profile_sector() now handles case "c"
#282
Conversation
| mutate(scenario = recode(.data$scenario, "1.5c rps" = "IPR 1.5c RPS", "nz 2050" = "WEO NZ 2050")) |> | ||
| mutate(scenario = ifelse(is.na(scenario), grouped_by, scenario)) |> | ||
| mutate(scenario = recode_scenario(.data$scenario)) |> |
There was a problem hiding this comment.
Here I write a custom helper to recode scenario in a more robust way. It may be wrong but the idea may be useful. I'll discuss in the helper itself.
| add_avg_matching_certainty("completion") |> | ||
| exclude_rows("risk_category") |
There was a problem hiding this comment.
Here is the main "bug".
This line removes rows where risk_category is NA and therefore drops the rows we would want to output.
| recode_scenario <- function(x) { | ||
| out <- gsub("^ipr|^weo", "", x, ignore.case = TRUE) | ||
| out <- gsub("_", " ", out) | ||
| tolower(trimws(out)) | ||
| } |
There was a problem hiding this comment.
This helper aims to recode the values of scenario in a more "programatic" way -- trying to avoid using explicit strings and letter-case. The exception here is "ipr" and "weo" -- which we could also eliminate by rewriting the code so that it removes the first word (whatever that is).
If you keep this function do test it because a test suggest we may want something like "ns 2050" but instead we get "ns 2050 2030".
| @@ -271,3 +271,37 @@ test_that("outputs `profile_ranking_avg` at company level", { | |||
| company <- unnest_company(out) | |||
| expect_true(hasName(company, "reduction_targets_avg")) | |||
| }) | |||
|
|
|||
| test_that("given a product in 'ipr', when scenarios has also 'weo', then the product-result includes a 'weo*' `scenario` and maps to `NA` in `sector_profile` (#279, tiltIndicator#739)", { | |||
There was a problem hiding this comment.
Here's the test for the specific case associated to this issue. Meeting this expectation breaks other tests so the job is incomplete.
profile_sector()fails to handle case "c" #279TODO