Use Unicode.normalize in _preprocess_name

TheRoniOne · Apr 7, 2024 · 222e1fb · 222e1fb
1 parent cd3469b
commit 222e1fb
Show file tree

Hide file tree

Showing 2 changed files with 5 additions and 2 deletions.
diff --git a/Project.toml b/Project.toml
@@ -6,8 +6,10 @@ version = "1.0.6"
 [deps]
 PrettyTables = "08abe8d2-0d0c-5749-adfa-8a2ac140af0d"
 Tables = "bd369af6-aec1-5ad0-b16a-f7cc5008161c"
+Unicode = "4ec0a83e-493e-50e2-b9ac-8f72acf5a8f5"
 
 [compat]
 PrettyTables = "1, 2"
 Tables = "1"
+Unicode = "1"
 julia = "1.6"
diff --git a/src/polish_names.jl b/src/polish_names.jl
@@ -1,10 +1,11 @@
 using Base: String
+using Unicode: normalize
 
 struct Style{T} end
 
 Style(s::Symbol) = Style{s}()
 
-const SPECIAL_CHARS = r"[\s\-\.\_\/\:\\\*\?\"\'\>\<\|]"
+const SPECIAL_CHARS = r"[\s\-\.\_\/\:\\\*\?\"\'\>\<\|\!\,\$\@\^\[\]\{\}]"
 
 """
     polish_names!(table::CleanTable; style::Symbol=:snake_case)
@@ -78,7 +79,7 @@ function generate_polished_names(names, ::Style)
 end
 
 function _preprocess_name(name)
-    preprocessed = String(name)
+    preprocessed = normalize(String(name), stripmark=true)
 
     matched = match(r"^[[:upper:]]+$", preprocessed)
     if matched !== nothing