handle CSV files with odd header rows, closes #94

NEWS.md

@@ -1,3 +1,7 @@
+# statnipokladna (development version)
+
+* `sp_get_table()` can now handle CSV files with non-standard header rows in the format '"Výkaz"ZC_VYKAZ:ZC_VYKAZ"', fixes #94
+
 # statnipokladna 0.7.2
 
 # statnipokladna 0.7.1

R/tables.R

@@ -106,12 +106,20 @@ sp_load_table <- function(path, ico = NULL) {
 
   suppressWarnings(suppressMessages(
     dt <- readr::read_csv2(path, col_types = readr::cols(.default = readr::col_character()))))
+
+  dt_new_names <- stringr::str_remove_all(names(dt), "\"[A-\\u017da-\\u017e\\s\\-\\./]*\"") |>
+    stringr::str_remove_all("/BIC/")
+
+  dt <- dt %>%
+    purrr::set_names(dt_new_names)
+
   # print(head(dt))
   if(max(stringr::str_length(dt$`ZC_ICO:ZC_ICO`), na.rm = TRUE) == 10) {
     dt <- dplyr::mutate(dt, `ZC_ICO:ZC_ICO` = stringr::str_sub(.data$`ZC_ICO:ZC_ICO`, 3, 10))
   }
   if(!is.null(ico)) dt <- dt[dt$`ZC_ICO:ZC_ICO` %in% ico,]
-  dt <- dt %>%
+
+  dt <- dt |>
     purrr::set_names(stringr::str_remove(names(dt), "^[A-Z_0-9/]*:")) %>%
     dplyr::mutate_at(dplyr::vars(dplyr::starts_with("ZU_")), ~switch_minus(.) %>% as.numeric(.)) %>%
     tidyr::extract(.data$`0FISCPER`, c("vykaz_year", "vykaz_month"), "([0-9]{4})0([0-9]{2})") %>%

tests/testthat/test-tables.R

@@ -9,3 +9,7 @@ test_that("List of tables make sense", {
 test_that("get_table works", {
   expect_error(sp_get_table("blah", "2012", "12"))
 })
+
+test_that("get_table works with newer files", {
+  expect_gt(nrow(sp_get_table("budget-local", "2022", "12")), 1000)
+})