From a3f9b1e8c3540b3c323a71874a69068c6acb855a Mon Sep 17 00:00:00 2001
From: AurumTheEnd <47597303+aurumtheend@users.noreply.github.com>
Date: Fri, 17 May 2024 15:41:27 +0200
Subject: [PATCH] refactor(parser): made PATTERN_SET regex lazy_static

---
 src/parser/structs/intermediate_token.rs | 26 +++++-------------------
 src/parser/tokenize.rs                   |  4 ++--
 src/parser/utils/mod.rs                  |  2 +-
 src/parser/utils/regex.rs                | 21 +++++++++++++++++--
 4 files changed, 27 insertions(+), 26 deletions(-)

diff --git a/src/parser/structs/intermediate_token.rs b/src/parser/structs/intermediate_token.rs
index f6ca202..fab2c24 100644
--- a/src/parser/structs/intermediate_token.rs
+++ b/src/parser/structs/intermediate_token.rs
@@ -1,6 +1,4 @@
-use regex::RegexSet;
-
-use crate::parser::utils::LITERAL_IDENTIFIER;
+use crate::parser::utils::PATTERN_SET;
 
 #[derive(PartialEq, Debug)]
 pub enum IntermediateToken<'a> {
@@ -136,26 +134,12 @@ impl<'a> IntermediateToken<'a> {
 
     // TODO make a trait method
     pub fn try_from(value: &'a str) -> Option<IntermediateToken> {
-        let input = Self::ALL_TOKEN_PATTERNS_FROM_LONGEST;
-
-        // escape the pattern so that e.g. "^" is not treated as regex, but as a literal character for the And operation
-        let set = RegexSet::new(input.iter().map(|pattern| {
-            format!(
-                r"(?i)^{}{}",
-                regex::escape(pattern),
-                if LITERAL_IDENTIFIER.is_match(pattern) {
-                    "([^-_a-zA-Z0-9]|$)"
-                } else {
-                    ""
-                }
-            )
-        }))
-        .unwrap();
-
-        let pattern_or_no_match = set
+        let patterns = Self::ALL_TOKEN_PATTERNS_FROM_LONGEST;
+
+        let pattern_or_no_match = PATTERN_SET
             .matches(value)
             .into_iter()
-            .map(|index| &input[index])
+            .map(|index| &patterns[index])
             .next();
 
         pattern_or_no_match.map(|value| Self::from(value))
diff --git a/src/parser/tokenize.rs b/src/parser/tokenize.rs
index ae9cb8f..d9a01b1 100644
--- a/src/parser/tokenize.rs
+++ b/src/parser/tokenize.rs
@@ -183,7 +183,7 @@ mod tests {
     };
     use crate::parser::error::EOL_VICINITY;
     use crate::parser::structs::FinalToken::*;
-    use crate::parser::utils::LITERAL_IDENTIFIER;
+    use regex::Regex;
 
     use super::*;
 
@@ -233,7 +233,7 @@ mod tests {
 
         // test sanity
         assert!(!all_tokens().contains(input));
-        assert!(!LITERAL_IDENTIFIER.is_match(input));
+        assert!(!Regex::new(r"[-_a-zA-Z0-9]+").unwrap().is_match(input));
 
         let actual = tokenize(input);
         let expected_err = UnknownSymbolError {
diff --git a/src/parser/utils/mod.rs b/src/parser/utils/mod.rs
index d0bef93..4963c54 100644
--- a/src/parser/utils/mod.rs
+++ b/src/parser/utils/mod.rs
@@ -1,6 +1,6 @@
 pub use peek_until::peek_until_n;
 pub use pop::pop_n_left;
-pub use regex::{LITERAL_IDENTIFIER, SHOULD_END_LITERAL};
+pub use regex::{PATTERN_SET, SHOULD_END_LITERAL};
 pub use trim_whitespace::trim_whitespace_left;
 
 mod peek_until;
diff --git a/src/parser/utils/regex.rs b/src/parser/utils/regex.rs
index c4c830e..53afc91 100644
--- a/src/parser/utils/regex.rs
+++ b/src/parser/utils/regex.rs
@@ -1,6 +1,23 @@
-use regex::Regex;
+use crate::parser::structs::IntermediateToken;
+use regex::{Regex, RegexSet};
 
 lazy_static::lazy_static! {
     pub static ref SHOULD_END_LITERAL: Regex = Regex::new(r"[^-_a-zA-Z0-9]").unwrap();
-    pub static ref LITERAL_IDENTIFIER: Regex = Regex::new(r"[-_a-zA-Z0-9]+").unwrap();
+    static ref LITERAL_IDENTIFIER: Regex = Regex::new(r"[-_a-zA-Z0-9]+").unwrap();
+
+    pub static ref PATTERN_SET: RegexSet = RegexSet::new(IntermediateToken::ALL_TOKEN_PATTERNS_FROM_LONGEST
+        .iter()
+        .map(|pattern| {
+            format!(
+                r"(?i)^{}{}",
+                // escape the pattern so that e.g. "^" is not treated as regex, but as a literal character for the And operation
+                regex::escape(pattern),
+                if LITERAL_IDENTIFIER.is_match(pattern) {
+                    "([^-_a-zA-Z0-9]|$)"
+                } else {
+                    ""
+                }
+            )
+        }))
+        .unwrap();
 }