add Clojure/ClojureScript pre processor

Author: RobinMalfait

crates/oxide/src/extractor/pre_processors/clojure.rs

@@ -0,0 +1,129 @@
+use crate::cursor;
+use crate::extractor::pre_processors::pre_processor::PreProcessor;
+
+#[derive(Debug, Default)]
+pub struct Clojure;
+
+impl PreProcessor for Clojure {
+    fn process(&self, content: &[u8]) -> Vec<u8> {
+        let len = content.len();
+        let mut result = content.to_vec();
+        let mut cursor = cursor::Cursor::new(content);
+
+        while cursor.pos < len {
+            match cursor.curr {
+                // Consume strings as-is
+                b'"' => {
+                    let len = cursor.input.len();
+                    cursor.advance();
+
+                    while cursor.pos < len {
+                        match cursor.curr {
+                            // Escaped character, skip ahead to the next character
+                            b'\\' => cursor.advance_twice(),
+
+                            // End of the string
+                            b'"' => break,
+
+                            // Everything else is valid
+                            _ => cursor.advance(),
+                        };
+                    }
+                }
+
+                b':' | b'.' => {
+                    result[cursor.pos] = b' ';
+                }
+
+                // Consume everything else
+                _ => {}
+            };
+
+            cursor.advance();
+        }
+
+        result
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::Clojure;
+    use crate::extractor::pre_processors::pre_processor::PreProcessor;
+
+    #[test]
+    fn test_clojure_pre_processor() {
+        for (input, expected) in [
+            (":div.flex-1.flex-2", " div flex-1 flex-2"),
+            (
+                ":.flex-3.flex-4 ;defaults to div",
+                "  flex-3 flex-4 ;defaults to div",
+            ),
+            ("{:class :flex-5.flex-6", "{ class  flex-5 flex-6"),
+            (r#"{:class "flex-7 flex-8"}"#, r#"{ class "flex-7 flex-8"}"#),
+            (
+                r#"{:class  ["flex-9" :flex-10]}"#,
+                r#"{ class  ["flex-9"  flex-10]}"#,
+            ),
+            (
+                r#"(dom/div {:class "flex-11 flex-12"})"#,
+                r#"(dom/div { class "flex-11 flex-12"})"#,
+            ),
+            ("(dom/div :.flex-13.flex-14", "(dom/div   flex-13 flex-14"),
+        ] {
+            Clojure::test(input, expected);
+        }
+    }
+
+    #[test]
+    fn test_extract_candidates() {
+        // https://github.com/luckasRanarison/tailwind-tools.nvim/issues/68#issuecomment-2660951258
+        let input = r#"
+            :div.c1.c2
+            :.c3.c4 ;defaults to div
+            {:class :c5.c6
+            {:class "c7 c8"}
+            {:class  ["c9" :c10]}
+            (dom/div {:class "c11 c12"})
+            (dom/div :.c13.c14
+        "#;
+
+        Clojure::test_extract_contains(
+            input,
+            vec![
+                "c1", "c2", "c3", "c4", "c5", "c6", "c7", "c8", "c9", "c10", "c11", "c12", "c13",
+                "c14",
+            ],
+        );
+
+        // Similar structure but using a real classes
+        let input = r#"
+            :div.flex-1.flex-2
+            :.flex-3.flex-4 ;defaults to div
+            {:class :flex-5.flex-6
+            {:class "flex-7 flex-8"}
+            {:class  ["flex-9" :flex-10]}
+            (dom/div {:class "flex-11 flex-12"})
+            (dom/div :.flex-13.flex-14
+        "#;
+
+        Clojure::test_extract_contains(
+            input,
+            vec![
+                "flex-1", "flex-2", "flex-3", "flex-4", "flex-5", "flex-6", "flex-7", "flex-8",
+                "flex-9", "flex-10", "flex-11", "flex-12", "flex-13", "flex-14",
+            ],
+        );
+    }
+
+    #[test]
+    fn test_special_characters_are_valid_in_strings() {
+        // In this case the `:` and `.` should not be replaced by ` ` because they are inside a
+        // string.
+        let input = r#"
+            (dom/div {:class "hover:flex px-1.5"})
+        "#;
+
+        Clojure::test_extract_contains(input, vec!["hover:flex", "px-1.5"]);
+    }
+}

crates/oxide/src/extractor/pre_processors/mod.rs

@@ -1,3 +1,4 @@
+pub mod clojure;
 pub mod haml;
 pub mod pre_processor;
 pub mod pug;
@@ -6,6 +7,7 @@ pub mod ruby;
 pub mod slim;
 pub mod svelte;
 
+pub use clojure::*;
 pub use haml::*;
 pub use pre_processor::*;
 pub use pug::*;

crates/oxide/src/lib.rs

@@ -468,6 +468,7 @@ pub fn pre_process_input(content: &[u8], extension: &str) -> Vec<u8> {
     use crate::extractor::pre_processors::*;
 
     match extension {
+        "clj" | "cljs" | "cljc" => Clojure.process(content),
         "cshtml" | "razor" => Razor.process(content),
         "haml" => Haml.process(content),
         "pug" => Pug.process(content),