Uh oh!
There was an error while loading. Please reload this page.
Sort colocated waypoint lists; close various minor issues - #144
Conversation
yakra
commented
Dec 10, 2018
yakra
commented
Dec 10, 2018
yakra
commented
Dec 10, 2018
yakra
commented
Dec 10, 2018
yakra
commented
Dec 10, 2018
That was the case for logs produced on noreaster. Not the case on my home machine. I've noticed differences in the way the FreeBSD and Ubuntu Python flavors iterate thru some container formats, such as dicts. I'll take a closer look at this later on. |
yakra
commented
Dec 10, 2018
yakra
commented
Dec 10, 2018
yakra
commented
Dec 13, 2018
I may have goofed in my fix for #134. Will look into this. Edit: I goofed. I'll test it out before committing, because it's now established that I can be careless. 8) |
done incorrectly in ed5e912
yakra
commented
Dec 14, 2018
Tested. Fixed in dc8a05c. |
jteresco
commented
Dec 20, 2018
Sorry I have not been following this one much. Is it ready to merge in and be used in production going forward? |
yakra
commented
Dec 20, 2018
It is. Currently testing out a datacheck for the OP in #57, which can be in a separate pull request. |
Deterministic concurrency detection portion. I've not yet addressed the per-traveler augments.
highwaydatastats.log:I didn't set out to make this deterministic, but 10 single-threaded trials & 10 multi-threaded trials have no DIFFs other than the initial
Travel Mapping highway mileage as of ...line. OK! :)This allows me to simplify the code for my earlier modifications to the VISIBLE_HIDDEN_COLOC and HIDDEN_JUNCTION datachecks.
Takes 1.6-1.7 seconds on my machine. Will be a little faster on Noreaster.
I think DIFFs are worthwhile. :)