refactor: Switch to bytes for symbol lookup

src/dict.rs

@@ -12,7 +12,7 @@ impl Dictionary {
         DICTIONARY.get(word).map(|s| *s)
     }
 
-    pub fn correct_bytes<'s>(&'s self, word: &[u8]) -> Option<&'s [u8]> {
+    pub fn correct_bytes<'s>(&'s self, word: &[u8]) -> Option<&'s str> {
-        std::str::from_utf8(word).ok().and_then(|word| DICTIONARY.get(word)).map(|s| s.as_bytes())
+        std::str::from_utf8(word).ok().and_then(|word| DICTIONARY.get(word)).map(|s| *s)
     }
 }

src/lib.rs

@@ -17,21 +17,20 @@ pub fn process_file(path: &std::path::Path, dictionary: &Dictionary, report: rep
     for (line_idx, line) in grep_searcher::LineIter::new(b'\n', &buffer).enumerate() {
         let line_num = line_idx + 1;
         for token in identifier::tokenize(line) {
-            if let Some(word) = std::str::from_utf8(token.token).ok() {
+            // Correct tokens as-is
-                // Correct tokens as-is
+            if let Some(correction) = dictionary.correct_bytes(token.token) {
-                if let Some(correction) = dictionary.correct_str(word) {
+                let word = String::from_utf8_lossy(token.token);
-                    let col_num = token.offset;
+                let col_num = token.offset;
-                    let msg = report::Message {
+                let msg = report::Message {
-                        path,
+                    path,
-                        line,
+                    line,
-                        line_num,
+                    line_num,
-                        col_num,
+                    col_num,
-                        word,
+                    word: word.as_ref(),
-                        correction,
+                    correction,
-                        non_exhaustive: (),
+                    non_exhaustive: (),
-                    };
+                };
-                    report(msg);
+                report(msg);
-                }
             }
         }
     }