Fix charset parsing logic of Content-Type header

integration/hurl/tests_ok/charset.hurl

@@ -11,6 +11,13 @@ Content-Type: text/html; charset=latin1
 body == "<p>café</p>"
 
 
+GET http://localhost:8000/charset/latin1/uppercase-key
+HTTP 200
+Content-Type: text/html; CHARSET=latin1
+[Asserts]
+body == "<p>café</p>"
+
+
 GET http://localhost:8000/charset/gb2312
 HTTP 200
 Content-Type: text/html; charset=gb2312
@@ -25,8 +32,15 @@ Content-Type: text/html; charset=cp1256
 body == "<p>مرحبا بالعالم</p>"
 
 
-GET http://localhost:8000/charset/uppercase
+GET http://localhost:8000/charset/uppercase-value
 HTTP 200
 Content-Type: text/html; charset=UTF-8
 [Asserts]
 body == "<p>Hello World!</p>"
+
+
+GET http://localhost:8000/charset/many-keys
+HTTP 200
+Content-Type: text/plain; version=0.0.4; charset=utf-8; escaping=values
+[Asserts]
+body == "<p>Hello World!</p>"

integration/hurl/tests_ok/charset.py

@@ -7,20 +7,36 @@ def charset_default():
  return "<p>Hello World!</p>"
 
 
-@app.route("/charset/uppercase")
-def charset_uppercase():
+@app.route("/charset/uppercase-value")
+def charset_uppercase_value():
  resp = make_response("<p>Hello World!</p>")
  resp.headers["Content-Type"] = "text/html; charset=UTF-8"
  return resp
 
 
+@app.route("/charset/many-keys")
+def charset_uppercase_many_keys():
+ resp = make_response("<p>Hello World!</p>")
+ resp.headers[
+ "Content-Type"
+ ] = "text/plain; version=0.0.4; charset=utf-8; escaping=values"
+ return resp
+
+
 @app.route("/charset/latin1")
 def charset_latin1():
  resp = make_response("<p>café</p>".encode("latin1"))
  resp.headers["Content-Type"] = "text/html; charset=latin1"
  return resp
 
 
+@app.route("/charset/latin1/uppercase-key")
+def charset_latin1_uppercase_key():
+ resp = make_response("<p>café</p>".encode("latin1"))
+ resp.headers["Content-Type"] = "text/html; CHARSET=latin1"
+ return resp
+
+
 @app.route("/charset/gb2312")
 def charset_gb2312():
  resp = make_response("<p>你好世界</p>".encode("gb2312"))

packages/hurl/src/http/mimetype.rs

@@ -36,9 +36,14 @@ pub fn is_html(content_type: &str) -> bool {
 
 /// Extracts charset from mime-type String
 pub fn charset(mime_type: &str) -> Option<String> {
- mime_type
- .find("charset=")
- .map(|index| mime_type[(index + 8)..].to_string())
+ let parts = mime_type.trim().split(';');
+ for part in parts {
+ let param = part.trim().split('=').collect::<Vec<_>>();
+ if param.len() == 2 && param[0].trim().eq_ignore_ascii_case("charset") {
+ return Some(param[1].trim().to_string());
+ }
+ }
+ None
 }
 
 #[cfg(test)]
@@ -51,10 +56,22 @@ pub mod tests {
  charset("text/plain; charset=utf-8"),
  Some("utf-8".to_string())
  );
+
  assert_eq!(
  charset("text/plain; charset=ISO-8859-1"),
  Some("ISO-8859-1".to_string())
  );
+
  assert_eq!(charset("text/plain;"), None);
+
+ assert_eq!(
+ charset("text/plain; CHARSET=ISO-8859-1"),
+ Some("ISO-8859-1".to_string())
+ );
+
+ assert_eq!(
+ charset("text/plain; version=0.0.4; charset=utf-8; escaping=values"),
+ Some("utf-8".to_string())
+ );
  }
 }