OF_VIEWS_IN_SESSION context-param. What is your opinion on this solution? Of course this solution only works with MyFaces Tomahawk's commandLink and commandButton. Ohter component sets that do not use a custom stateManager might use
Class missing from Source
Hello there, I am trying to debug a myfaces 1.2.0 app but the sources download does not contain the WebXml class. I have checked both the maven repository and the main download site. The snapshot version for 1.2.1 also does not have this class included. Thanks, Carl -- View this message in context: http://www.nabble.com/Class-missing-from-Source-tp14433768p14433768.html Sent from the My Faces - Dev mailing list archive at Nabble.com.
Re: [Trinidad] Maven Plugins
ok... looks like myfaces-build-tools won the race. structure will be /myfaces-build-tools -/maven2-plugins -/all the current (trin) plugins -tbc -M On Dec 15, 2007 11:12 PM, Matthias Wessendorf [EMAIL PROTECTED] wrote: Hi, there is this maven plugins folder for the Trinidad Plugins (see [1]). The Trinidad inside the name may confuse people. The plugins are generally usable outside of Trinidad as well. The maven-faces-plugin for instance, is used in other projects, such as MyFaces Core, or MyFaces Commons. The idea is to move them into a different location. A name like myfaces developers toolkit, myfaces dev tools, or what ever you like ? The idea would be to keep it abstract like /myfaces-dev-tools -/maven2-plugins -tbc that would allow us to have maven3 or ant or what ever support as well. Greetings, Matze Manfred [1] https://svn.apache.org/repos/asf/myfaces/trinidad-maven/trunk/ -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: Class missing from Source
Hi, that is part of the shared; package org.apache.myfaces.shared_impl.webapp.webxml check this JAR: http://repo1.maven.org/maven2/org/apache/myfaces/shared/myfaces-shared-impl/3.0.0/myfaces-shared-impl-3.0.0-sources.jar (was used to build 1.2.0) -M On Dec 20, 2007 11:14 AM, Carl Howarth [EMAIL PROTECTED] wrote: Hello there, I am trying to debug a myfaces 1.2.0 app but the sources download does not contain the WebXml class. I have checked both the maven repository and the main download site. The snapshot version for 1.2.1 also does not have this class included. Thanks, Carl -- View this message in context: http://www.nabble.com/Class-missing-from-Source-tp14433768p14433768.html Sent from the My Faces - Dev mailing list archive at Nabble.com. -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: Class missing from Source
But the source download for myfaces really should include the shared classes too. The .class files are in the myfaces impl jar, so the (repackaged) sources should be in the corresponding source jarfile. I would *really* like to see this fixed for the next release. It's a pain to be stepping through code and then not be able to find the source without adding a source jar that is not part of the declared dependencies at all.. Regards, Simon Matthias Wessendorf [EMAIL PROTECTED] schrieb: Hi, that is part of the shared; package org.apache.myfaces.shared_impl.webapp.webxml check this JAR: http://repo1.maven.org/maven2/org/apache/myfaces/shared/myfaces-shared-impl/3.0.0/myfaces-shared-impl-3.0.0-sources.jar (was used to build 1.2.0) -M On Dec 20, 2007 11:14 AM, Carl Howarth [EMAIL PROTECTED] wrote: Hello there, I am trying to debug a myfaces 1.2.0 app but the sources download does not contain the WebXml class. I have checked both the maven repository and the main download site. The snapshot version for 1.2.1 also does not have this class included. Thanks, Carl -- View this message in context: http://www.nabble.com/Class-missing-from-Source-tp14433768p14433768.html Sent from the My Faces - Dev mailing list archive at Nabble.com. -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: Class missing from Source
On Dec 20, 2007 12:36 PM, Simon Kitching [EMAIL PROTECTED] wrote: But the source download for myfaces really should include the shared classes too. The .class files are in the myfaces impl jar, so the (repackaged) sources should be in the corresponding source jarfile. currently the sources are just sources for the module; an enhancement would be nice, sure. I would *really* like to see this fixed for the next release. It's a pain to be stepping through code and then not be able to find the source without adding a source jar that is not part of the declared dependencies at all.. yes, I agree; feel free to add this :-) Regards, Simon Matthias Wessendorf [EMAIL PROTECTED] schrieb: Hi, that is part of the shared; package org.apache.myfaces.shared_impl.webapp.webxml check this JAR: http://repo1.maven.org/maven2/org/apache/myfaces/shared/myfaces-shared-impl/3.0.0/myfaces-shared-impl-3.0.0-sources.jar (was used to build 1.2.0) -M On Dec 20, 2007 11:14 AM, Carl Howarth [EMAIL PROTECTED] wrote: Hello there, I am trying to debug a myfaces 1.2.0 app but the sources download does not contain the WebXml class. I have checked both the maven repository and the main download site. The snapshot version for 1.2.1 also does not have this class included. Thanks, Carl -- View this message in context: http://www.nabble.com/Class-missing-from-Source-tp14433768p14433768.html Sent from the My Faces - Dev mailing list archive at Nabble.com. -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: Class missing from Source
Matthias Wessendorf [EMAIL PROTECTED] schrieb: On Dec 20, 2007 12:36 PM, Simon Kitching [EMAIL PROTECTED] wrote: But the source download for myfaces really should include the shared classes too. The .class files are in the myfaces impl jar, so the (repackaged) sources should be in the corresponding source jarfile. currently the sources are just sources for the module; an enhancement would be nice, sure. Well, it depends on whether you count the stuff pulled in from shared-impl as part of the sources of the module or not. I think there is a good reason for saying they are. Ideally, the build process would not be the shared project generating impl, tomahawk, etc versions but instead other projects pulling a standard shared-sources.jar file, then doing the package-munging themselves, effectively importing the source. I would *really* like to see this fixed for the next release. It's a pain to be stepping through code and then not be able to find the source without adding a source jar that is not part of the declared dependencies at all.. yes, I agree; feel free to add this :-) The shared code is unpacked into the target dir as part of the build process. So it is just a matter of including that dir when building the source jar. Should be 1 or 2 lines in the pom. I won't be able to get this done in the next week; work xmas party tonight then leave on holiday for about a week at 6am tomorrow morning. It would be great if someone else could have a go at this. Regards, Simon
Re: Class missing from Source
On Dec 20, 2007 12:45 PM, Simon Kitching [EMAIL PROTECTED] wrote: Matthias Wessendorf [EMAIL PROTECTED] schrieb: On Dec 20, 2007 12:36 PM, Simon Kitching [EMAIL PROTECTED] wrote: But the source download for myfaces really should include the shared classes too. The .class files are in the myfaces impl jar, so the (repackaged) sources should be in the corresponding source jarfile. currently the sources are just sources for the module; an enhancement would be nice, sure. Well, it depends on whether you count the stuff pulled in from shared-impl as part of the sources of the module or not. I think there is a good reason for saying they are. Ideally, the build process would not be the shared project generating impl, tomahawk, etc versions but instead other projects pulling a standard shared-sources.jar file, then doing the package-munging themselves, effectively importing the source. +1 that sounds indeed better; I would *really* like to see this fixed for the next release. It's a pain to be stepping through code and then not be able to find the source without adding a source jar that is not part of the declared dependencies at all.. yes, I agree; feel free to add this :-) The shared code is unpacked into the target dir as part of the build process. So it is just a matter of including that dir when building the source jar. Should be 1 or 2 lines in the pom. I won't be able to get this done in the next week; work xmas party tonight then leave on holiday for about a week at 6am tomorrow morning. It would be great if someone else could have a go at this. not today; let me try tomorrow. -M Regards, Simon -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: Class missing from Source
As Leonardo is working on the release anyways, he might take a stab? regards, Martin On 12/20/07, Simon Kitching [EMAIL PROTECTED] wrote: Matthias Wessendorf [EMAIL PROTECTED] schrieb: On Dec 20, 2007 12:36 PM, Simon Kitching [EMAIL PROTECTED] wrote: But the source download for myfaces really should include the shared classes too. The .class files are in the myfaces impl jar, so the (repackaged) sources should be in the corresponding source jarfile. currently the sources are just sources for the module; an enhancement would be nice, sure. Well, it depends on whether you count the stuff pulled in from shared-impl as part of the sources of the module or not. I think there is a good reason for saying they are. Ideally, the build process would not be the shared project generating impl, tomahawk, etc versions but instead other projects pulling a standard shared-sources.jar file, then doing the package-munging themselves, effectively importing the source. I would *really* like to see this fixed for the next release. It's a pain to be stepping through code and then not be able to find the source without adding a source jar that is not part of the declared dependencies at all.. yes, I agree; feel free to add this :-) The shared code is unpacked into the target dir as part of the build process. So it is just a matter of including that dir when building the source jar. Should be 1 or 2 lines in the pom. I won't be able to get this done in the next week; work xmas party tonight then leave on holiday for about a week at 6am tomorrow morning. It would be great if someone else could have a go at this. Regards, Simon -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Re: Class missing from Source
I still think the ideal way would be to do the repackaging on release only - for the normal build process, we could leave it as is. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 12:45 PM, Simon Kitching [EMAIL PROTECTED] wrote: Matthias Wessendorf [EMAIL PROTECTED] schrieb: On Dec 20, 2007 12:36 PM, Simon Kitching [EMAIL PROTECTED] wrote: But the source download for myfaces really should include the shared classes too. The .class files are in the myfaces impl jar, so the (repackaged) sources should be in the corresponding source jarfile. currently the sources are just sources for the module; an enhancement would be nice, sure. Well, it depends on whether you count the stuff pulled in from shared-impl as part of the sources of the module or not. I think there is a good reason for saying they are. Ideally, the build process would not be the shared project generating impl, tomahawk, etc versions but instead other projects pulling a standard shared-sources.jar file, then doing the package-munging themselves, effectively importing the source. +1 that sounds indeed better; I would *really* like to see this fixed for the next release. It's a pain to be stepping through code and then not be able to find the source without adding a source jar that is not part of the declared dependencies at all.. yes, I agree; feel free to add this :-) The shared code is unpacked into the target dir as part of the build process. So it is just a matter of including that dir when building the source jar. Should be 1 or 2 lines in the pom. I won't be able to get this done in the next week; work xmas party tonight then leave on holiday for about a week at 6am tomorrow morning. It would be great if someone else could have a go at this. not today; let me try tomorrow. -M Regards, Simon -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Re: Class missing from Source
hrm, I think the regular build should be similar to the release, But for the release do things like signing etc as an extra step; -M On Dec 20, 2007 12:50 PM, Martin Marinschek [EMAIL PROTECTED] wrote: I still think the ideal way would be to do the repackaging on release only - for the normal build process, we could leave it as is. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 12:45 PM, Simon Kitching [EMAIL PROTECTED] wrote: Matthias Wessendorf [EMAIL PROTECTED] schrieb: On Dec 20, 2007 12:36 PM, Simon Kitching [EMAIL PROTECTED] wrote: But the source download for myfaces really should include the shared classes too. The .class files are in the myfaces impl jar, so the (repackaged) sources should be in the corresponding source jarfile. currently the sources are just sources for the module; an enhancement would be nice, sure. Well, it depends on whether you count the stuff pulled in from shared-impl as part of the sources of the module or not. I think there is a good reason for saying they are. Ideally, the build process would not be the shared project generating impl, tomahawk, etc versions but instead other projects pulling a standard shared-sources.jar file, then doing the package-munging themselves, effectively importing the source. +1 that sounds indeed better; I would *really* like to see this fixed for the next release. It's a pain to be stepping through code and then not be able to find the source without adding a source jar that is not part of the declared dependencies at all.. yes, I agree; feel free to add this :-) The shared code is unpacked into the target dir as part of the build process. So it is just a matter of including that dir when building the source jar. Should be 1 or 2 lines in the pom. I won't be able to get this done in the next week; work xmas party tonight then leave on holiday for about a week at 6am tomorrow morning. It would be great if someone else could have a go at this. not today; let me try tomorrow. -M Regards, Simon -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
[Build-Tools] Archetypes ?
Hi, IMO the archetypes could be moved into the build-tools as well. /myfaces-build-tools -/maven2-plugins -/maven2-archetype -here they are -M -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
[jira] Resolved: (TOMAHAWK-1171) forceId doesn't work when page rendered after an 'immediate' action
[ https://issues.apache.org/jira/browse/TOMAHAWK-1171?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Eric Price resolved TOMAHAWK-1171. -- Resolution: Fixed I realized the component referenced by the binding attribute on the t:inputText component was a javax.faces.component.html.HtmlInputText. When I changed the object to an org.apache.myfaces.component.html.ext.HtmlInputText component, the component ID assignment worked as expected. forceId doesn't work when page rendered after an 'immediate' action --- Key: TOMAHAWK-1171 URL: https://issues.apache.org/jira/browse/TOMAHAWK-1171 Project: MyFaces Tomahawk Issue Type: Bug Components: ForceId Affects Versions: 1.1.3 Environment: java 1.4.2_13, myfaces 1.1.4, tomahawk 1.1.3, windows XP Reporter: Eric Price Priority: Minor I have a jsp with the following components: t:outputText value=Processed Date/Time:/ t:panelGroup rendered=#{reportBean.dateCleared} t:inputText styleClass=deactiveText title=-MM-dd size=10 maxlength=10 readonly=true/ t:inputText styleClass=deactiveText title=HH:mm:ss size=8 maxlength=8 readonly=true/ /t:panelGroup t:panelGroup rendered=#{!reportBean.dateCleared} t:inputText binding=#{reportBean.startDateInput} size=10 maxlength=10 immediate=true value=#{reportBean.startDate} title=-MM-dd forceId=true id=processedDateStartDateInput t:convertDateTime type=date pattern=-MM-dd/ /t:inputText t:inputText binding=#{reportBean.startTimeInput} size=8 maxlength=8 immediate=true value=#{reportBean.startTime} title=HH:mm:ss forceId=true id=processedDateStartTimeInput t:convertDateTime type=time pattern=HH:mm:ss/ /t:inputText /t:panelGroup When the page is initially displayed the value of reportBean.dateCleared is false, so the first two inputText components are displayed. The source of the page looks like: td class=fieldLabelColumnProcessed Date/Time:/td td class=defaultColumn input id=mainform:_idJsp393 name=mainform:_idJsp393 type=text value= maxlength=10 readonly=readonly size=10 title=-MM-dd class=deactiveText/ input id=mainform:_idJsp396 name=mainform:_idJsp396 type=text value= maxlength=8 readonly=readonly size=8 title=HH:mm:ss class=deactiveText/ /td... which shows what I would expect for the componet IDs. The page also contains a jsCookMenu with an action method that causes the processed date fields to be set according to the user selection. The action method is executed with immediate=true set on the menu component. Once the date is set, the reportBean.dateCleared is set to false and the current page is re-rendered. The selected date is correctly displayed in the date fields, but the component IDs are not set correctly. After setting the date and the page re-rendered, the source looks like: td class=fieldLabelColumnProcessed Date/Time:/td td class=defaultColumn input id=mainform:processedDateStartDateInput name=mainform:processedDateStartDateInput type=text value=2007-12-15 maxlength=10 size=10 title=-MM-dd / input id=mainform:processedDateStartTimeInput name=mainform:processedDateStartTimeInput type=text value=23:59:59 maxlength=8 size=8 title=HH:mm:ss/ /td The component IDs have the specified value, but the form name is prepended. Once I navigate to the next page of the application, and then navigate back to this page (via an action method on a button), I see the following source: td class=fieldLabelColumnProcessed Date/Time:/td td class=defaultColumn input id=processedDateStartDateInput name=processedDateStartDateInput type=text value=2007-12-15 maxlength=10 size=10 title=-MM-dd / input id=processedDateStartTimeInput name=processedDateStartTimeInput type=text value=23:59:59 maxlength=8 size=8 title=HH:mm:ss/ /td Now the component IDs are what I expected, just the ID without the form name prepended. Any insight as to why the forceId doesn't appear to work when rendering the page after an (immediate) action, but it does work when rendering the page via navigation? Thanks. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
Re: [Trinidad] Maven Plugins
Did a relocate of the plugins to: https://svn.apache.org/repos/asf/myfaces/myfaces-build-tools/trunk/maven2-plugins/ the current trinidad-pluings are still in their trinidad-maven trunk; I renamed the plugins to be compliant to the naming-schema maven-blah-plugin means that blah is a plugin from the maven folks; they are now named: myfaces-whatever-plugin (like myfaces-jdev-plugin) Need to work on the groupIds, the java packages etc. Nice task for between the years :-) On Dec 20, 2007 11:28 AM, Matthias Wessendorf [EMAIL PROTECTED] wrote: ok... looks like myfaces-build-tools won the race. structure will be /myfaces-build-tools -/maven2-plugins -/all the current (trin) plugins -tbc -M On Dec 15, 2007 11:12 PM, Matthias Wessendorf [EMAIL PROTECTED] wrote: Hi, there is this maven plugins folder for the Trinidad Plugins (see [1]). The Trinidad inside the name may confuse people. The plugins are generally usable outside of Trinidad as well. The maven-faces-plugin for instance, is used in other projects, such as MyFaces Core, or MyFaces Commons. The idea is to move them into a different location. A name like myfaces developers toolkit, myfaces dev tools, or what ever you like ? The idea would be to keep it abstract like /myfaces-dev-tools -/maven2-plugins -tbc that would allow us to have maven3 or ant or what ever support as well. Greetings, Matze Manfred [1] https://svn.apache.org/repos/asf/myfaces/trinidad-maven/trunk/ -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: [Build-Tools] Archetypes ?
+1 yes, sure On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: Hi, IMO the archetypes could be moved into the build-tools as well. /myfaces-build-tools -/maven2-plugins -/maven2-archetype -here they are -M -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
[Build-Tools] GroupId
Hi, I like to use groupIdorg.apache.myfaces.build/groupId as the groupId And the java packages like: -org.apache.myfaces.build.wagon -org.apache.myfaces.build.faces -... WDYT ? -M -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: [Build-Tools] GroupId
yes, ok or org.apache.myfaces.buildtools ? --Manfred On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: Hi, I like to use groupIdorg.apache.myfaces.build/groupId as the groupId And the java packages like: -org.apache.myfaces.build.wagon -org.apache.myfaces.build.faces -... WDYT ? -M -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Re: [Build-Tools] GroupId
sounds good; On Dec 20, 2007 2:08 PM, Manfred Geiler [EMAIL PROTECTED] wrote: yes, ok or org.apache.myfaces.buildtools ? --Manfred On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: Hi, I like to use groupIdorg.apache.myfaces.build/groupId as the groupId And the java packages like: -org.apache.myfaces.build.wagon -org.apache.myfaces.build.faces -... WDYT ? -M -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: [Build-Tools] GroupId
Matthias Wessendorf [EMAIL PROTECTED] schrieb: Hi, I like to use groupIdorg.apache.myfaces.build/groupId as the groupId And the java packages like: -org.apache.myfaces.build.wagon -org.apache.myfaces.build.faces -... WDYT ? +1
Re: [Trinidad] Maven Plugins
Thanks for the build-tools stuff Matthias. That's a very nice tidyup. Having myfaces-core depend on stuff called trinidad was rather confusing... Matthias Wessendorf [EMAIL PROTECTED] schrieb: Did a relocate of the plugins to: https://svn.apache.org/repos/asf/myfaces/myfaces-build-tools/trunk/maven2-plugins/ the current trinidad-pluings are still in their trinidad-maven trunk; I renamed the plugins to be compliant to the naming-schema maven-blah-plugin means that blah is a plugin from the maven folks; they are now named: myfaces-whatever-plugin (like myfaces-jdev-plugin) Need to work on the groupIds, the java packages etc. Nice task for between the years :-) On Dec 20, 2007 11:28 AM, Matthias Wessendorf [EMAIL PROTECTED] wrote: ok... looks like myfaces-build-tools won the race. structure will be /myfaces-build-tools -/maven2-plugins -/all the current (trin) plugins -tbc -M On Dec 15, 2007 11:12 PM, Matthias Wessendorf [EMAIL PROTECTED] wrote: Hi, there is this maven plugins folder for the Trinidad Plugins (see [1]). The Trinidad inside the name may confuse people. The plugins are generally usable outside of Trinidad as well. The maven-faces-plugin for instance, is used in other projects, such as MyFaces Core, or MyFaces Commons. The idea is to move them into a different location. A name like myfaces developers toolkit, myfaces dev tools, or what ever you like ? The idea would be to keep it abstract like /myfaces-dev-tools -/maven2-plugins -tbc that would allow us to have maven3 or ant or what ever support as well. Greetings, Matze Manfred [1] https://svn.apache.org/repos/asf/myfaces/trinidad-maven/trunk/ -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: [Build-Tools] Archetypes ?
+1 On Dec 20, 2007 2:33 PM, Manfred Geiler [EMAIL PROTECTED] wrote: +1 yes, sure On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: Hi, IMO the archetypes could be moved into the build-tools as well. /myfaces-build-tools -/maven2-plugins -/maven2-archetype -here they are -M -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Re: [Build-Tools] GroupId
+1 On 20/12/2007, Manfred Geiler [EMAIL PROTECTED] wrote: yes, ok or org.apache.myfaces.buildtools ? --Manfred On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: Hi, I like to use groupIdorg.apache.myfaces.build/groupId as the groupId And the java packages like: -org.apache.myfaces.build.wagon -org.apache.myfaces.build.faces -... WDYT ? -M -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Re: [Build-Tools] Archetypes ?
+1 On 20/12/2007, Cagatay Civici [EMAIL PROTECTED] wrote: +1 On Dec 20, 2007 2:33 PM, Manfred Geiler [EMAIL PROTECTED] wrote: +1 yes, sure On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: Hi, IMO the archetypes could be moved into the build-tools as well. /myfaces-build-tools -/maven2-plugins -/maven2-archetype -here they are -M -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Re: svn commit: r605927 - /myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java
On Dec 20, 2007 2:45 PM, [EMAIL PROTECTED] wrote: Author: skitching Date: Thu Dec 20 05:45:06 2007 New Revision: 605927 URL: http://svn.apache.org/viewvc?rev=605927view=rev Log: Remove accidental use of java 1.6 method String.isEmpty isn't the build catching this ? -M Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java URL: http://svn.apache.org/viewvc/myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java?rev=605927r1=605926r2=605927view=diff == --- myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java (original) +++ myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Thu Dec 20 05:45:06 2007 @@ -838,7 +838,7 @@ private static void formatPattern(DateFormatSymbols symbols, ParserContext context, String patternSub, boolean yearIsWeekYear, StringBuffer out) { -if ((patternSub == null) || (patternSub.isEmpty())) +if ((patternSub == null) || (patternSub.length() == 0)) { return; } @@ -1137,7 +1137,7 @@ public Date parse(String dateStr) { -if ((dateStr==null) || dateStr.isEmpty()) +if ((dateStr==null) || (dateStr.length() == 0)) return null; ParserContext context = parseOps(symbols, yearIsWeekYear, firstDayOfWeek, ops, dateStr); -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: svn commit: r605927 - /myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java
No - if you are building with 1.6, it won't. Only the language level is checked. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 2:45 PM, [EMAIL PROTECTED] wrote: Author: skitching Date: Thu Dec 20 05:45:06 2007 New Revision: 605927 URL: http://svn.apache.org/viewvc?rev=605927view=rev Log: Remove accidental use of java 1.6 method String.isEmpty isn't the build catching this ? -M Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java URL: http://svn.apache.org/viewvc/myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java?rev=605927r1=605926r2=605927view=diff == --- myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java (original) +++ myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Thu Dec 20 05:45:06 2007 @@ -838,7 +838,7 @@ private static void formatPattern(DateFormatSymbols symbols, ParserContext context, String patternSub, boolean yearIsWeekYear, StringBuffer out) { -if ((patternSub == null) || (patternSub.isEmpty())) +if ((patternSub == null) || (patternSub.length() == 0)) { return; } @@ -1137,7 +1137,7 @@ public Date parse(String dateStr) { -if ((dateStr==null) || dateStr.isEmpty()) +if ((dateStr==null) || (dateStr.length() == 0)) return null; ParserContext context = parseOps(symbols, yearIsWeekYear, firstDayOfWeek, ops, dateStr); -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Re: [Build-Tools] Archetypes ?
+1, regards, Martin On 12/20/07, Bruno Aranda [EMAIL PROTECTED] wrote: +1 On 20/12/2007, Cagatay Civici [EMAIL PROTECTED] wrote: +1 On Dec 20, 2007 2:33 PM, Manfred Geiler [EMAIL PROTECTED] wrote: +1 yes, sure On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: Hi, IMO the archetypes could be moved into the build-tools as well. /myfaces-build-tools -/maven2-plugins -/maven2-archetype -here they are -M -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Re: svn commit: r605927 - /myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java
I thought the compile was configured for source/target not 1.5 || 1.6 -M On Dec 20, 2007 3:18 PM, Martin Marinschek [EMAIL PROTECTED] wrote: No - if you are building with 1.6, it won't. Only the language level is checked. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 2:45 PM, [EMAIL PROTECTED] wrote: Author: skitching Date: Thu Dec 20 05:45:06 2007 New Revision: 605927 URL: http://svn.apache.org/viewvc?rev=605927view=rev Log: Remove accidental use of java 1.6 method String.isEmpty isn't the build catching this ? -M Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java URL: http://svn.apache.org/viewvc/myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java?rev=605927r1=605926r2=605927view=diff == --- myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java (original) +++ myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Thu Dec 20 05:45:06 2007 @@ -838,7 +838,7 @@ private static void formatPattern(DateFormatSymbols symbols, ParserContext context, String patternSub, boolean yearIsWeekYear, StringBuffer out) { -if ((patternSub == null) || (patternSub.isEmpty())) +if ((patternSub == null) || (patternSub.length() == 0)) { return; } @@ -1137,7 +1137,7 @@ public Date parse(String dateStr) { -if ((dateStr==null) || dateStr.isEmpty()) +if ((dateStr==null) || (dateStr.length() == 0)) return null; ParserContext context = parseOps(symbols, yearIsWeekYear, firstDayOfWeek, ops, dateStr); -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: svn commit: r605927 - /myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java
On Dec 20, 2007 3:19 PM, Matthias Wessendorf [EMAIL PROTECTED] wrote: I thought the compile was configured for source/target not 1.5 || 1.6 (the compiler plugin) -M On Dec 20, 2007 3:18 PM, Martin Marinschek [EMAIL PROTECTED] wrote: No - if you are building with 1.6, it won't. Only the language level is checked. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 2:45 PM, [EMAIL PROTECTED] wrote: Author: skitching Date: Thu Dec 20 05:45:06 2007 New Revision: 605927 URL: http://svn.apache.org/viewvc?rev=605927view=rev Log: Remove accidental use of java 1.6 method String.isEmpty isn't the build catching this ? -M Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java URL: http://svn.apache.org/viewvc/myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java?rev=605927r1=605926r2=605927view=diff == --- myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java (original) +++ myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Thu Dec 20 05:45:06 2007 @@ -838,7 +838,7 @@ private static void formatPattern(DateFormatSymbols symbols, ParserContext context, String patternSub, boolean yearIsWeekYear, StringBuffer out) { -if ((patternSub == null) || (patternSub.isEmpty())) +if ((patternSub == null) || (patternSub.length() == 0)) { return; } @@ -1137,7 +1137,7 @@ public Date parse(String dateStr) { -if ((dateStr==null) || dateStr.isEmpty()) +if ((dateStr==null) || (dateStr.length() == 0)) return null; ParserContext context = parseOps(symbols, yearIsWeekYear, firstDayOfWeek, ops, dateStr); -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: svn commit: r605927 - /myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java
Yes, so 1.5/1.6 syntax will correctly be found and disallowed. But API changes are not checked for, so a newly added method will not be found. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 3:19 PM, Matthias Wessendorf [EMAIL PROTECTED] wrote: I thought the compile was configured for source/target not 1.5 || 1.6 (the compiler plugin) -M On Dec 20, 2007 3:18 PM, Martin Marinschek [EMAIL PROTECTED] wrote: No - if you are building with 1.6, it won't. Only the language level is checked. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 2:45 PM, [EMAIL PROTECTED] wrote: Author: skitching Date: Thu Dec 20 05:45:06 2007 New Revision: 605927 URL: http://svn.apache.org/viewvc?rev=605927view=rev Log: Remove accidental use of java 1.6 method String.isEmpty isn't the build catching this ? -M Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java URL: http://svn.apache.org/viewvc/myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java?rev=605927r1=605926r2=605927view=diff == --- myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java (original) +++ myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Thu Dec 20 05:45:06 2007 @@ -838,7 +838,7 @@ private static void formatPattern(DateFormatSymbols symbols, ParserContext context, String patternSub, boolean yearIsWeekYear, StringBuffer out) { -if ((patternSub == null) || (patternSub.isEmpty())) +if ((patternSub == null) || (patternSub.length() == 0)) { return; } @@ -1137,7 +1137,7 @@ public Date parse(String dateStr) { -if ((dateStr==null) || dateStr.isEmpty()) +if ((dateStr==null) || (dateStr.length() == 0)) return null; ParserContext context = parseOps(symbols, yearIsWeekYear, firstDayOfWeek, ops, dateStr); -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Re: svn commit: r605927 - /myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java
So, you say that a configuration like: plugin groupIdorg.apache.maven.plugins/groupId artifactIdmaven-compiler-plugin/artifactId inheritedtrue/inherited configuration source1.5/source target1.5/target /configuration /plugin does allow you to use JDK6 methods? That would be a bad bug, IMO. -M On Dec 20, 2007 3:22 PM, Martin Marinschek [EMAIL PROTECTED] wrote: Yes, so 1.5/1.6 syntax will correctly be found and disallowed. But API changes are not checked for, so a newly added method will not be found. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 3:19 PM, Matthias Wessendorf [EMAIL PROTECTED] wrote: I thought the compile was configured for source/target not 1.5 || 1.6 (the compiler plugin) -M On Dec 20, 2007 3:18 PM, Martin Marinschek [EMAIL PROTECTED] wrote: No - if you are building with 1.6, it won't. Only the language level is checked. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 2:45 PM, [EMAIL PROTECTED] wrote: Author: skitching Date: Thu Dec 20 05:45:06 2007 New Revision: 605927 URL: http://svn.apache.org/viewvc?rev=605927view=rev Log: Remove accidental use of java 1.6 method String.isEmpty isn't the build catching this ? -M Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java URL: http://svn.apache.org/viewvc/myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java?rev=605927r1=605926r2=605927view=diff == --- myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java (original) +++ myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Thu Dec 20 05:45:06 2007 @@ -838,7 +838,7 @@ private static void formatPattern(DateFormatSymbols symbols, ParserContext context, String patternSub, boolean yearIsWeekYear, StringBuffer out) { -if ((patternSub == null) || (patternSub.isEmpty())) +if ((patternSub == null) || (patternSub.length() == 0)) { return; } @@ -1137,7 +1137,7 @@ public Date parse(String dateStr) { -if ((dateStr==null) || dateStr.isEmpty()) +if ((dateStr==null) || (dateStr.length() == 0)) return null; ParserContext context = parseOps(symbols, yearIsWeekYear, firstDayOfWeek, ops, dateStr); -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: svn commit: r605927 - /myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java
yes, that is what I'm saying. Saying that this is a bug is maybe pushing it too hard... the attributes just translate into a compiler language level - that does not mean that someone automagically knows all API changes from 1.4 to 1.6 and automatically applies them. The only chance for you to catch all these problems is to actually compile with 1.4. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: So, you say that a configuration like: plugin groupIdorg.apache.maven.plugins/groupId artifactIdmaven-compiler-plugin/artifactId inheritedtrue/inherited configuration source1.5/source target1.5/target /configuration /plugin does allow you to use JDK6 methods? That would be a bad bug, IMO. -M On Dec 20, 2007 3:22 PM, Martin Marinschek [EMAIL PROTECTED] wrote: Yes, so 1.5/1.6 syntax will correctly be found and disallowed. But API changes are not checked for, so a newly added method will not be found. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 3:19 PM, Matthias Wessendorf [EMAIL PROTECTED] wrote: I thought the compile was configured for source/target not 1.5 || 1.6 (the compiler plugin) -M On Dec 20, 2007 3:18 PM, Martin Marinschek [EMAIL PROTECTED] wrote: No - if you are building with 1.6, it won't. Only the language level is checked. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 2:45 PM, [EMAIL PROTECTED] wrote: Author: skitching Date: Thu Dec 20 05:45:06 2007 New Revision: 605927 URL: http://svn.apache.org/viewvc?rev=605927view=rev Log: Remove accidental use of java 1.6 method String.isEmpty isn't the build catching this ? -M Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java URL: http://svn.apache.org/viewvc/myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java?rev=605927r1=605926r2=605927view=diff == --- myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java (original) +++ myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Thu Dec 20 05:45:06 2007 @@ -838,7 +838,7 @@ private static void formatPattern(DateFormatSymbols symbols, ParserContext context, String patternSub, boolean yearIsWeekYear, StringBuffer out) { -if ((patternSub == null) || (patternSub.isEmpty())) +if ((patternSub == null) || (patternSub.length() == 0)) { return; } @@ -1137,7 +1137,7 @@ public Date parse(String dateStr) { -if ((dateStr==null) || dateStr.isEmpty()) +if ((dateStr==null) || (dateStr.length() == 0)) return null; ParserContext context = parseOps(symbols, yearIsWeekYear, firstDayOfWeek, ops, dateStr); -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Re: Server state saving and multiple frames
Yes, well, with Trinidad the suggested changes won't work - but Trinidad has some internal means of doing something similar to what is proposed here, so first we should check what Trinidad is doing. Can you check/send a mail with Trinidad in the subject line? regards, Martin On 12/20/07, Cristi Toth [EMAIL PROTECTED] wrote: What Thomas said about Trinidad and any other component library with their own state manager is very true. It would be nice to have the SerializedViewCollection (the view pool from JspStateManagerImpl) to be delegated to handle the actual storing of the views. But that still implies modifying those state managers On Dec 20, 2007 11:01 AM, Nicu Mercioiu [EMAIL PROTECTED] wrote: My solution doesn't work only with commandLinks but works as well with outputLink or JavaScript by adding a GET parameter like in the following example: h:outputLink value=multipleFrames.jsf target=_blank f:param name=target value=_blank/ /h:outputLink It also works for any submitted form with a target attribute, for example a form submitted by commandButton. Regards, Nicu On Dec 20, 2007 10:41 AM, Simon Kitching [EMAIL PROTECTED] wrote: Orchestra provides a tag that can be explicitly wrapped around links (eg h:outputLink). It then encodes a window id into the url params. So it also partially addresses the second part of the problem (identifying windows/frames). It covers more cases than Nicu's suggestion, but it does require modifications to pages. Orchestra then uses this window-id to provide a new set of conversation-scoped beans per window. But nothing in orchestra addresses the problem of the view-state pool, so unless I misunderstand something this Orchestra feature is not really usable; the backing beans will work right but when the viewstate gets screwed up after 4 page-views that is no consolation. I think myfaces core needs to provide a per-window viewstate cache. It can then just allow something else to figure out how to identify the windows, allowing the Orchestra approach or Nicu's approach, or whatever people come up with. Yes, it would be interesting to hear how Trinidad dialogs handle this.. Regards, Simon Martin Marinschek [EMAIL PROTECTED] schrieb: Hi Nicu, we should include Mario in this discussion - he implemented a solution for this in Orchestra. Also, how about Trinidad, in Trinidad there is dialog handling as well, how is this done? regards, Martin On 12/19/07, simon [EMAIL PROTECTED] wrote: Hi Nicu, I haven't got time to look at this closely, but IMO siomething like this is definitely needed in MyFaces. A user with multiple windows is certainly going to have trouble at the moment. I think a modification to the view pool to include a window id (or frame id) is definitely a good idea. The second part of the problem still remains: how to associate a different id with each window/frame. Checking CommandLink components for a target attribute is clever; it doesn't solve all the cases but does solve some. Regards, Simon On Tue, 2007-12-18 at 19:07 +0200, Nicu Mercioiu wrote: Hi, There is a problem in JSF when more than one window are opened in an application. There are only a maximum number of NUMBER_OF_VIEWS_IN_SESSION view states saved at one moment (when server side state saving is enabled). If you have 2 windows opened and you navigate on one of them for NUMBER_OF_VIEWS_IN_SESSION times, you will lose the other window's state. I've been facing this problem while developing a project so I've implemented a solution for it. The solution is having a number of view states saved for each opened window at one moment. For determining when a new window (frame) is opened, the target of the submitting component (or its enclosing form) is used. This is obtained in the HtmlLinkRendererBase's and HtmlButtonRendererBase's decode methods and it is set in the RequestMap. Using the submitted target, the JspStateManagerImpl figures out whether a new frame was opened. If so, a new frame id is generated. In the renderResponse phase, the frameId is encoded in the javax.faces.ViewState field and is used along with the viewId to save the state in a SerializedViewCollection. In the restore view phase the frameId is decoded from the javax.faces.ViewState field and is used along with the viewId to restore the corresponding state from the SerializedViewCollection. In SerializedViewCollection instead of a list of recently used views, now a list is kept for each frameId. The following context params are
Re: svn commit: r605927 - /myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java
I'd be fine with Java5 instead of 1.4 -M On Dec 20, 2007 3:37 PM, Martin Marinschek [EMAIL PROTECTED] wrote: yes, that is what I'm saying. Saying that this is a bug is maybe pushing it too hard... the attributes just translate into a compiler language level - that does not mean that someone automagically knows all API changes from 1.4 to 1.6 and automatically applies them. The only chance for you to catch all these problems is to actually compile with 1.4. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: So, you say that a configuration like: plugin groupIdorg.apache.maven.plugins/groupId artifactIdmaven-compiler-plugin/artifactId inheritedtrue/inherited configuration source1.5/source target1.5/target /configuration /plugin does allow you to use JDK6 methods? That would be a bad bug, IMO. -M On Dec 20, 2007 3:22 PM, Martin Marinschek [EMAIL PROTECTED] wrote: Yes, so 1.5/1.6 syntax will correctly be found and disallowed. But API changes are not checked for, so a newly added method will not be found. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 3:19 PM, Matthias Wessendorf [EMAIL PROTECTED] wrote: I thought the compile was configured for source/target not 1.5 || 1.6 (the compiler plugin) -M On Dec 20, 2007 3:18 PM, Martin Marinschek [EMAIL PROTECTED] wrote: No - if you are building with 1.6, it won't. Only the language level is checked. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 2:45 PM, [EMAIL PROTECTED] wrote: Author: skitching Date: Thu Dec 20 05:45:06 2007 New Revision: 605927 URL: http://svn.apache.org/viewvc?rev=605927view=rev Log: Remove accidental use of java 1.6 method String.isEmpty isn't the build catching this ? -M Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java URL: http://svn.apache.org/viewvc/myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java?rev=605927r1=605926r2=605927view=diff == --- myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java (original) +++ myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Thu Dec 20 05:45:06 2007 @@ -838,7 +838,7 @@ private static void formatPattern(DateFormatSymbols symbols, ParserContext context, String patternSub, boolean yearIsWeekYear, StringBuffer out) { -if ((patternSub == null) || (patternSub.isEmpty())) +if ((patternSub == null) || (patternSub.length() == 0)) { return; } @@ -1137,7 +1137,7 @@ public Date parse(String dateStr) { -if ((dateStr==null) || dateStr.isEmpty()) +if ((dateStr==null) || (dateStr.length() == 0)) return null; ParserContext context = parseOps(symbols, yearIsWeekYear, firstDayOfWeek, ops, dateStr); -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions:
Re: [Build-Tools] Archetypes ?
relocated to: https://svn.apache.org/repos/asf/myfaces/myfaces-build-tools/trunk/maven2-archetypes/ renamed them to fit into the proper naming schema. will update the wiki as well. (name, location, etc) -M On Dec 20, 2007 3:18 PM, Martin Marinschek [EMAIL PROTECTED] wrote: +1, regards, Martin On 12/20/07, Bruno Aranda [EMAIL PROTECTED] wrote: +1 On 20/12/2007, Cagatay Civici [EMAIL PROTECTED] wrote: +1 On Dec 20, 2007 2:33 PM, Manfred Geiler [EMAIL PROTECTED] wrote: +1 yes, sure On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: Hi, IMO the archetypes could be moved into the build-tools as well. /myfaces-build-tools -/maven2-plugins -/maven2-archetype -here they are -M -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Java version compatibility (was Re: svn commit: r605927 - /myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java)
This problem happens no matter what java versions are being talked about. When the aim is to support java N, but the latest is java X then it is difficult to prevent accidents. Compiler property source prevents the use of *language features* that are not supported in the specified version Compiler property target ensures that the generated bytecode is readable by the specified version. But nothing checks that you are using only APIs provided by the runtime library distributed by that version. Currently the only way to do that is to compile against the runtime lib distributed with that version (which basically means *using* that version to compile). The CLIRR project does binary-compatibility testing on jars. I have always suspected that it would only need a minor enhancement to generate an API summary for the rt.jar of version N of java, and then to be able to automatically check that an app only makes calls to the APIs defined by that summary file. Just need that old 40 hour day in order to get time to do it :-) Cheers, Simon Matthias Wessendorf [EMAIL PROTECTED] schrieb: I'd be fine with Java5 instead of 1.4 -M On Dec 20, 2007 3:37 PM, Martin Marinschek [EMAIL PROTECTED] wrote: yes, that is what I'm saying. Saying that this is a bug is maybe pushing it too hard... the attributes just translate into a compiler language level - that does not mean that someone automagically knows all API changes from 1.4 to 1.6 and automatically applies them. The only chance for you to catch all these problems is to actually compile with 1.4. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: So, you say that a configuration like: plugin groupIdorg.apache.maven.plugins/groupId artifactIdmaven-compiler-plugin/artifactId inheritedtrue/inherited configuration source1.5/source target1.5/target /configuration /plugin does allow you to use JDK6 methods? That would be a bad bug, IMO. -M On Dec 20, 2007 3:22 PM, Martin Marinschek [EMAIL PROTECTED] wrote: Yes, so 1.5/1.6 syntax will correctly be found and disallowed. But API changes are not checked for, so a newly added method will not be found. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 3:19 PM, Matthias Wessendorf [EMAIL PROTECTED] wrote: I thought the compile was configured for source/target not 1.5 || 1.6 (the compiler plugin) -M On Dec 20, 2007 3:18 PM, Martin Marinschek [EMAIL PROTECTED] wrote: No - if you are building with 1.6, it won't. Only the language level is checked. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 2:45 PM, [EMAIL PROTECTED] wrote: Author: skitching Date: Thu Dec 20 05:45:06 2007 New Revision: 605927 URL: http://svn.apache.org/viewvc?rev=605927view=rev Log: Remove accidental use of java 1.6 method String.isEmpty isn't the build catching this ? -M Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java URL: http://svn.apache.org/viewvc/myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java?rev=605927r1=605926r2=605927view=diff == --- myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java (original) +++ myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Thu Dec 20 05:45:06 2007 @@ -838,7 +838,7 @@ private static void formatPattern(DateFormatSymbols symbols, ParserContext context, String patternSub, boolean yearIsWeekYear, StringBuffer out) { -if ((patternSub == null) || (patternSub.isEmpty())) +if ((patternSub == null) || (patternSub.length() == 0)) { return; } @@ -1137,7 +1137,7 @@ public Date parse(String dateStr) { -if ((dateStr==null) || dateStr.isEmpty()) +if ((dateStr==null) || (dateStr.length() == 0)) return null; ParserContext context = parseOps(symbols,
Re: [Build-Tools] GroupId
+1 Simon Kitching schrieb: Matthias Wessendorf [EMAIL PROTECTED] schrieb: Hi, I like to use groupIdorg.apache.myfaces.build/groupId as the groupId And the java packages like: -org.apache.myfaces.build.wagon -org.apache.myfaces.build.faces -... WDYT ? +1
[continuum] used jdk
I saw in the logs, that the zone uses JDK6. Is there a special reason for that ? -M -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
[jira] Commented: (TRINIDAD-65) The JS e.getFacesMessage method does not work when trinidad was compiled with Java 6 (JSLocaleElementsGenerator does not work with Java6)
[ https://issues.apache.org/jira/browse/TRINIDAD-65?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12553711 ] Matthias Weßendorf commented on TRINIDAD-65: I just used the Trinidad 1.0.5-SNAPSHOT from MyFaces's continuum server. I tried the client validator/converter examples. They are working; The continuum server is running on JDK 1.6 (if the log's aren't lying). Can you guys try @home and check as well ? -Matthias The JS e.getFacesMessage method does not work when trinidad was compiled with Java 6 (JSLocaleElementsGenerator does not work with Java6) - Key: TRINIDAD-65 URL: https://issues.apache.org/jira/browse/TRINIDAD-65 Project: MyFaces Trinidad Issue Type: Bug Components: Build Affects Versions: 1.0.0-incubating-plugins, 1.0.0-incubating-core, 1.0.1-plugins Environment: Windows XP, Java 6 Reporter: David Übelacker Assignee: Matthias Weßendorf Attachments: JSLocaleElementsGenerator.patch, LocaleDataResolver.java If trinidad was compiled with Java 6, you get the JS error e.getFacesMessage is no Function. The reason for this problem is, that the JSLocaleElementsGenerator of the maven-i18n-plugin does not work with Java 6. The JSLocaleElementsGenerator uses some java resource bundles to generate the LocaleElements files. These resource bundles are part of the rt.jar file until java 5. With java 6 these bundles doesn't exist anymore, are hard coded and can be accessed through sun.util.resources.LocaleData (not on the public API). -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
Re: Java version compatibility (was Re: svn commit: r605927 - /myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java)
Can't you do something like this to force the rt.jar to be 1.4? http://java.sun.com/j2se/1.5.0/docs/tooldocs/windows/javac.html % javac -target 1.4 -bootclasspath jdk1.4.2/lib/classes.zip -extdirs OldCode.java On Dec 20, 2007 10:11 AM, Simon Kitching [EMAIL PROTECTED] wrote: This problem happens no matter what java versions are being talked about. When the aim is to support java N, but the latest is java X then it is difficult to prevent accidents. Compiler property source prevents the use of *language features* that are not supported in the specified version Compiler property target ensures that the generated bytecode is readable by the specified version. But nothing checks that you are using only APIs provided by the runtime library distributed by that version. Currently the only way to do that is to compile against the runtime lib distributed with that version (which basically means *using* that version to compile). The CLIRR project does binary-compatibility testing on jars. I have always suspected that it would only need a minor enhancement to generate an API summary for the rt.jar of version N of java, and then to be able to automatically check that an app only makes calls to the APIs defined by that summary file. Just need that old 40 hour day in order to get time to do it :-) Cheers, Simon Matthias Wessendorf [EMAIL PROTECTED] schrieb: I'd be fine with Java5 instead of 1.4 -M On Dec 20, 2007 3:37 PM, Martin Marinschek [EMAIL PROTECTED] wrote: yes, that is what I'm saying. Saying that this is a bug is maybe pushing it too hard... the attributes just translate into a compiler language level - that does not mean that someone automagically knows all API changes from 1.4 to 1.6 and automatically applies them. The only chance for you to catch all these problems is to actually compile with 1.4. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: So, you say that a configuration like: plugin groupIdorg.apache.maven.plugins/groupId artifactIdmaven-compiler-plugin/artifactId inheritedtrue/inherited configuration source1.5/source target1.5/target /configuration /plugin does allow you to use JDK6 methods? That would be a bad bug, IMO. -M On Dec 20, 2007 3:22 PM, Martin Marinschek [EMAIL PROTECTED] wrote: Yes, so 1.5/1.6 syntax will correctly be found and disallowed. But API changes are not checked for, so a newly added method will not be found. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 3:19 PM, Matthias Wessendorf [EMAIL PROTECTED] wrote: I thought the compile was configured for source/target not 1.5 || 1.6 (the compiler plugin) -M On Dec 20, 2007 3:18 PM, Martin Marinschek [EMAIL PROTECTED] wrote: No - if you are building with 1.6, it won't. Only the language level is checked. regards, Martin On 12/20/07, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 2:45 PM, [EMAIL PROTECTED] wrote: Author: skitching Date: Thu Dec 20 05:45:06 2007 New Revision: 605927 URL: http://svn.apache.org/viewvc?rev=605927view=rev Log: Remove accidental use of java 1.6 method String.isEmpty isn't the build catching this ? -M Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Modified: myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java URL: http://svn.apache.org/viewvc/myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java?rev=605927r1=605926r2=605927view=diff == --- myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java (original) +++ myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java Thu Dec 20 05:45:06 2007 @@ -838,7 +838,7 @@ private static void formatPattern(DateFormatSymbols symbols, ParserContext context, String patternSub, boolean yearIsWeekYear, StringBuffer out) { -if ((patternSub == null) || (patternSub.isEmpty())) +if ((patternSub == null) || (patternSub.length()
Re: [Vote] release for Trinidad Plugins 1.2.6
+1 On Dec 20, 2007 12:59 AM, Bruno Aranda [EMAIL PROTECTED] wrote: +1 Thank you Leonardo! Bruno On 20/12/2007, Martin Marinschek [EMAIL PROTECTED] wrote: +1 regards, Martin On 12/20/07, Bernd Bohmann [EMAIL PROTECTED] wrote: +1 Matthias Wessendorf schrieb: yes, that is correct. -M On Dec 20, 2007 8:27 AM, Bernd Bohmann [EMAIL PROTECTED] wrote: Hello, are the release notes correct? http://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12310661styleName=Htmlversion=12312879 Regards Bernd Matthias Wessendorf schrieb: +1 On Dec 20, 2007 7:39 AM, Leonardo Uribe [EMAIL PROTECTED] wrote: Hi, I was running the needed tasks to get the 1.2.6 release of Trinidad Plugins out. This task is necessary for release Myfaces 1.2.1, because it uses faces plugin. Please note that this vote concerns all of the following parts: 1. Maven artifact group org.apache.myfaces.trinidadbuild v 1.2.6 [1] The artifacts are deployed to my private Apache account ([1]). Please take a look at the 1.2.6 artifacts and vote! Please note: This vote is majority approval with a minimum of three +1 votes (see [3]). -- -- [ ] +1 for community members who have reviewed the bits [ ] +0 [ ] -1 for fatal flaws that should cause these bits not to be released, and why.. Thanks, Leonardo Uribe [1] http://people.apache.org/~lu4242/trinidad-build126http://people.apache.org/%7Elu4242/trinidad-build126 [2] http://www.apache.org/foundation/voting.html#ReleaseVotes -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces -- Grant Smith
[VOTE] release for MyFaces 1.2.1
Hi, I was running the needed tasks to get the 1.2.1 release of Apache MyFaces core out. This mail is to make clear that the vote still continues and the problems discussed before already has been solved. Please note that this vote concerns all of the following parts: 1. Maven artifact group org.apache.myfaces.shared v3.0.1 [1] 2. Maven artifact group org.apache.myfaces.core v1.2.1 [1] The artifacts are deployed to my private Apache account ([1]). Please take a look at the 1.2.0 artifacts and vote! Please note: This vote is majority approval with a minimum of three +1 votes (see [3]). [ ] +1 for community members who have reviewed the bits [ ] +0 [ ] -1 for fatal flaws that should cause these bits not to be released, and why.. Thanks, Leonardo Uribe [1] http://people.apache.org/~lu4242/myfaces121 [2] http://www.apache.org/foundation/voting.html#ReleaseVotes
Re: [VOTE] release for MyFaces 1.2.1
On Dec 20, 2007 5:42 PM, Leonardo Uribe [EMAIL PROTECTED] wrote: Hi, I was running the needed tasks to get the 1.2.1 release of Apache MyFaces core out. This mail is to make clear that the vote still continues and the problems discussed before already has been solved. Please note that this vote concerns all of the following parts: 1. Maven artifact group org.apache.myfaces.shared v3.0.1 [1] 2. Maven artifact group org.apache.myfaces.core v1.2.1 [1] The artifacts are deployed to my private Apache account ([1]). Please take a look at the 1.2.0 artifacts and vote! 1.2.1 :-) +1 Please note: This vote is majority approval with a minimum of three +1 votes (see [3]). [ ] +1 for community members who have reviewed the bits [ ] +0 [ ] -1 for fatal flaws that should cause these bits not to be released, and why.. Thanks, Leonardo Uribe [1] http://people.apache.org/~lu4242/myfaces121 [2] http://www.apache.org/foundation/voting.html#ReleaseVotes -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: [VOTE] release for MyFaces 1.2.1
+1 On Dec 20, 2007 8:47 AM, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 5:42 PM, Leonardo Uribe [EMAIL PROTECTED] wrote: Hi, I was running the needed tasks to get the 1.2.1 release of Apache MyFaces core out. This mail is to make clear that the vote still continues and the problems discussed before already has been solved. Please note that this vote concerns all of the following parts: 1. Maven artifact group org.apache.myfaces.shared v3.0.1 [1] 2. Maven artifact group org.apache.myfaces.core v1.2.1 [1] The artifacts are deployed to my private Apache account ([1]). Please take a look at the 1.2.0 artifacts and vote! 1.2.1 :-) +1 Please note: This vote is majority approval with a minimum of three +1 votes (see [3]). [ ] +1 for community members who have reviewed the bits [ ] +0 [ ] -1 for fatal flaws that should cause these bits not to be released, and why.. Thanks, Leonardo Uribe [1] http://people.apache.org/~lu4242/myfaces121http://people.apache.org/%7Elu4242/myfaces121 [2] http://www.apache.org/foundation/voting.html#ReleaseVotes -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Grant Smith
Re: [VOTE] release for MyFaces 1.2.1
+1 On 20/12/2007, Grant Smith [EMAIL PROTECTED] wrote: +1 On Dec 20, 2007 8:47 AM, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 5:42 PM, Leonardo Uribe [EMAIL PROTECTED] wrote: Hi, I was running the needed tasks to get the 1.2.1 release of Apache MyFaces core out. This mail is to make clear that the vote still continues and the problems discussed before already has been solved. Please note that this vote concerns all of the following parts: 1. Maven artifact group org.apache.myfaces.shared v3.0.1 [1] 2. Maven artifact group org.apache.myfaces.core v1.2.1 [1] The artifacts are deployed to my private Apache account ([1]). Please take a look at the 1.2.0 artifacts and vote! 1.2.1 :-) +1 Please note: This vote is majority approval with a minimum of three +1 votes (see [3]). [ ] +1 for community members who have reviewed the bits [ ] +0 [ ] -1 for fatal flaws that should cause these bits not to be released, and why.. Thanks, Leonardo Uribe [1] http://people.apache.org/~lu4242/myfaces121 [2] http://www.apache.org/foundation/voting.html#ReleaseVotes -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Grant Smith
Re: [continuum] used jdk
nevermind. I got confused by vmbuild (welcome back to life :-) ) On Dec 20, 2007 4:42 PM, Matthias Wessendorf [EMAIL PROTECTED] wrote: I saw in the logs, that the zone uses JDK6. Is there a special reason for that ? -M -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: [VOTE] release for MyFaces 1.2.1
+1 - I looked through the artifacts, and they are ok - the shared sources are missing, something we can include in the next release. regards, Martin On 12/20/07, Bruno Aranda [EMAIL PROTECTED] wrote: +1 On 20/12/2007, Grant Smith [EMAIL PROTECTED] wrote: +1 On Dec 20, 2007 8:47 AM, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 5:42 PM, Leonardo Uribe [EMAIL PROTECTED] wrote: Hi, I was running the needed tasks to get the 1.2.1 release of Apache MyFaces core out. This mail is to make clear that the vote still continues and the problems discussed before already has been solved. Please note that this vote concerns all of the following parts: 1. Maven artifact group org.apache.myfaces.shared v3.0.1 [1] 2. Maven artifact group org.apache.myfaces.core v1.2.1 [1] The artifacts are deployed to my private Apache account ([1]). Please take a look at the 1.2.0 artifacts and vote! 1.2.1 :-) +1 Please note: This vote is majority approval with a minimum of three +1 votes (see [3]). [ ] +1 for community members who have reviewed the bits [ ] +0 [ ] -1 for fatal flaws that should cause these bits not to be released, and why.. Thanks, Leonardo Uribe [1] http://people.apache.org/~lu4242/myfaces121 [2] http://www.apache.org/foundation/voting.html#ReleaseVotes -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Grant Smith -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Fwd: [continuum] BUILD ERROR: Apache MyFaces Trinidad Examples
Could someone please fix the continuum configuration, or at least disable the scheduled build for trinidad? Since the svn changes of today, there has been a regular stream of error emails to [EMAIL PROTECTED] I presume these are because pom files have now moved, and that fixing the config is reasonably simple.. Thanks.---BeginMessage--- Online report : http://vmbuild.apache.org/continuum/buildResult.action?buildId=29537projectId=18 Build statistics: State: Error Previous State: Error Started at: Thu 20 Dec 2007 09:02:54 -0800 Finished at: Thu 20 Dec 2007 09:03:00 -0800 Total time: 5s Build Trigger: Schedule Build Number: 0 Exit code: 0 Building machine hostname: vmbuild.apache.org Operating system : Linux(unknown) Java Home version : java version 1.6.0 Java(TM) SE Runtime Environment (build 1.6.0-b105) Java HotSpot(TM) Client VM (build 1.6.0-b105, mixed mode, sharing) Builder version : Maven version: 2.0.7 Java version: 1.6.0 OS name: linux version: 2.6.20-16-server arch: i386 SCM Changes: No files changed Dependencies Changes: No dependencies changed Build Error: Provider message: The svn command failed. Command output: --- svn: REPORT request failed on '/repos/asf/!svn/vcc/default' svn: Target path does not exist --- ---End Message--- ---BeginMessage--- Online report : http://vmbuild.apache.org/continuum/buildResult.action?buildId=29537projectId=18 Build statistics: State: Error Previous State: Error Started at: Thu 20 Dec 2007 09:02:54 -0800 Finished at: Thu 20 Dec 2007 09:03:00 -0800 Total time: 5s Build Trigger: Schedule Build Number: 0 Exit code: 0 Building machine hostname: vmbuild.apache.org Operating system : Linux(unknown) Java Home version : java version 1.6.0 Java(TM) SE Runtime Environment (build 1.6.0-b105) Java HotSpot(TM) Client VM (build 1.6.0-b105, mixed mode, sharing) Builder version : Maven version: 2.0.7 Java version: 1.6.0 OS name: linux version: 2.6.20-16-server arch: i386 SCM Changes: No files changed Dependencies Changes: No dependencies changed Build Error: Provider message: The svn command failed. Command output: --- svn: REPORT request failed on '/repos/asf/!svn/vcc/default' svn: Target path does not exist --- ---End Message--- ---BeginMessage--- Online report : http://vmbuild.apache.org/continuum/buildResult.action?buildId=29537projectId=18 Build statistics: State: Error Previous State: Error Started at: Thu 20 Dec 2007 09:02:54 -0800 Finished at: Thu 20 Dec 2007 09:03:00 -0800 Total time: 5s Build Trigger: Schedule Build Number: 0 Exit code: 0 Building machine hostname: vmbuild.apache.org Operating system : Linux(unknown) Java Home version : java version 1.6.0 Java(TM) SE Runtime Environment (build 1.6.0-b105) Java HotSpot(TM) Client VM (build 1.6.0-b105, mixed mode, sharing) Builder version : Maven version: 2.0.7 Java version: 1.6.0 OS name: linux version: 2.6.20-16-server arch: i386 SCM Changes: No files changed Dependencies Changes: No dependencies changed Build Error: Provider message: The svn command failed. Command output:
Re: Java version compatibility (was Re: svn commit: r605927 - /myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java)
Mike Kienenberger [EMAIL PROTECTED] schrieb: Can't you do something like this to force the rt.jar to be 1.4? http://java.sun.com/j2se/1.5.0/docs/tooldocs/windows/javac.html % javac -target 1.4 -bootclasspath jdk1.4.2/lib/classes.zip -extdirs OldCode.java Yep. However that cannot be put into the pom because the location of the jdk1.4 will be different on everyone's machine - if they have it installed at all. And because it isn't automatic, most people won't bother to set it up. I'm not sure I would...
Re: Java version compatibility (was Re: svn commit: r605927 - /myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java)
I agree, but it could be set up on the continuum server (if it's not already set up there). I know nothing about maven, but maybe the rt.jar can be pulled down into the repository, where it should be in a standard location. On Dec 20, 2007 12:36 PM, Simon Kitching [EMAIL PROTECTED] wrote: Mike Kienenberger [EMAIL PROTECTED] schrieb: Can't you do something like this to force the rt.jar to be 1.4? http://java.sun.com/j2se/1.5.0/docs/tooldocs/windows/javac.html % javac -target 1.4 -bootclasspath jdk1.4.2/lib/classes.zip -extdirs OldCode.java Yep. However that cannot be put into the pom because the location of the jdk1.4 will be different on everyone's machine - if they have it installed at all. And because it isn't automatic, most people won't bother to set it up. I'm not sure I would...
Re: Java version compatibility (was Re: svn commit: r605927 - /myfaces/tomahawk/trunk/core/src/main/java/org/apache/myfaces/dateformat/SimpleDateFormatter.java)
On 12/20/07, Simon Kitching [EMAIL PROTECTED] wrote: Mike Kienenberger [EMAIL PROTECTED] schrieb: Can't you do something like this to force the rt.jar to be 1.4? http://java.sun.com/j2se/1.5.0/docs/tooldocs/windows/javac.html % javac -target 1.4 -bootclasspath jdk1.4.2/lib/classes.zip -extdirs OldCode.java Yep. However that cannot be put into the pom because the location of the jdk1.4 will be different on everyone's machine - if they have it installed at all. And because it isn't automatic, most people won't bother to set it up. I'm not sure I would... Continuum 1.1 has the ability to execute builds using a specific JDK, build tool version and/or environment variables in a kind of Build Environment. (It's currently called a Build Profile which is horribly confusing, since it's completely different from the profile element in the pom. I'm working on getting it renamed. :) ) -- Wendy
[jira] Created: (TRINIDAD-879) NLS: Translation to always honor view locale not formatting locale
NLS: Translation to always honor view locale not formatting locale -- Key: TRINIDAD-879 URL: https://issues.apache.org/jira/browse/TRINIDAD-879 Project: MyFaces Trinidad Issue Type: Bug Reporter: Jeanne Waldman Assignee: Jeanne Waldman Trinidad embedded translations should always honor view locale. Problem: 1. Usually Trinidad embedded translations are honoring view locale 2. However, when formatting locale is specified, translations will be following formatting locale instead. In theory, only localization definitions (such as date format pattern) should follow formatting locale. Here's a testcase. The idea is: - we would like to have formatting support according to locale Traditional Chinese-Hong Kong (zh-HK). - However, language translations in Traditional Chinese (zh-TW) due to that fact that we do not provide zh-HK translations. facestring.jspx === ?xml version=1.0 encoding=iso-8859-1 standalone=yes ? jsp:root xmlns:jsp=http://java.sun.com/JSP/Page; version=2.0 xmlns:f=http://java.sun.com/jsf/core; xmlns:tr=http://myfaces.apache.org/trinidad; jsp:directive.page contentType=text/html;charset=utf-8/ f:view locale=zh-TW tr:document title=InputText Demo tr:form p tr:outputText value=view locale #{view.locale}/ /p tr:inputText label=test value=23245 tr:convertDateTime dateStyle=short/ /tr:inputText tr:commandButton text=Submit/ /tr:form /tr:document /f:view /jsp:root faces-config.xml ... supported-localezh-TW/supported-locale supported-localezh-HK/supported-locale trinidad-config.xml === ?xml version=1.0 encoding=UTF-8? trinidad-config xmlns=http:?/?/myfaces.apache.org?/trinidad?/config formatting-localezh-HK/formatting-locale /trinidad-config Problem: While formatting is presented in zh-HK, language translations are not shown in Traditional Chinese. Valid example: 98ChineseChar11ChineseChar29ChineseChar As a Reference Testcase: Remove formatting-locale in trinidad-config.xml (i.e. only view locale = zh-tw below) Traditional Chinese translations present and shown correctly. ChineseChars 23245 ChineseChars. ChineseChars: 1998/11/29. While Trinidad supports more than 130 locales in terms of localization formatting (there are 132 files of META-INF/adf/jsLibs/resources/LocaleElements_*.js), - there are only 32 languages provided for translations (there are 33 files of oracle/adfinternal/view/faces/renderkit/rich/resource/RichBundle_*.class) A fallback approach will be often required for appropriate i18n support. For example: locale formatting language translations = = zh-HK zh-TW zh-SG zh-CN zh-MO zh-TW etc. This bug and analysis was found by our NLS expert. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.
[Trinidad] NLS bug re: formatting locale
Hi there, Our local NLS expert brought this bug to my attention. I created a JIRA issue https://issues.apache.org/jira/browse/TRINIDAD-879 Here is what I have there. Let me know if you think that this is not a bug. I plan to fix it. Thanks! Jeanne Trinidad embedded translations should always honor view locale. Problem: 1. Usually Trinidad embedded translations are honoring view locale 2. However, when formatting locale is specified, translations will be following formatting locale instead. In theory, only localization definitions (such as date format pattern) should follow formatting locale. Here's a testcase. The idea is: - we would like to have formatting support according to locale Traditional Chinese-Hong Kong (zh-HK). - However, language translations in Traditional Chinese (zh-TW) due to that fact that we do not provide zh-HK translations. facestring.jspx === ?xml version=1.0 encoding=iso-8859-1 standalone=yes ? jsp:root xmlns:jsp=http://java.sun.com/JSP/Page; version=2.0 xmlns:f=http://java.sun.com/jsf/core; xmlns:tr=http://myfaces.apache.org/trinidad; jsp:directive.page contentType=text/html;charset=utf-8/ f:view locale=zh-TW tr:document title=InputText Demo tr:form p tr:outputText value=view locale #{view.locale}/ /p tr:inputText label=test value=23245 tr:convertDateTime dateStyle=short/ /tr:inputText tr:commandButton text=Submit/ /tr:form /tr:document /f:view /jsp:root faces-config.xml ... supported-localezh-TW/supported-locale supported-localezh-HK/supported-locale trinidad-config.xml === ?xml version=1.0 encoding=UTF-8? trinidad-config xmlns=http:?/?/myfaces.apache.org?/trinidad?/config formatting-localezh-HK/formatting-locale /trinidad-config Problem: While formatting is presented in zh-HK, language translations are not shown in Traditional Chinese. Valid example: 98ChineseChar11ChineseChar29ChineseChar As a Reference Testcase: Remove formatting-locale in trinidad-config.xml (i.e. only view locale = zh-tw below) Traditional Chinese translations present and shown correctly. ChineseChars 23245 ChineseChars. ChineseChars: 1998/11/29. While Trinidad supports more than 130 locales in terms of localization formatting (there are 132 files of META-INF/adf/jsLibs/resources/LocaleElements_*.js), - there are only 32 languages provided for translations (there are 33 files of oracle/adfinternal/view/faces/renderkit/rich/resource/RichBundle_*.class) A fallback approach will be often required for appropriate i18n support. For example: locale formatting language translations = = zh-HK zh-TW zh-SG zh-CN zh-MO zh-TW etc.
Re: [VOTE] release for MyFaces 1.2.1
+1 On Dec 20, 2007 7:26 PM, Martin Marinschek [EMAIL PROTECTED] wrote: +1 - I looked through the artifacts, and they are ok - the shared sources are missing, something we can include in the next release. regards, Martin On 12/20/07, Bruno Aranda [EMAIL PROTECTED] wrote: +1 On 20/12/2007, Grant Smith [EMAIL PROTECTED] wrote: +1 On Dec 20, 2007 8:47 AM, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 5:42 PM, Leonardo Uribe [EMAIL PROTECTED] wrote: Hi, I was running the needed tasks to get the 1.2.1 release of Apache MyFaces core out. This mail is to make clear that the vote still continues and the problems discussed before already has been solved. Please note that this vote concerns all of the following parts: 1. Maven artifact group org.apache.myfaces.shared v3.0.1 [1] 2. Maven artifact group org.apache.myfaces.core v1.2.1 [1] The artifacts are deployed to my private Apache account ([1]). Please take a look at the 1.2.0 artifacts and vote! 1.2.1 :-) +1 Please note: This vote is majority approval with a minimum of three +1 votes (see [3]). [ ] +1 for community members who have reviewed the bits [ ] +0 [ ] -1 for fatal flaws that should cause these bits not to be released, and why.. Thanks, Leonardo Uribe [1] http://people.apache.org/~lu4242/myfaces121http://people.apache.org/%7Elu4242/myfaces121 [2] http://www.apache.org/foundation/voting.html#ReleaseVotes -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Grant Smith -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
Re: [VOTE] release for MyFaces 1.2.1
I did a test-drive w/ trinidad 1.2.5-SNAPSHOT all cool over there. good job guys! On Dec 20, 2007 5:47 PM, Matthias Wessendorf [EMAIL PROTECTED] wrote: On Dec 20, 2007 5:42 PM, Leonardo Uribe [EMAIL PROTECTED] wrote: Hi, I was running the needed tasks to get the 1.2.1 release of Apache MyFaces core out. This mail is to make clear that the vote still continues and the problems discussed before already has been solved. Please note that this vote concerns all of the following parts: 1. Maven artifact group org.apache.myfaces.shared v3.0.1 [1] 2. Maven artifact group org.apache.myfaces.core v1.2.1 [1] The artifacts are deployed to my private Apache account ([1]). Please take a look at the 1.2.0 artifacts and vote! 1.2.1 :-) +1 Please note: This vote is majority approval with a minimum of three +1 votes (see [3]). [ ] +1 for community members who have reviewed the bits [ ] +0 [ ] -1 for fatal flaws that should cause these bits not to be released, and why.. Thanks, Leonardo Uribe [1] http://people.apache.org/~lu4242/myfaces121 [2] http://www.apache.org/foundation/voting.html#ReleaseVotes -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: [continuum] BUILD ERROR: Apache MyFaces Trinidad Examples
yah, tomorrow :-) On Dec 20, 2007 6:32 PM, Simon Kitching [EMAIL PROTECTED] wrote: Could someone please fix the continuum configuration, or at least disable the scheduled build for trinidad? Since the svn changes of today, there has been a regular stream of error emails to [EMAIL PROTECTED] I presume these are because pom files have now moved, and that fixing the config is reasonably simple.. Thanks. -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
Re: [continuum] BUILD ERROR: Apache MyFaces Trinidad Examples
fixed (deleted) On Dec 20, 2007 11:21 PM, Matthias Wessendorf [EMAIL PROTECTED] wrote: yah, tomorrow :-) On Dec 20, 2007 6:32 PM, Simon Kitching [EMAIL PROTECTED] wrote: Could someone please fix the continuum configuration, or at least disable the scheduled build for trinidad? Since the svn changes of today, there has been a regular stream of error emails to [EMAIL PROTECTED] I presume these are because pom files have now moved, and that fixing the config is reasonably simple.. Thanks. -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org -- Matthias Wessendorf further stuff: blog: http://matthiaswessendorf.wordpress.com/ sessions: http://www.slideshare.net/mwessendorf mail: matzew-at-apache-dot-org
[jira] Created: (TRINIDAD-880) Clientside validation errors when using PPR
Clientside validation errors when using PPR --- Key: TRINIDAD-880 URL: https://issues.apache.org/jira/browse/TRINIDAD-880 Project: MyFaces Trinidad Issue Type: Bug Affects Versions: 1.2.4-core, 1.2.3-core Environment: Windows XP Pro SP2, Facelets 1.1.13, Sun Java System Application Server 9.1, JSF1.2_07-b03 Reporter: Graeme Steyn To reproduce the problem deploy the attached sample application and execute the following sequence of actions. 1. All radio buttons should be blank when form is first displayed - press Submit - validation error occurs for first question (expected). 2. Select Yes for Is English your first language? - press Submit - validation error is cleared and form re-displayed. 3. Select No for Is English your first language? - press Submit - validation error occurs for second question (expected). 4. Select Yes for Is English your first language? - ppr results in second question being disabled, not required and clears validation message. 5. Press Submit - validation error occurs for second question (fault). 6. Select No for Is English your first language? 7. Select Yes for second question. Notice that a new required field appears. 8. Press Submit - validation error occurs for the new field (expected). 8. Select No for second question. The additional required field disappears. 9. Press Submit - Error appears: Field Error [certDetails] - Enter a value (fault). 10. Acknowledge and select Yes for Is English your first language? - press Submit - 2 validation error messages occur (fault). Using Firebug, it appears that the ppr is taking place correctly, but checking the DOM for _english_Validators shows that all three validators are eventually listed. I am assuming that this is what is causing the problem, as when the lower two fields are disabled/hidden, the validators are not removed from _english_Validators. It appears that the problem may be related to TRINIDAD-96 - Validators should be added incrementally, not set as a whole. The problem is not present in Trinidad 1.2.2, but does appear in 1.2.3 and 1.2.4. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.