On 26 April 2012 21:00, Philippe Mouawad <[email protected]> wrote: > Hello, > I was wondering if this code is right: > * try { > convertSubTree(subTree); > } catch (Exception err) { > log.warn("Error converting subtree "+err); > }* > > Why isn't exception thrown ?
No idea - I added the log message to at least record any problems. > is it right to use the subtree if an error occurs in convertSubtree ? Again, no idea. > I reviewed code, it seems right to me. > But I still have no save dialog when using shortcut CMD + Q on Mac OSX. That is something you will have to debug; I don't have access to a Mac. The code to intercept the close is set up in MainFrame#init() which calls addWindowListener(new WindowHappenings()); Maybe that needs to be tweaked for the Mac? > Regards > Philippe > > On Wed, Apr 25, 2012 at 3:31 PM, sebb <[email protected]> wrote: > >> On 25 April 2012 14:06, sebb <[email protected]> wrote: >> > On 24 April 2012 21:58, Philippe Mouawad <[email protected]> >> wrote: >> >> Hello, >> >> Note that I am on a Mac OSX. >> >> I am testing on trunk. >> > >> > I am seeing the problem now; I think my test plan was too short. >> > >> > The moved call to >> > ActionRouter.getInstance().doActionNow(new ActionEvent(saveTree, >> > e.getID(), ActionNames.SUB_TREE_SAVED)); >> > seems to be the root of the problem. >> > >> > I'm investigating now. >> >> I think I've now fixed it - try and break it, please! >> >> The problem was that convertSubTree changed the tree so checkDirty no >> longer worked. >> Also fixed another issue - the Action saves the current tree for later >> comparison. >> Does not make sense to do this for partial tree saves, as changes >> outside the selected portion would be missed. >> >> >> Regards >> >> Philippe >> >> >> >> On Tue, Apr 24, 2012 at 10:15 PM, sebb <[email protected]> wrote: >> >> >> >>> On 24 April 2012 20:08, Philippe Mouawad <[email protected]> >> >>> wrote: >> >>> > IN fact it happens If I use shortcut but not if I use menu on Mac >> OSX. >> >>> > >> >>> > On Tue, Apr 24, 2012 at 9:05 PM, Philippe Mouawad < >> >>> > [email protected]> wrote: >> >>> > >> >>> >> I think I didn't test correctly, in fact to simulate failure I >> threw an >> >>> >> exception in code before this test, that's why I failed detecting >> it. >> >>> >> >>> I made the test plan read-only for testing. >> >>> >> >>> >> And yes , it seems I haven't created any test plan since yesterday >> :-) , >> >>> >> by lack of chance I am creating a Test Plan but using an already >> created >> >>> >> one. >> >>> >> >> >>> >> So I have the same result as you on canRead / canWrite. >> >>> >> >> >>> >> >>> OK, useful to know. >> >>> >> >>> >> By the way I noticed the following strange behaviour related to save >> >>> >> (which seems to exist also in 2.6): >> >>> >> >> >>> >> - Start JMeter >> >>> >> - Add Thread Group and other Test Eleemnt >> >>> >> - Quit >> >>> >> - JMeter does not ask you to save, shouldn't be so ? >> >>> >> >>> I don't get the same behvaiour in the current SVN code. >> >>> >> >>> >> >> >>> >> Regards >> >>> >> >> >>> >> Philippe >> >>> >> >> >>> >> On Tue, Apr 24, 2012 at 8:07 PM, sebb <[email protected]> wrote: >> >>> >> >> >>> >>> On 24 April 2012 18:09, Philippe Mouawad < >> [email protected]> >> >>> >>> wrote: >> >>> >>> > As you always say, good catch :-) >> >>> >>> > And sorry for being too hasty on this fix. >> >>> >>> >> >>> >>> No problem. >> >>> >>> >> >>> >>> I'm just a bit surprised you did not notice the problem, unless >> >>> >>> canWrite() behaves differently for your (or you did not create any >> new >> >>> >>> plans yet!). >> >>> >>> >> >>> >>> > Regards >> >>> >>> > Philippe >> >>> >>> > >> >>> >>> > On Tue, Apr 24, 2012 at 2:38 PM, <[email protected]> wrote: >> >>> >>> > >> >>> >>> >> https://issues.apache.org/bugzilla/show_bug.cgi?id=52997 >> >>> >>> >> >> >>> >>> >> --- Comment #3 from Sebb <[email protected]> 2012-04-24 12:38:21 >> UTC >> >>> --- >> >>> >>> >> Fixed in SVN: >> >>> >>> >> >> >>> >>> >> URL: http://svn.apache.org/viewvc?rev=1329669&view=rev >> >>> >>> >> Log: >> >>> >>> >> Bug 52997 - Jmeter should not exit without saving Test Plan if >> >>> saving >> >>> >>> >> before >> >>> >>> >> exit fails >> >>> >>> >> >> >>> >>> >> Modified: >> >>> >>> >> jmeter/trunk/src/core/org/apache/jmeter/gui/action/Save.java >> >>> >>> >> >> >>> >>> >> >> >>> >>> >> Note: I removed the setDirty(true) method call and the clearing >> of >> >>> the >> >>> >>> file >> >>> >>> >> name. >> >>> >>> >> Failure to save the file should not affect the dirty status. >> >>> >>> >> Clearing the filename does not really achieve anything. >> >>> >>> >> >> >>> >>> >> Leaving the issue open for now, in case there are further >> problems. >> >>> >>> >> >> >>> >>> >> -- >> >>> >>> >> Configure bugmail: >> >>> >>> >> https://issues.apache.org/bugzilla/userprefs.cgi?tab=email >> >>> >>> >> ------- You are receiving this mail because: ------- >> >>> >>> >> You are the assignee for the bug. >> >>> >>> >> >> >>> >>> > >> >>> >>> > >> >>> >>> > >> >>> >>> > -- >> >>> >>> > Cordialement. >> >>> >>> > Philippe Mouawad. >> >>> >>> >> >>> >> >> >>> >> >> >>> >> >> >>> >> -- >> >>> >> Cordialement. >> >>> >> Philippe Mouawad. >> >>> >> >> >>> >> >> >>> >> >> >>> >> >> >>> > >> >>> > >> >>> > -- >> >>> > Cordialement. >> >>> > Philippe Mouawad. >> >>> >> >> >> >> >> >> >> >> -- >> >> Cordialement. >> >> Philippe Mouawad. >> > > > > -- > Cordialement. > Philippe Mouawad.
