Start writing some tests for decoding functionality

Fix up descriptions on some decoding functions

Author: dralley

src/encoding.rs

@@ -105,10 +105,9 @@ impl Decoder {
     }
 }
 
-/// Decodes the provided bytes using the specified encoding, ignoring the BOM
-/// if it is present in the `bytes`.
+/// Decodes the provided bytes using the specified encoding.
 ///
-/// Returns an error in case of malformed sequences in the `bytes`.
+/// Returns an error in case of malformed or non-representable sequences in the `bytes`.
 #[cfg(feature = "encoding")]
 pub fn decode<'b>(bytes: &'b [u8], encoding: &'static Encoding) -> Result<Cow<'b, str>> {
     encoding
@@ -119,7 +118,7 @@ pub fn decode<'b>(bytes: &'b [u8], encoding: &'static Encoding) -> Result<Cow<'b
 /// Decodes a slice with an unknown encoding, removing the BOM if it is present
 /// in the bytes.
 ///
-/// Returns an error in case of malformed sequences in the `bytes`.
+/// Returns an error in case of malformed or non-representable sequences in the `bytes`.
 #[cfg(feature = "encoding")]
 pub fn decode_with_bom_removal<'b>(bytes: &'b [u8]) -> Result<Cow<'b, str>> {
     if let Some(encoding) = detect_encoding(bytes) {
@@ -185,5 +184,3 @@ pub fn detect_encoding(bytes: &[u8]) -> Option<&'static Encoding> {
         _ => None,
     }
 }
-
-// TODO: add some tests for functions

tests/documents/utf8.xml

@@ -0,0 +1,3 @@
+<?xml version="1.0"?>
+<project name="project-name">
+</project>

tests/encodings.rs

@@ -0,0 +1,47 @@
+#[cfg(feature = "encoding")]
+mod decode {
+    use encoding_rs::{UTF_16BE, UTF_16LE, UTF_8};
+    use quick_xml::encoding::*;
+    use std::borrow::Cow;
+
+    static UTF16BE_TEXT_WITH_BOM: &[u8] = include_bytes!("./documents/utf16be.xml");
+    static UTF16LE_TEXT_WITH_BOM: &[u8] = include_bytes!("./documents/utf16le.xml");
+    static UTF8_TEXT_WITH_BOM: &[u8] = include_bytes!("./documents/utf8.xml");
+
+    static UTF8_TEXT: &str = r#"<?xml version="1.0"?>
+<project name="project-name">
+</project>
+"#;
+
+    #[test]
+    fn test_removes_bom() {
+        // No BOM
+        assert_eq!(
+            decode_with_bom_removal(UTF8_TEXT.as_bytes()).unwrap(),
+            Cow::Borrowed(UTF8_TEXT)
+        );
+        // BOM
+        assert_eq!(
+            decode_with_bom_removal(UTF8_TEXT_WITH_BOM).unwrap(),
+            Cow::Borrowed(UTF8_TEXT)
+        );
+        assert_eq!(
+            decode_with_bom_removal(UTF16BE_TEXT_WITH_BOM).unwrap(),
+            Cow::Borrowed(UTF8_TEXT).into_owned()
+        );
+        assert_eq!(
+            decode_with_bom_removal(UTF16LE_TEXT_WITH_BOM).unwrap(),
+            Cow::Borrowed(UTF8_TEXT).into_owned()
+        );
+    }
+
+    #[test]
+    fn test_detect_encoding() {
+        // No BOM
+        assert_eq!(detect_encoding(UTF8_TEXT.as_bytes()), Some(UTF_8));
+        // BOM
+        assert_eq!(detect_encoding(UTF8_TEXT_WITH_BOM), Some(UTF_8));
+        assert_eq!(detect_encoding(UTF16BE_TEXT_WITH_BOM), Some(UTF_16BE));
+        assert_eq!(detect_encoding(UTF16LE_TEXT_WITH_BOM), Some(UTF_16LE));
+    }
+}

tests/test.rs

@@ -89,24 +89,6 @@ fn test_comment_starting_with_gt() {
     }
 }
 
-#[test]
-#[cfg(feature = "encoding")]
-fn test_koi8_r_encoding() {
-    let src = include_bytes!("documents/opennews_all.rss").as_ref();
-    let mut buf = vec![];
-    let mut r = Reader::from_reader(src);
-    r.trim_text(true).expand_empty_elements(false);
-    loop {
-        match r.read_event_into(&mut buf) {
-            Ok(Text(e)) => {
-                e.unescape().unwrap();
-            }
-            Ok(Eof) => break,
-            _ => (),
-        }
-    }
-}
-
 #[test]
 fn fuzz_53() {
     let data: &[u8] = b"\xe9\x00\x00\x00\x00\x00\x00\x00\x00\