I reverted most (if not all) of the code + I did notice the misbehavior in some ajax responses that Zcombor mentioned so I've added the related logic in InjectAjaxComponentRenderWorker.buildMethodCallBack
Both of the tests you mentioned now pass, so your apps should now work. However, I did notice that Bug85 fails, i.e. Autocompleters in a loop... I remember this was working in 4.0.1 final, but i'll be away for the weekend and won't be able to investigate more - if anyone wants to talk a look, please, go on! andyhot wrote: > Leo, Zombi: i'd be nice to also have the automated tests for bug142 & > bug143 but what you've done is a good start > > Zombi: i'm working on this - thanks for the debugging. > > I haven't seen this issue in Tapestry 4.1.x versions. And as for > upgrading to Tapestry 4.1.2, > it really depends on which tacos components you use... > For instance there's no progress bar yet but on the other hand there's > the powerful > EventListener annotation - all in all the upgrade will certainly take > some days but it's worth it, > esp. if you can use java 1.5 features. > > >> On 6/7/07, *Andreas Andreou* < [EMAIL PROTECTED] >> <mailto:[EMAIL PROTECTED]>> wrote: >> >> Hi, perhaps im also responsible here... >> I should have taken a closer look, but i've heard lots of >> complaints about that >> change + i didn't get any response from Viktor as to why it was >> needed or any test case >> >> And i think i've emailed him twice about that - and not because i >> was using it - but because >> other users complained... so, now that Leonardo also confirmed this... >> >> Anyway, the good thing is that we now have tests that show both >> problems, so we can nail it down >> & everyone will be happy. >> >> Leo, i too hate those formatting changes in commit messages, >> exactly because i like to go through >> all of them - so, please, try to avoid them. >> >> Anyway, ttyl - im trying the tests now :) >> >> >> >> >> Ok, i've managed to debug the issues further. So I think the main >> problem ,is that during the rewind cycle, the components use the same >> 'NestedMarkupWriterImpl' which gets injected in the >> InjectAjaxComponentRenderWorker.buildMethodCallBack however in the >> rewind cycle, it should render to a NullWriter. So in the Bug142, the >> ajax response contains the followings: >> <response type="element" id="testZone"> >> <fieldset> >> <legend>Test Zone</legend> >> <p> >> <strong>This zone should appear only once.</strong> >> </p> >> <p> >> Lorem ipsum dolor sit amet, consectetur adipisicing elit, >> sed do eiusmod tempor incididunt ut labore et dolore magna >> aliqua. Ut enim ad minim veniam, quis nostrud exercitation >> ullamco laboris nisi ut aliquip ex ea commodo consequat. >> Duis aute irure dolor in reprehenderit in voluptate velit >> esse cillum dolore eu fugiat nulla pariatur. Excepteur sint >> occaecat cupidatat non proident, sunt in culpa qui officia >> deserunt mollit anim id est laborum. >> </p> >> >> >> </fieldset> >> <div id="testZone"> >> <fieldset> >> <legend>Test Zone</legend> >> <p> >> <strong>This zone should appear only once.</strong> >> </p> >> <p> >> So basicly the same component tree, twice, once without a proper 'div' >> node, and once with a div node. The first happens in the rewind cycle, >> the difference between the two response comes from the fact, that many >> components do not render their tags, in the rewind step, only their >> body. Because of other problems in the javascript side, hides this >> error, so it worked correclty. I'm not sure exactly why. The recent >> changes which fixed the xml correctness issues caused this double >> rendering in a lot's of cornercase. >> I think the correct fix is to modify the buildMethodCallBack to check >> for the rewinding phase, and do not substitute the null writer if it's >> in the rewind. My main question, is this just missed somehow in the >> tacos library? Is it fixed in the tapestry 4.1 branch, which I heared >> has an integrated ajax functionality ? And lastly, how far the >> tacos-4.0 branch from tapestry 4.1 ? How painfull the upgrade? >> >> BR, >> Zsombor >> >> ui. please, please, if you do not have time, or intention to fix >> problems caused by commits made by you in other persons applications, >> please do not modify the code, without discussion. I try to follow >> this practice, thanks. >> >> >> ------------------------------------------------------------------------ >> >> ------------------------------------------------------------------------- >> This SF.net email is sponsored by DB2 Express >> Download DB2 Express C - the FREE version of DB2 express and take >> control of your XML. No limits. Just data. Click to get it now. >> http://sourceforge.net/powerbar/db2/ >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> Tacos-devel mailing list >> Tacos-devel@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/tacos-devel >> >> > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by DB2 Express > Download DB2 Express C - the FREE version of DB2 express and take > control of your XML. No limits. Just data. Click to get it now. > http://sourceforge.net/powerbar/db2/ > _______________________________________________ > Tacos-devel mailing list > Tacos-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/tacos-devel > > ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ Tacos-devel mailing list Tacos-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/tacos-devel
