Add support for BOM

5 files changed, 24 insertions, 10 deletions

diff --git a/readme.md b/readme.md
index 6c2e57d..86b6b6e 100644
--- a/readme.md
+++ b/readme.md
@@ -66,7 +66,6 @@ cargo doc --document-private-items
 
 ### Small things
 
-- [ ] (1) Handle BOM at start
 - [ ] (1) Parse initial and final whitespace of paragraphs (in text)
 - [ ] (1) Add docs to subtokenize
 - [ ] (1) Add module docs to parser
@@ -171,6 +170,7 @@ cargo doc --document-private-items
 - [x] (1) Parse whitespace in each flow construct
 - [x] (1) Connect `ChunkString` in label, destination, title
 - [x] (1) Add support for line endings in `string`
+- [x] (1) Handle BOM at start
 
 ### Extensions
 
diff --git a/src/subtokenize.rs b/src/subtokenize.rs
index 4a29a01..0623a37 100644
--- a/src/subtokenize.rs
+++ b/src/subtokenize.rs
@@ -15,6 +15,10 @@ pub fn subtokenize(events: Vec<Event>, codes: &[Code]) -> (Vec<Event>, bool) {
     let mut link_to_info: HashMap<usize, (usize, usize, usize)> = HashMap::new();
     let mut done = true;
 
+    if events.is_empty() {
+        return (events, true);
+    }
+
     while index < events.len() {
         let event = &events[index];
 
diff --git a/src/tokenizer.rs b/src/tokenizer.rs
index d31c8c5..c0a7105 100644
--- a/src/tokenizer.rs
+++ b/src/tokenizer.rs
@@ -680,10 +680,20 @@ fn attempt_impl(
 // To do: handle BOM at start?
 pub fn as_codes(value: &str) -> Vec<Code> {
     let mut codes: Vec<Code> = vec![];
+    let mut at_start = true;
     let mut at_carriage_return = false;
     let mut column = 1;
 
     for char in value.chars() {
+        if at_start {
+            if char == '\u{feff}' {
+                // Ignore.
+                continue;
+            }
+
+            at_start = false;
+        }
+
         // Send a CRLF.
         if at_carriage_return && '\n' == char {
             at_carriage_return = false;
diff --git a/tests/misc_bom.rs b/tests/misc_bom.rs
index 9805616..44f661e 100644
--- a/tests/misc_bom.rs
+++ b/tests/misc_bom.rs
@@ -1,15 +1,13 @@
 extern crate micromark;
-// use micromark::micromark;
+use micromark::micromark;
 
 #[test]
 fn bom() {
-    // // To do: BOM.
-    // assert_eq!(micromark("\u{FEFF}"), "", "should ignore just a bom");
+    assert_eq!(micromark("\u{FEFF}"), "", "should ignore just a bom");
 
-    // // To do: BOM.
-    // assert_eq!(
-    //     micromark("\u{FEFF}# hea\u{FEFF}ding"),
-    //     "<h1>hea\u{FEFF}ding</h1>",
-    //     "should ignore a bom"
-    // );
+    assert_eq!(
+        micromark("\u{FEFF}# hea\u{FEFF}ding"),
+        "<h1>hea\u{FEFF}ding</h1>",
+        "should ignore a bom"
+    );
 }
diff --git a/tests/misc_zero.rs b/tests/misc_zero.rs
index 946a3e2..47aa8ed 100644
--- a/tests/misc_zero.rs
+++ b/tests/misc_zero.rs
@@ -3,6 +3,8 @@ use micromark::micromark;
 
 #[test]
 fn zero() {
+    assert_eq!(micromark(""), "", "should support no markdown");
+
     assert_eq!(
         micromark("asd\0asd"),
         "<p>asd�asd</p>",