← Back to team overview

gtg-contributors team mailing list archive

Refactorisation progress : an indicible truth?

 

Hello everybody,

Good news first: 

My refactorisation did a huge step today by removing a huge performance
lock. It doesn't mean that we don't have to optimize anymore but that
this refactorisation will improve **a lot** the performances of GTG.


Bad news: 

I discovered that our tagtree, that I didn't want to touch, was, in
facto, completely broken. I mean, even the basic asumption in it were
completely wrong (the get_path method was always giving an invalid
path). I don't explain how but this remained hidden because of the way
tasks were slowly added. With the refactorisation, it became apparent,
thus more work for me.

I also discovered that the browser was even more the heart of the
application than previously thought. Did you know that the browser was
responsible to inform tagstore and taskstore were tags and tasks were
modified ? That tags were created in the browser ? That all filtering
logic was in the browser.

Instead of having a frontend showing a backend state, we had a frontend
controlling a backend and displaying what it wanted.


Conclusion:

Each time I'm solving something, I discover something else, it makes
progress really slow :-(


Lionel




Follow ups