dod.met() is broken - Githubissues

dankelley commented 1 year ago

See https://github.com/dankelley/oce/issues/2151 but note that I plan to fix it here, in dod, porting it back to oce only if I find it works well.

Another approach, which I find quite appealing, is to "deprecated" the oce version. I just don't like having a function that requires major changes on a year or even sub-year basis, and that is what happens with met data from Environment Canada (or whatever it names itself this week).

dankelley commented 1 year ago

I'm seeing the same thing with Gavin Simpson's canadaHCD package. Actually, it doesn't work because it has some errors that relate, I think, to changes in tibble-related code (see below), but I went into his code to see the URL, and when I download that (this time as CSV) I get headings but no data. Clearly env-can has -- yet again -- changed the format.

> library(canadaHCD)
> find_station("Halifax")
# A tibble: 13 × 5
   Name                      Province         StationID LatitudeDD LongitudeDD
   <fct>                     <fct>            <fct>          <dbl>       <dbl>
 1 BABINE LAKE HALIFAX BAY   British Columbia 455             55.0      -126. 
 2 HALIFAX                   Nova Scotia      6355            44.6       -63.6
 3 HALIFAX                   Nova Scotia      6356            44.6       -63.6
 4 HALIFAX CITADEL           Nova Scotia      6357            44.6       -63.6
 5 HALIFAX COMMONS           Nova Scotia      49128           44.6       -63.6
 6 HALIFAX DOCKYARD          Nova Scotia      43405           44.7       -63.6
 7 HALIFAX INTL A            Nova Scotia      53938           44.9       -63.5
 8 HALIFAX STANFIELD INT'L A Nova Scotia      6358            44.9       -63.5
 9 HALIFAX INTL A            Nova Scotia      50620           44.9       -63.5
10 HALIFAX KOOTENAY          Nova Scotia      43124           44.6       -63.6
11 HALIFAX WINDSOR PARK      Nova Scotia      43403           44.7       -63.6
12 HALIFAX RADAR             Nova Scotia      9030            44.9       -63.5
13 HALIFAX METOC             Nova Scotia      10879           44.7       -63.6
> d <- hcd_hourly(6357,2023,09)
Rows: 720 Columns: 30                                                                                    
── Column specification ──────────────────────────────────────────────
Delimiter: ","
chr   (3): Station Name, Month, Day
dbl   (4): Longitude (x), Latitude (y), Climate ID, Year
lgl  (21): Temp (°C), Temp Flag, Dew Point Temp (°C), Dew Point Te...
dttm  (1): Date/Time (LST)
time  (1): Time (LST)

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Error:
! Names repair functions can't return `NA` values.
Run `rlang::last_trace()` to see where the error occurred.

dankelley commented 1 year ago

It's not broken, actually. The station codes have changed, I guess. If I use e.g.

library(dod)
metFile <- dod.met(43405, 2023, 9, age=1, type="csv")

it works. (That is the Halifax Dockyard dataset.)

Maybe the code number for Halifax Stanfield has changed, etc. You'd think I could check on that by going to https://weather.gc.ca/city/pages/ns-19_metric_e.html and I do see a link there that brings me to https://climate.weather.gc.ca/climate_data/daily_data_e.html?StationID=50620 so I guess the code for the airport is now 50620 (which is number 9 in the list from canadaHCD::find_station().

Actually, there are 3 items in that list that are for the airport. Maybe when they change equipment they change the number. Or maybe they keep moving the station? I've no idea, but I do know that this issue -- as defined by its title -- is ready to be closed.

dankelley / dod

dod.met() is broken #7