Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!uwm.edu!zaphod.mps.ohio-state.edu!samsung!emory!cambridge.apple.com!bloom-beacon!convex.UUCP!root From: root@convex.UUCP (Superuser) Newsgroups: comp.windows.x Subject: Lost mail for you Message-ID: <9006071737.AA16426@sushi> Date: 7 Jun 90 17:37:29 GMT Sender: root@athena.mit.edu (Wizard A. Root) Organization: The Internet Lines: 48 From pescadero.stanford.edu!expo.lcs.mit.edu!xpert-mailer Sat Jun 2 15:46:21 1990 remote from convex Received: by sushi (5.51/7.0) id AA03778; Sat, 2 Jun 90 15:46:21 CDT Received: by convex.COM (5.51/4.7) id AA28559; Sat, 2 Jun 90 15:46:15 CDT Received: from Erebus.Stanford.EDU by uxc.cso.uiuc.edu with SMTP (5.62+/IDA-1.2.8-900601) id AA04261 (for salevin%drlc1.UUCP@convex.com); Sat, 2 Jun 90 15:42:08 -0500 Received: from Hanauma.Stanford.EDU by erebus.Stanford.EDU with TCP; Sat, 2 Jun 90 13:42:04 PDT Received: by hanauma.stanford.edu (5.51/7.0) id AA15675; Sat, 2 Jun 90 13:41:54 PDT Received: from EXPO.LCS.MIT.EDU by Pescadero.Stanford.EDU (5.59/25-eef) id AA27827; Sat, 2 Jun 90 13:40:55 PDT Received: by expo.lcs.mit.edu; Sat, 2 Jun 90 13:31:00 EDT Received: from bloom-beacon.MIT.EDU by expo.lcs.mit.edu; Sat, 2 Jun 90 13:30:47 EDT Received: by bloom-beacon.MIT.EDU (5.61/25-eef) id AA18191; Sat, 2 Jun 90 13:25:35 EDT Received: from USENET by bloom-beacon.mit.edu with netnews for xpert@expo.lcs.mit.edu (xpert@expo.lcs.mit.edu) (contact usenet@bloom-beacon.mit.edu if you have questions) Date: 1 Jun 90 09:56:23 GMT From: convex!uunet.uu.net!vyncke%esat%kulesat%kulcs%ub4b%mcsun Organization: Katholieke Universiteit Leuven,ESAT - dep. Elect. Engineering, Belgium Subject: again : avoiding event loops Message-Id: <2998.26663bc7@esat.kuleuven.ac.be> Sender: convex!expo.lcs.mit.edu!xpert-request To: xpert@expo.lcs.mit.edu I'm writing an application where the graphical interface is not the main part of the program. Since I want my windows to be updated after expose events, I need a kind of asynchrounous behaviour. I first build the program on a VAX station 3100 and there I had a routine called XSelectAsyncInput(...) . This worked fine. But two months ago I switched to a Sun Sparc with X11R4. There I tried to catch the events in a background process. This however was not failsafe. Now I implemented the method of John Bruner : > My approach is to get the file descriptor for the X server (with > ConnectionNumber(display)) and catch SIGIO. This works fine but there is still one problem : If I catch an interrupt when I'm in a X-call, my programm crashes. I worked around this by postponing the interrupt handling while I'm doing X-calls. This is however elaborate, and I'm wondering if there is not an other way to prevent the Xserver from receiving two concurrent X-calls Please send any reply by Email -- Frank Vyncke . Department of Robot Vision, KU Leuven, Belgium vyncke@esat.kuleuven.ac.be