Feature #1129
Generic Timeout for CoCoA-5?
Description
CoCoALib has a generic "timeout" facility (via objects of type CpuTimeLimit).
Can something easy-to-use be ported into CoCoA-5?
History
#1
Updated by John Abbott over 6 years ago
Updated by - Status changed from New to In Progress
- % Done changed from 0 to 10
In CoCoALib I have used the RAII "design pattern": the dtor for a CpuTimeLimit object stops the "time bomb".
So C++ scoping is the usual way of saying when to stop monitoring CPU time.
How to achieve something reasonable in CoCoA-5?
One simplistic possibility is
TimerVariable := CreateNewCpuTimeLimit(2.0); RGB := ReducedGBasis(I); TimerVariable := 0; // to stop monitoring CPU time
The point being that assigning anything to TimerVariable effectively destroys the old value (which I had presumed would be a CpuTimeLimit object). However it is a bit unnatural... perhaps it is better than nothing?
#2
Updated by Anna Maria Bigatti over 6 years ago
Updated by John Abbott wrote:
In CoCoALib I have used the RAII "design pattern": the dtor for a
CpuTimeLimitobject stops the "time bomb".
So C++ scoping is the usual way of saying when to stop monitoring CPU time.How to achieve something reasonable in CoCoA-5?
One simplistic possibility is
[...]The point being that assigning anything to
TimerVariableeffectively destroys the old value (which I had presumed would be aCpuTimeLimitobject).
What about
SetTimeout(2.0); RGB := ReducedGBasis(I); UnsetTimeout(); // to stop monitoring CPU time
?
#3
Updated by John Abbott over 6 years ago
The interface SetTimeout and UnsetTimeout is simple to understand. BUT it does imply use of a global variable -- not too much of a problem until CoCoA-5 becomes multi-threaded. It is a little disappointing to have to add 2 functions.
My suggestion is less natural (esp. assigning any value to TimerVar as the means to disable the timeout).
In theory my suggestion should allow "nested" timeouts so long as functions do not use TopLevel variables for storing the timer. Note that nested timeouts could happen easily if several functions which call each other have their own timeouts. Anna's suggestion would not work well with nested timeouts (at least I do not see how to make it work).