html decode the input before highlighting

3 files changed, 20 insertions, 1 deletions

diff --git a/Cargo.lock b/Cargo.lock
index 7a75cc9..e2fb471 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -97,6 +97,15 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "9229cfe53dfd69f0609a49f65461bd93001ea1ef889cd5529dd176593f5338a1"
 
 [[package]]
+name = "html-escape"
+version = "0.2.13"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6d1ad449764d627e22bfd7cd5e8868264fc9236e07c752972b4080cd351cb476"
+dependencies = [
+ "utf8-width",
+]
+
+[[package]]
 name = "icu_collections"
 version = "2.0.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -502,6 +511,12 @@ dependencies = [
 ]
 
 [[package]]
+name = "utf8-width"
+version = "0.1.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "86bd8d4e895da8537e5315b8254664e6b769c4ff3db18321b297a1e7004392e3"
+
+[[package]]
 name = "utf8_iter"
 version = "1.0.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -604,6 +619,7 @@ name = "yah"
 version = "0.1.0"
 dependencies = [
  "dirs",
+ "html-escape",
  "libloading",
  "tree-sitter",
  "tree-sitter-highlight",
diff --git a/Cargo.toml b/Cargo.toml
index 25da5ac..ccc6bd8 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -5,6 +5,7 @@ edition = "2024"
 
 [dependencies]
 dirs = "6.0.0"
+html-escape = "0.2.13"
 libloading = "0.8.8"
 tree-sitter = "0.25.8"
 tree-sitter-highlight = "0.25.8"
diff --git a/src/main.rs b/src/main.rs
index d96cc46..0904b55 100644
--- a/src/main.rs
+++ b/src/main.rs
@@ -148,6 +148,8 @@ fn get_config<'a>(
 
 fn main() {
     if let [_, lang, code] = args().collect::<Vec<String>>().as_slice() {
+        // decode code as it may be html encoded
+        let code = html_escape::decode_html_entities(code);
         let mut highlighter = Highlighter::new();
         let configs = RefCell::new(HashMap::new());
 
@@ -174,6 +176,6 @@ fn main() {
         }).unwrap();
         print!("{}", String::from_utf8(renderer.html).unwrap())
     } else {
-        panic!("Need <lang> <code> as arguments")
+        panic!("Need <lang> <code> as arguments, got {:?}", args().collect::<Vec<String>>())
     }
 }