From d8a9cc6428e76d1e075d58c88fa09f762bc3c3d1 Mon Sep 17 00:00:00 2001
From: "Azat S." <to@azat.io>
Date: Sun, 20 Oct 2024 21:09:56 +0300
Subject: [PATCH] fix: fix todo comments identification

---
 Cargo.lock                     |  1 +
 Cargo.toml                     |  1 +
 src/identify_supported_file.rs |  9 ++++++--
 src/identify_todo_comment.rs   | 41 ++++++++++++++++++++++------------
 tests/identify_todo_comment.rs | 24 +++++++++++++++++---
 5 files changed, 57 insertions(+), 19 deletions(-)

diff --git a/Cargo.lock b/Cargo.lock
index 13bf49a..35a12f6 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -1222,6 +1222,7 @@ dependencies = [
  "clap",
  "futures",
  "indicatif",
+ "lazy_static",
  "mockall",
  "open",
  "oxc",
diff --git a/Cargo.toml b/Cargo.toml
index 1f396c0..a1eff1f 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -8,6 +8,7 @@ chrono = { version = "0.4.38", features = ["serde"] }
 clap = { version = "4.5.20", features = ["derive"] }
 futures = "0.3.31"
 indicatif = "0.17.8"
+lazy_static = "1.5.0"
 open = "5.3.0"
 oxc = "0.31.0"
 regex = "1.11.0"
diff --git a/src/identify_supported_file.rs b/src/identify_supported_file.rs
index 1d8393c..1b1c7f3 100644
--- a/src/identify_supported_file.rs
+++ b/src/identify_supported_file.rs
@@ -1,6 +1,11 @@
+use lazy_static::lazy_static;
 use regex::Regex;
 
+lazy_static! {
+    static ref SUPPORTED_FILE_REGEX: Regex =
+        Regex::new(r"\.[cm]?[jt]sx?$").unwrap();
+}
+
 pub fn identify_supported_file(file_name: &str) -> bool {
-    let re = Regex::new(r"\.[cm]?[jt]sx?$").unwrap();
-    re.is_match(file_name)
+    SUPPORTED_FILE_REGEX.is_match(file_name)
 }
diff --git a/src/identify_todo_comment.rs b/src/identify_todo_comment.rs
index dce6011..3d68fae 100644
--- a/src/identify_todo_comment.rs
+++ b/src/identify_todo_comment.rs
@@ -1,9 +1,12 @@
-pub const PRIMARY_TODO_KEYWORDS: [&str; 9] = [
-    "TODO", "FIXME", "CHANGED", "XXX", "HACK", "BUG", "OPTIMIZE", "REFACTOR",
-    "TEMP",
+use lazy_static::lazy_static;
+use regex::Regex;
+
+pub const PRIMARY_TODO_KEYWORDS: [&str; 8] = [
+    "TODO", "FIXME", "XXX", "HACK", "BUG", "OPTIMIZE", "REFACTOR", "TEMP",
 ];
 
-pub const SECONDARY_TODO_KEYWORDS: [&str; 13] = [
+pub const SECONDARY_TODO_KEYWORDS: [&str; 14] = [
+    "CHANGED",
     "IDEA",
     "NOTE",
     "REVIEW",
@@ -19,23 +22,33 @@ pub const SECONDARY_TODO_KEYWORDS: [&str; 13] = [
     "COMBAK",
 ];
 
+lazy_static! {
+    static ref PRIMARY_KEYWORD_REGEXES: Vec<Regex> = PRIMARY_TODO_KEYWORDS
+        .iter()
+        .map(|keyword| Regex::new(&format!(
+            r"(?i)\b{}\b",
+            regex::escape(keyword)
+        ))
+        .unwrap())
+        .collect();
+    static ref SECONDARY_KEYWORD_REGEX: Regex =
+        Regex::new(r"(?i)^[^\w]*(\w+)([\s\p{P}]*)(:|[\p{P}\s]|$)").unwrap();
+}
+
 pub fn identify_todo_comment(comment_text: &str) -> Option<String> {
     let trimmed_text = comment_text.trim();
 
-    let words_with_separators: Vec<&str> =
-        trimmed_text.split_whitespace().collect();
-
-    for keyword in PRIMARY_TODO_KEYWORDS.iter() {
-        for word in &words_with_separators {
-            if word.to_uppercase().contains(&keyword.to_uppercase()) {
-                return Some(keyword.to_string());
-            }
+    for (i, re) in PRIMARY_KEYWORD_REGEXES.iter().enumerate() {
+        if re.is_match(trimmed_text) {
+            return Some(PRIMARY_TODO_KEYWORDS[i].to_string());
         }
     }
 
-    if let Some(first_word) = words_with_separators.first() {
+    if let Some(captures) = SECONDARY_KEYWORD_REGEX.captures(trimmed_text) {
+        let first_word = captures.get(1).unwrap().as_str();
+
         for keyword in SECONDARY_TODO_KEYWORDS.iter() {
-            if first_word.to_uppercase() == keyword.to_uppercase() {
+            if first_word.eq_ignore_ascii_case(keyword) {
                 return Some(keyword.to_string());
             }
         }
diff --git a/tests/identify_todo_comment.rs b/tests/identify_todo_comment.rs
index ad72dae..5e2f737 100644
--- a/tests/identify_todo_comment.rs
+++ b/tests/identify_todo_comment.rs
@@ -25,9 +25,15 @@ async fn test_primary_keyword_lowercase() {
 }
 
 #[tokio::test]
-async fn test_secondary_keyword_at_start() {
-    let comment = "// FIXME: There is a bug.";
-    assert_eq!(identify_todo_comment(comment), Some("FIXME".to_string()));
+async fn test_secondary_keyword_at_start_with_colon() {
+    let comment = "// NB: There is a nota bene.";
+    assert_eq!(identify_todo_comment(comment), Some("NB".to_string()));
+}
+
+#[tokio::test]
+async fn test_secondary_keyword_at_start_with_no_colon() {
+    let comment = "// DEBUG: There is a debug.";
+    assert_eq!(identify_todo_comment(comment), Some("DEBUG".to_string()));
 }
 
 #[tokio::test]
@@ -71,3 +77,15 @@ async fn test_no_keyword_found() {
     let comment = "// This is just a comment without anything.";
     assert_eq!(identify_todo_comment(comment), None);
 }
+
+#[tokio::test]
+async fn test_no_keyword_found_with_similar_primary_word() {
+    let comment = "// I love todoctor";
+    assert_eq!(identify_todo_comment(comment), None);
+}
+
+#[tokio::test]
+async fn test_no_keyword_found_with_similar_secondary_word() {
+    let comment = "// Dangerous stuff";
+    assert_eq!(identify_todo_comment(comment), None);
+}