Hi Kerry, to keep the GUI active while you are doing other things (like filling histograms) call regularly: gSystem->ProcessEvents(). This call will process all pending GUI events. For examples look into the $ROOTSYS/tutorials directory and search for ProcessEvents. This kind of processing is synchronous, does not use threads but should give the same effect. Cheers, Fons. On Tue, 2004-01-20 at 16:39, LEE, KERRY T. (JSC-SR) (UHCL) wrote: > ROOT team, > > I sent this message last week and since I have been spoiled in the past > with quick replys I think it is possible that it has been overlooked. > > I use RH 9.0 gcc 3.2.2, and tried my program with ROOT versions 3.10.02 > and 4.00.00 with the same results. > > Thanks > Kerry > > -----Original Message----- > From: owner-roottalk@pcroot.cern.ch > To: 'roottalk@pcroot.cern.ch' > Sent: 1/16/2004 4:46 PM > Subject: [ROOT] GUI classes with TThread > > Dear ROOTers, > > I am attempting to create an online histogram monitoring GUI, in which > many > histograms are filled , but only displaying a few at a time. I > currently > have a TListBox containing a list that selects which histograms are to > be > displayed, and a TROOTEnbededCanvas to display the histograms in. This > worked perfectly as long as the histograms were finished being filled > (after > my simulated online run). While the run was in progress I could not > interact with the GUI and change the histogram display. My next step > was to > introduce TThreads via the TThreadframe example. Unfortunately I did > this > before coming across the previous message > > http://atlassw1.phy.bnl.gov/doc/root/RootHtmlDoc/roottalk/roottalk99/161 > 9.ht > ml > > which states that "Access to TCanvas, TPad and other graphics from > threads. > Does not work, even if access is locked explicitly. > It seems that the errors occur in the X event handling > outside root. Moving the mouse accelerates the crashes considerably." > Prior > to reading this I concluded (from stack dump) that there was probably > something wrong with X event handling. My program ran fine as long as I > didn't select a different set of histograms to be displayed, and > occasionally I could display different histogram sets but a crash was > inevitable if I did that. Further in the above mail it was stated that > this > problem was not likely to ever be fixed. > > Therefore, if anyone has similar monitoring requirements I'd like to > hear > how you overcame these problems. Any help is appreciated. > > Thanks > Kerry -- Org: CERN, European Laboratory for Particle Physics. Mail: 1211 Geneve 23, Switzerland E-Mail: Fons.Rademakers@cern.ch Phone: +41 22 7679248 WWW: http://www.rademakers.org/fons/ Fax: +41 22 7679480
This archive was generated by hypermail 2b29 : Sun Jan 02 2005 - 05:50:05 MET