frecency: Boost interactive selection frecency #114
Conversation
|
I should have commented, but I'm leaving this unmerged pending one integ test. And for that integ test, we need to support |
This gives a 3x weighted boost to the result of an interactive selection on the assumption that if the user went out of their way to do an interactive selection to choose something it's something they wish to visit more often. This does the necessary plumbing to implement the feature, and implements it.
This adds the needed testing apparatus for asserting the output of z -i and hopefully lays the groundwork for more tests of the like. It also adds an 'interactive list' test to start things off.
euank
force-pushed
the
tmp/pr/pazi-jump-bump
branch
from
f528d4c to
b7eedfb
Compare
I wrote tests for it, they didn't work, and I realized that's because I reworked main such that the jump code that's actually in use is elsewhere. So yeah, the tests caught that, woo!
|
It has been over a year, and I have come back to this! |
|
Self-reviewed; this still looks good. bors r+ |
114: frecency: Boost interactive selection frecency r=euank a=euank This gives a 3x weighted boost to the result of an interactive selection on the assumption that if the user went out of their way to do an interactive selection to choose something it's something they wish to visit more often. This does the necessary plumbing to implement the feature, and implements it. Thank you @lclarkmichalek for the idea! Co-authored-by: Euan Kemp <[email protected]>
|
Build failed: |
|
Hm. Looks like one of the new tests flakes on travis-ci sometimes (hangs); it doesn't trivially reproduce locally, so I'll have to dig into that. |
This gives a 3x weighted boost to the result of an interactive selection
on the assumption that if the user went out of their way to do an
interactive selection to choose something it's something they wish to
visit more often.
This does the necessary plumbing to implement the feature, and
implements it.
Thank you @lclarkmichalek for the idea!