From: mkr <mkr@opencascade.com>
Date: Tue, 31 Jan 2006 09:12:03 +0000 (+0000)
Subject: Fix for bug IPAL11408 : "Run()" method returns "0" after trying to start graph execution.
X-Git-Tag: T2_2_9pre~1
X-Git-Url: http://git.salome-platform.org/gitweb/?a=commitdiff_plain;h=da0d472a794a7f39307707da93c39f216f88e5bb;p=modules%2Fsuperv.git

Fix for bug IPAL11408 : "Run()" method returns "0" after trying to start graph execution.
---

diff --git a/src/Supervision/Graph_Impl.cxx b/src/Supervision/Graph_Impl.cxx
index dd0df07..cc57cb5 100644
--- a/src/Supervision/Graph_Impl.cxx
+++ b/src/Supervision/Graph_Impl.cxx
@@ -1764,6 +1764,13 @@ long Graph_Impl::SuspendedThreads() {
 
 bool Graph_Impl::Begin() {
   bool RetVal = false ;
+  // mkr : IPAL11408 : "Run()" method returns "0" after trying to start graph execution.
+  // If we call Run() method from Python script two times for one graph, we have to be sure,
+  // that first execution of this graph ended (i.e. GraphEditor::DataFlow::Editing() method was called).
+  // Otherwise, the second call of Run() method will return 0. Therefore, we have to waiting for
+  // the first execution finish. 
+  // NB! From GUI we cann't run dataflow if its previous execution not finished.
+  while ( !DataFlowEditor()->IsEditing() ) {} // waiting for the previous execution of this graph ended
   if ( DataFlowEditor()->IsEditing() ) {
     if ( pthread_mutex_lock( &_MutexExecutorWait ) ) {
       perror("pthread_mutex_lock _MutexExecutorWait") ;