Michael Chirico

relevant issue https://github.com/Rdatatable/data.table/issues/2634

Hi Roger, much appreciate your reaching out and pointing to all the resources. However, I feel a bit overwhelmed by it all and am struggling to turn this into action...

@Nowosad great! Is this hosted on dockerhub?

confirming that the error is consistent across illegal inputs: ``` gh_decode(c('a','i','l','o')) # lat lng lat_error lng_error # 1 -67.5 22.5 22.5 22.5 # 2 -67.5 22.5 22.5 22.5 # 3...

I have the following test built into a function wrapping `gh_decode` as a stop-gap: ``` if (any(idx

I don't know how this can produce anything but NA. 0,0 is demonstrably incorrect. On Fri, Jun 1, 2018, 1:36 AM Seth Wenchel wrote: > i have a patch that...

according to the comment you reference (with which I pretty heavily disagree -- for example, any user is free to set warnings-as-errors through options(warn = 2)), the preferred behavior is...

I'm just noticing the `gh_neighbour` internal function... with that, the names of the neighbors are well-defined, i.e. `geohash:::gh_neighbour('dqcjqc', c(1, -2))` gives `dqcjq6` unambiguously. Corollary: - This could be used to...

precisely! and grey/gray On Feb 6, 2018 8:43 AM, "Os Keyes" wrote: > Works for me! See colors()/colours() > > — > You are receiving this because you authored the...

Bumping here, it's still a pretty common issue to wind up with dim-0 results and get spurious warnings. Any thoughts on the right solution @flying-sheep? Happy to file a PR...