mikejohnson51 / HydroData

An R 📦 for finding and getting geospatial earth systems data
https://mikejohnson51.github.io/HydroData/
MIT License
41 stars 5 forks source link

Multiple functions break because AOI::getBoundingBox is not found #30

Open jsta opened 4 years ago

jsta commented 4 years ago

Looks like it has been replaced by AOI::bbox_get

mikejohnson51 commented 4 years ago

Hi,

Yes, after a fantastic code review of AOI by @dblodgett-usgs I refined some function names and scope and haven't got around to fixing the packages relying on it. It is on my to-do list and I would welcome any contribution given your great R talents,

Mike

jsta commented 4 years ago

In process on this. But I'm stumped on AOI::aoiProj. What does that command map to in the current version?

mikejohnson51 commented 4 years ago

That was a naive (bad) way of storing EPSG 4326 with the corresponding proj4 string. I think they should all be replaceable by 4326.

Thanks!

mikejohnson51 commented 4 years ago

Sooo... I am super embarrassed now ... I fixed most of these issues and never pushed to GitHub 🤦 The code could still use a good amount of work/review and I hope to eventually write it up as a publication (maybe EMS) with some compelling use cases. Would you be interested in pushing it in that direction as a collaboration?

I will get that push up tonight/tomorrow morning...

Again, I feel so bad and hope you didn't spend to much time on this!

jsta commented 4 years ago

No problem I didn't spend too much time. Yes, I'd be interested in helping out with a paper effort. My latest interest in the package is the wrapper to NLDI so maybe that's an area where I could contribute a use case..

dblodgett-usgs commented 4 years ago

Just lurking here -- @ldecicco-USGS and I talked about moving an NLDI wrapper client from nhdplusTools out to dataRetrieval. Maybe a little collaboration to make sure it's good for all the use cases we have would be in order...

mikejohnson51 commented 4 years ago

@dblodgett-usgs yeah what you raise is part of the larger point where I stopped focusing on this package - with the intention of coming back... The idea of a single data aggregator was getting to be heavy on maintenance and better things were being done in packages like nhdplusTools. Mostly I was waiting for dust to settle :)

That said, perhaps part of our next talk can be figuring out where nhd "discovery" tools belong. Some really useful ones sit in nhdplusTools, here, and in nwmHistoric. Knowing where they go and consolidating them would be great!

The one aspect I like about this wrapper is that it builds out many NLDI requests and returns a list of spatial (sf) objects. I also like the generalization (in a naive sense) of the feature types (e.g. just the USGS siteID rather then "USGS-#")