Difference between revisions of "Bugs"
From OdaWiki
(→How to report a bug) |
|||
| Line 1: | Line 1: | ||
| + | == Why bother? == | ||
| + | |||
| + | * Help odamex | ||
| + | * Improve your IT and testing skills | ||
| + | * You might get mentioned in the credits if your report was very helpful | ||
| + | |||
== How to report a bug == | == How to report a bug == | ||
The [http://odamex.net/bugs Odamex Bug Tracker] is an active list of issues. Look through the bugs to see if the bug you want to report is already listed. Please only post reproducible bugs, with steps to reproduce them. | The [http://odamex.net/bugs Odamex Bug Tracker] is an active list of issues. Look through the bugs to see if the bug you want to report is already listed. Please only post reproducible bugs, with steps to reproduce them. | ||
| + | |||
| + | If the bug is new and did not appear in a previous [[svn]] revision, it would also be helpful to determine which revision caused the bug. You may beed to do regression testing to work this out. | ||
| + | |||
| + | == Regression testing == | ||
| + | |||
| + | Find a revision from which the bug was absent. Then find a revision in which the bug was present. Pick a revision in the middle, and test that. Now you know which suspect block of revision to test next. Divide and conquer. A bug in a block of 1024 revisions should only take about 10 tests. | ||
Revision as of 14:16, 29 March 2006
Why bother?
- Help odamex
- Improve your IT and testing skills
- You might get mentioned in the credits if your report was very helpful
How to report a bug
The Odamex Bug Tracker is an active list of issues. Look through the bugs to see if the bug you want to report is already listed. Please only post reproducible bugs, with steps to reproduce them.
If the bug is new and did not appear in a previous svn revision, it would also be helpful to determine which revision caused the bug. You may beed to do regression testing to work this out.
Regression testing
Find a revision from which the bug was absent. Then find a revision in which the bug was present. Pick a revision in the middle, and test that. Now you know which suspect block of revision to test next. Divide and conquer. A bug in a block of 1024 revisions should only take about 10 tests.