|Anonymous | Login||10-23-2021 14:02 UTC|
|Main | My View | View Issues | Change Log | Docs|
|Viewing Issue Simple Details [ Jump to Notes ]||[ View Advanced ] [ Issue History ] [ Print ]|
|ID||Category||Severity||Reproducibility||Date Submitted||Last Update|
|0007111||[Squeak] SUnit||feature||N/A||07-06-08 00:31||07-10-08 10:07|
|Reporter||nicolas cellier||View Status||public|
|Summary||0007111: [RFE] SUnit support for testing infinite recursion loop|
I propose to add following extension:
self should: [myObject doSomething] notSendDeeperThan: 10000.
I propose a MessageTally-spy-like solution to this.
I'm still interested if anyone finds a way to assert infinite ITERATIVE loops (whileTrue and whileFalse...)
Note that this solution does not handle correctly an Error...
An Error would pass...
Up to Block/Exception gurus to give their advice.
SUnit-infiniteRecursionSupport-M7111-nice.1.cs [^] (2,191 bytes) 07-06-08 00:38
SUnit-infiniteRecursionSupport-Test-M7111-nice.1.cs [^] (2,092 bytes) 07-06-08 00:39
(0012352 - 516 - 564 - 766 - 766 - 766 - 766)
Since squeak-dev thread tend to show i was misunderstood, I add a few complements:
My aim is not to solve the halting problem! This is not possible.
I don't want to fight against maths, logic nor Godel.
I just want to be able to count loops so that I can provide a hint and decide to stop the test.
Monitoring resources consumption (CPU and memory) at an upper level would be of greater interest than this partial support...
See http://lists.squeakfoundation.org/pipermail/squeak-dev/2008-July/130058.html [^]
|07-06-08 00:31||nicolas cellier||New Issue|
|07-06-08 00:38||nicolas cellier||File Added: SUnit-infiniteRecursionSupport-M7111-nice.1.cs|
|07-06-08 00:39||nicolas cellier||File Added: SUnit-infiniteRecursionSupport-Test-M7111-nice.1.cs|
|07-10-08 10:07||nicolas cellier||Note Added: 0012352|
| Mantis 1.0.8[^]
Copyright © 2000 - 2007 Mantis Group
35 total queries executed.|
26 unique queries executed.