Closed hernanmd closed 2 months ago
Thank you Hernan.
However I would not like to do it like that, we spent a lot of time removing similar code from the debugger. It solves the problem (I think) but it only add unnecessary complexity to the debugger, using globals etc.
I would like to find time to discuss the solution, for instance to move all settings of the debugger into a single "debugger setting" object instead of class variables.
Please also come to the debugging team to discuss before implementing without us knowing, because we can share a bit of vision and technical constraints and objectives.
@adri09070 can you have a look?
Thank you Hernan.
However I would not like to do it like that, we spent a lot of time removing similar code from the debugger. It solves the problem (I think) but it only add unnecessary complexity to the debugger, using globals etc.
I would like to find time to discuss the solution, for instance to move all settings of the debugger into a single "debugger setting" object instead of class variables.
Please also come to the debugging team to discuss before implementing without us knowing, because we can share a bit of vision and technical constraints and objectives.
Hi Steven,
I've updated the PR with a cleaner StDebugger "remember position" implementation by moving the settings methods to a new class, StDebuggerSettings. Please let me know of any appropriate enhancements.
I've fixed some additional issues, so the CI should pass and can be reviewed "safely"
In addition to my comments, we need tests ensuring the new introduced behavior.
Done in the last PR commit.
Guys, yesterday this PR was green and today I've seen some conflicts so it would be nice to have it integrated on time before other PR's make it obsolete, otherwise, I have to work on this forever.
Now I'm checking this again because StDebugger initializeWindow: was changed somewhere.
The failing test is unrelated.
Guys, yesterday this PR was green and today I've seen some conflicts so it would be nice to have it integrated on time before other PR's make it obsolete, otherwise, I have to work on this forever.
Now I'm checking this again because StDebugger initializeWindow: was changed somewhere.
Done. Could you have a look @StevenCostiou ?
Perhaps tomorrow, I am not available today. But worry not, the debugger code does not change that often, that is just bad luck.
For me now there are two problems left:
Concerning the failing tests, they are not related
For me now there are two problems left:
* what about those nil checks? can we do something? * tests are all red (I didn't check what's happening) If those are fixed, can we agree to merge?
I removed the ifNil: checks since I need to work on other issues.
Add lastKnownExtent and lastKnownPosition to preserve the last extent and position for the next debug request window. Add a setting to enable this behavior (enabled by default after this PR).