Lifecycle of a tapestry request and the MarkupRendererFilter
Hi, I have been successfully using the CacheControlMarkupRenderer from the ioko-tapestry-commons http://tapestry.formos.com/projects/ioko-tapestry-commons/ library. This is a java class that implements the MarkupRendererFilter and Partial MarkupRendererFilter in order to apply caching headers to pages served by our application. This is all working wonderfully, however we now have a component which is returning a StreamResponse object out of the onActivate method of the component. This of course does not invoke the renderMarkup(MarkupWriter writer, MarkupRenderer renderer) method on the CacheControlMarkupRenderer as it is not markup. Is there an equivalent filter we could use that would enable us to set headers on the Response object but would be applied to both StreamResponse objects as well as the output from normal components with markup in a tml file. Thanks, Ian
RE: Lifecycle of a tapestry request and the MarkupRendererFilter
Hi, We do actually have a load balanced caching system sitting in front our app server, however we control the caching behaviour by setting headers on the HTTP responses. For example in this particular case we need to set: response.setHeader(CACHE_CONTROL_HEADER, no-cache); response.setHeader(PRAGMA_HEADER, No-Cache); -Original Message- From: cordenier christophe [mailto:christophe.corden...@gmail.com] Sent: 15 October 2009 14:58 To: Tapestry users Subject: Re: Lifecycle of a tapestry request and the MarkupRendererFilter Hello From my point of view this kind of caching mechanism should be done in front of your application server. Regards Christophe. 2009/10/15 Ian Petzer ian.pet...@ioko.com Hi, I have been successfully using the CacheControlMarkupRenderer from the ioko-tapestry-commons http://tapestry.formos.com/projects/ioko-tapestry-commons/ library. This is a java class that implements the MarkupRendererFilter and Partial MarkupRendererFilter in order to apply caching headers to pages served by our application. This is all working wonderfully, however we now have a component which is returning a StreamResponse object out of the onActivate method of the component. This of course does not invoke the renderMarkup(MarkupWriter writer, MarkupRenderer renderer) method on the CacheControlMarkupRenderer as it is not markup. Is there an equivalent filter we could use that would enable us to set headers on the Response object but would be applied to both StreamResponse objects as well as the output from normal components with markup in a tml file. Thanks, Ian - To unsubscribe, e-mail: users-unsubscr...@tapestry.apache.org For additional commands, e-mail: users-h...@tapestry.apache.org
RE: Lifecycle of a tapestry request and the MarkupRendererFilter
Hi, This is exactly how I am currently setting the headers, however I don't want to do this on an individual basis for all StreamResponses. What I'm looking for a is a filter or similar mechanism that can be used to post-process all responses irrespective of whether they are StreamResponse objects or the output from normal components with markup in a tml file. Essentially a filter like the MarkupRendererFilter that works on StreamResponses as well Thanks Ian -Original Message- From: cordenier christophe [mailto:christophe.corden...@gmail.com] Sent: 15 October 2009 15:31 To: Tapestry users Subject: Re: Lifecycle of a tapestry request and the MarkupRendererFilter What not using the prepareResponse of the StreamResponse Tapestry interface to set headers ? 2009/10/15 Ian Petzer ian.pet...@ioko.com Hi, We do actually have a load balanced caching system sitting in front our app server, however we control the caching behaviour by setting headers on the HTTP responses. For example in this particular case we need to set: response.setHeader(CACHE_CONTROL_HEADER, no-cache); response.setHeader(PRAGMA_HEADER, No-Cache); -Original Message- From: cordenier christophe [mailto:christophe.corden...@gmail.com] Sent: 15 October 2009 14:58 To: Tapestry users Subject: Re: Lifecycle of a tapestry request and the MarkupRendererFilter Hello From my point of view this kind of caching mechanism should be done in front of your application server. Regards Christophe. 2009/10/15 Ian Petzer ian.pet...@ioko.com Hi, I have been successfully using the CacheControlMarkupRenderer from the ioko-tapestry-commons http://tapestry.formos.com/projects/ioko-tapestry-commons/ library. This is a java class that implements the MarkupRendererFilter and Partial MarkupRendererFilter in order to apply caching headers to pages served by our application. This is all working wonderfully, however we now have a component which is returning a StreamResponse object out of the onActivate method of the component. This of course does not invoke the renderMarkup(MarkupWriter writer, MarkupRenderer renderer) method on the CacheControlMarkupRenderer as it is not markup. Is there an equivalent filter we could use that would enable us to set headers on the Response object but would be applied to both StreamResponse objects as well as the output from normal components with markup in a tml file. Thanks, Ian - To unsubscribe, e-mail: users-unsubscr...@tapestry.apache.org For additional commands, e-mail: users-h...@tapestry.apache.org - To unsubscribe, e-mail: users-unsubscr...@tapestry.apache.org For additional commands, e-mail: users-h...@tapestry.apache.org
Re: Deep linking to tapestry pages from jps. Conversion
Hi Howard, Unfortantely this approach of encoding the information is request params is not really ideal. As Russel has said, we are halfway through converting our application. We have created Tapestry pages to supply some new functionality (various RSS feeds in this case) however we need to link to this functionality from existing jsp pages which haven't been converted yet and new tapestry pages. The new tapestry pages are a doddle as we can simply use a pageLink and pass through the relative url as a param in the context for the actionLink. What I'm really looking for is guidance on hwo to create a Link to mimic what is created by the pageLink component. Even if you could just point me at the right class which encodes the param values and introduces the special behaviour for slashes in param values being double encoded. Thanks for your help, Ian On Mon, Oct 27, 2008 at 10:32 AM, Russell Brown [EMAIL PROTECTED]wrote: Hi Borut, We are migrating a Struts app at the moment so if you have any specific questions about stuff then fire away as we _may_ have solved one or two problems that you might face already. Such a guide would be a good idea: it is ideal to start a nice new app with a nice new framework but in the real world most people are working on legacy applications and migration is the normal use case. If T5 wants any real penetration then thinking about migrating users from Struts to T5 is a must. Although it is essentially a marketing problem and most developers struggle with marketing. Cheers Russell -Original Message- From: Borut Bolčina [mailto:[EMAIL PROTECTED] Sent: 24 October 2008 21:04 To: Tapestry users Subject: Re: Deep linking to tapestry pages from jps. Conversion Hi, it would be great if there was also a JSP transition guide. We have a large JSP application at the moment and my intention is to replace it gradually with Tapestry 5. Any advice on how to do it *the right way* would be more than welcome! Regards, Borut 2008/10/24 Howard Lewis Ship [EMAIL PROTECTED] For deep linking purposes, could you pass the information as normal query parameters? You could then @Inject the Request object to retrieve those values. You can also use the ContextValueEncoder service to decode them to appropriate types. This work would be done in the page's activate event handler method. This is a more servlet-style of development, which may be the right approach for a transitional application. On Thu, Oct 23, 2008 at 8:33 AM, Ian Petzer [EMAIL PROTECTED] wrote: Hi, We are busy converting our Struts/Tiles based application to Tapestry. Until this is complete I have am required to generate links to the new Tapestry pages correctly from the jsp pages. Normally this isn't a problem as I used the url pattern of: HOST/PAGE/PARAM1/PARAM2/PARAM3 Now, please consider the situation where my params have following values: PARAM1 = Hello PARAM2 = relative/url PARAM3 = colon:seperated So if I follow my previous url pattern i would get HOST/PAGE/Hello/relative/url/colon:seperated This of course doesn't work as that url is interpreted as having four params: [Hello, relative, url, colon:seperated] instead of the three that I define. When I use a pageLink object on Tapestry pages this isn't a problem as param2 is double url encoded, so that relative/url -- relative%252Furl However the colon in param3 isn't double url encoded. I would really appreciate it if someone could indicate the rules which are used when forming these urls or if someone could suggest an approach to generate these urls. Possibly a reverse @Inject of the relevant Tapestry object into my Spring service where I could call a method to convert the context values. Thanks, Ian -- Howard M. Lewis Ship Creator Apache Tapestry and Apache HiveMind - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Deep linking to tapestry pages from jps. Conversion
Hi, We are busy converting our Struts/Tiles based application to Tapestry. Until this is complete I have am required to generate links to the new Tapestry pages correctly from the jsp pages. Normally this isn't a problem as I used the url pattern of: HOST/PAGE/PARAM1/PARAM2/PARAM3 Now, please consider the situation where my params have following values: PARAM1 = Hello PARAM2 = relative/url PARAM3 = colon:seperated So if I follow my previous url pattern i would get HOST/PAGE/Hello/relative/url/colon:seperated This of course doesn't work as that url is interpreted as having four params: [Hello, relative, url, colon:seperated] instead of the three that I define. When I use a pageLink object on Tapestry pages this isn't a problem as param2 is double url encoded, so that relative/url -- relative%252Furl However the colon in param3 isn't double url encoded. I would really appreciate it if someone could indicate the rules which are used when forming these urls or if someone could suggest an approach to generate these urls. Possibly a reverse @Inject of the relevant Tapestry object into my Spring service where I could call a method to convert the context values. Thanks, Ian
