Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update loanbook demo NACE codes #338

Merged
merged 5 commits into from
Feb 15, 2024
+94 −34
Merged

Update loanbook demo NACE codes #338

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
880ddf6
add failing test
jdhoffa Feb 15, 2024
5c17f7c
Functionalize NACE conversion
jdhoffa Feb 15, 2024
b2262e9
Run NACE conversion on loanbook_demo
jdhoffa Feb 15, 2024
2fa52de
Update snaps
jdhoffa Feb 15, 2024
e43555a
Reformat loanbook_demo sector class codes
jdhoffa Feb 15, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 5 additions & 32 deletions data-raw/classification_bridge.R
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,38 +7,11 @@ nace_classification_raw <- read_bridge(
file.path("data-raw", "nace_classification.csv")
)

nace_classification <- nace_classification_raw %>%
mutate(
prepend_value = case_when(
original_code %in% LETTERS ~ "",
trunc(as.numeric(original_code)) %in% seq(1, 3) ~ "A",
trunc(as.numeric(original_code)) %in% seq(5, 9) ~ "B",
trunc(as.numeric(original_code)) %in% seq(10, 33) ~ "C",
trunc(as.numeric(original_code)) == 35 ~ "D",
trunc(as.numeric(original_code)) %in% seq(36, 39) ~ "E",
trunc(as.numeric(original_code)) %in% seq(41, 43) ~ "F",
trunc(as.numeric(original_code)) %in% seq(45, 47) ~ "G",
trunc(as.numeric(original_code)) %in% seq(49, 53) ~ "H",
trunc(as.numeric(original_code)) %in% seq(55, 56) ~ "I",
trunc(as.numeric(original_code)) %in% seq(58, 63) ~ "J",
trunc(as.numeric(original_code)) %in% seq(64, 66) ~ "K",
trunc(as.numeric(original_code)) == 68 ~ "L",
trunc(as.numeric(original_code)) %in% seq(69, 75) ~ "M",
trunc(as.numeric(original_code)) %in% seq(77, 82) ~ "N",
trunc(as.numeric(original_code)) == 84 ~ "O",
trunc(as.numeric(original_code)) == 85 ~ "P",
trunc(as.numeric(original_code)) %in% seq(86, 88) ~ "Q",
trunc(as.numeric(original_code)) %in% seq(90, 93) ~ "R",
trunc(as.numeric(original_code)) %in% seq(94, 96) ~ "S",
trunc(as.numeric(original_code)) %in% seq(97, 98) ~ "T",
trunc(as.numeric(original_code)) == 99 ~ "U",
TRUE ~ "Z" #debug value, see unit tests)
)
) %>%
mutate(
code = paste0(prepend_value, original_code),
prepend_value = NULL
)
nace_classification <- convert_superseded_nace_code(
nace_classification_raw,
col_from = "original_code",
col_to = "code"
)

use_data(nace_classification, overwrite = TRUE)

Expand Down
24 changes: 24 additions & 0 deletions data-raw/loanbook_demo.R
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,4 +48,28 @@ loanbook_demo <- loanbook_demo %>%
)
)

nace_classification_raw <- read_bridge(
file.path("data-raw", "nace_classification.csv")
)

# this gets the `original_code` back to the `loanbook_demo` dataset
loanbook_demo <- mutate(
loanbook_demo,
sector_classification_direct_loantaker = as.character(sector_classification_direct_loantaker)
)

loanbook_demo <- left_join(
loanbook_demo,
select(nace_classification_raw, original_code, code),
by = c("sector_classification_direct_loantaker" = "code")
)

loanbook_demo <- convert_superseded_nace_code(
loanbook_demo,
col_from = "original_code",
col_to = "sector_classification_direct_loantaker"
)

loanbook_demo <- mutate(loanbook_demo, original_code = NULL)

usethis::use_data(loanbook_demo, overwrite = TRUE)
39 changes: 39 additions & 0 deletions data-raw/utils.R
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -83,3 +83,42 @@ abort_missing_names <- function(missing_names) {
)
)
}

convert_superseded_nace_code <- function(data,
col_from,
col_to) {
data <- mutate(
data,
prepend_value = case_when(
.data[[col_from]] %in% LETTERS ~ "",
trunc(as.numeric(.data[[col_from]])) %in% seq(1, 3) ~ "A",
trunc(as.numeric(.data[[col_from]])) %in% seq(5, 9) ~ "B",
trunc(as.numeric(.data[[col_from]])) %in% seq(10, 33) ~ "C",
trunc(as.numeric(.data[[col_from]])) == 35 ~ "D",
trunc(as.numeric(.data[[col_from]])) %in% seq(36, 39) ~ "E",
trunc(as.numeric(.data[[col_from]])) %in% seq(41, 43) ~ "F",
trunc(as.numeric(.data[[col_from]])) %in% seq(45, 47) ~ "G",
trunc(as.numeric(.data[[col_from]])) %in% seq(49, 53) ~ "H",
trunc(as.numeric(.data[[col_from]])) %in% seq(55, 56) ~ "I",
trunc(as.numeric(.data[[col_from]])) %in% seq(58, 63) ~ "J",
trunc(as.numeric(.data[[col_from]])) %in% seq(64, 66) ~ "K",
trunc(as.numeric(.data[[col_from]])) == 68 ~ "L",
trunc(as.numeric(.data[[col_from]])) %in% seq(69, 75) ~ "M",
trunc(as.numeric(.data[[col_from]])) %in% seq(77, 82) ~ "N",
trunc(as.numeric(.data[[col_from]])) == 84 ~ "O",
trunc(as.numeric(.data[[col_from]])) == 85 ~ "P",
trunc(as.numeric(.data[[col_from]])) %in% seq(86, 88) ~ "Q",
trunc(as.numeric(.data[[col_from]])) %in% seq(90, 93) ~ "R",
trunc(as.numeric(.data[[col_from]])) %in% seq(94, 96) ~ "S",
trunc(as.numeric(.data[[col_from]])) %in% seq(97, 98) ~ "T",
trunc(as.numeric(.data[[col_from]])) == 99 ~ "U",
TRUE ~ "Z" #debug value, see unit tests)
)
)

data <- mutate(
data,
{{col_to}} := paste0(prepend_value, .data[[col_from]]),
prepend_value = NULL
)
}
Binary file modified data/loanbook_demo.rda
View file
Open in desktop
Binary file not shown.
4 changes: 2 additions & 2 deletions tests/testthat/_snaps/loanbook_demo.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -86,9 +86,9 @@
"value": ["Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code", "Code"]
},
{
"type": "double",
"type": "character",
"attributes": {},
"value": [3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3511, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 3011, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 291, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 2351, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 241, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 610, 2910, 2910, 5110, 5110, 5110, 5110, 3030, 2910, 5200, 2410]
"value": ["D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "D35.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C30.11", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C29.1", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C23.51", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "C24.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "B6.1", "C29.10", "C29.10", "H51.1", "H51.1", "H51.1", "H51.1", "C30.3", "C29.10", "B5.2", "C24.1"]
},
{
"type": "character",
Expand Down
24 changes: 24 additions & 0 deletions tests/testthat/test-loanbook_demo.R
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,3 +1,27 @@
test_that("hasn't changed", {
expect_snapshot_value(loanbook_demo, style = "json2")
})

test_that("loanbook_demo has at least one valid sector classification (#336)", {
#base R left_join
out <- merge(
loanbook_demo,
sector_classifications,
by.x = c(
"sector_classification_system",
"sector_classification_direct_loantaker"
),
by.y = c(
"code_system",
"code"
),
all.x = TRUE
)

number_of_sector_matches <- nrow(
out[!is.na(out[["sector"]]), ]
)

expect_true(number_of_sector_matches > 0)

})
Loading