Implement hashing for many basic enums #164
Merged
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I don't have a use for implementing Hash for Trips or StopTimes yet, nor a good workaround for
Option<f32>
types.I'm working on compression of schedules and I think at least the enums should implement hashing to allow direct comparison of time intervals. Most implementions of schedule parsers, such as Google Maps, likely do not store the entire dataset. I attempted to do that and it wound up being over 200 GB for the entire world. Implementations typically store the difference between each stop as an itinerary, and then trips are referenced via an offset to the itinerary. I can copy over the pickup and dropoff fields, and timepoint into my own struct and compute a hash of it, which becomes the Itinerary, then use it to match trip ids with a previous itinerary, along with an offset.
This paper describes it quite well: https://ad-publications.informatik.uni-freiburg.de/SIGSPATIAL_Frequency_based_Search_BBS_2014.pdf
Like my previous
Clone
pull request, this does increase compile time but is not a breaking change, nor does it worsen runtime.Thank you so much once again,
Kyler Chin