⚡ Optimize res_to_m with vectorized regex

e-kotov · e-kotov · commit c84f9acadcc9 · 2026-01-16T09:57:31.000+01:00
diff --git a/R/inspire_id_to_coords.R b/R/inspire_id_to_coords.R
@@ -174,9 +174,11 @@ determine_final_crs <- function(parsed_df, user_crs) {
 }
 
 res_to_m <- function(res) {
-  is_km <- grepl("(?<=[0-9])km", res, perl = TRUE)
-  numbers <- as.numeric(regex_match(res, "^[0-9]+", i = 1))
-  numbers <- ifelse(is_km, numbers * 1000, numbers)
+  is_km <- endsWith(res, "km")
+  numbers <- as.numeric(sub("k?m$", "", res))
+  # which() handles NAs safely (drops them)
+  idx_km <- which(is_km)
+  numbers[idx_km] <- numbers[idx_km] * 1000
   numbers
 }
 

Original file line number	Diff line number	Diff line change
`@@ -174,9 +174,11 @@ determine_final_crs <- function(parsed_df, user_crs) {`
`174`	`174`	`}`
`175`	`175`
`176`	`176`	`res_to_m <- function(res) {`
`177`		`- is_km <- grepl("(?<=[0-9])km", res, perl = TRUE)`
`178`		`- numbers <- as.numeric(regex_match(res, "^[0-9]+", i = 1))`
`179`		`- numbers <- ifelse(is_km, numbers * 1000, numbers)`
	`177`	`+ is_km <- endsWith(res, "km")`
	`178`	`+ numbers <- as.numeric(sub("k?m$", "", res))`
	`179`	`+ # which() handles NAs safely (drops them)`
	`180`	`+ idx_km <- which(is_km)`
	`181`	`+ numbers[idx_km] <- numbers[idx_km] * 1000`
`180`	`182`	`numbers`
`181`	`183`	`}`
`182`	`184`