diff --git a/tika-parsers/src/main/java/org/apache/tika/parser/html/HtmlParser.java b/tika-parsers/src/main/java/org/apache/tika/parser/html/HtmlParser.java
index 66100f1..4baa359 100644
--- a/tika-parsers/src/main/java/org/apache/tika/parser/html/HtmlParser.java
+++ b/tika-parsers/src/main/java/org/apache/tika/parser/html/HtmlParser.java
@@ -50,8 +50,9 @@ public class HtmlParser implements Parser {
"Content-Type['\\\"]\\s+content\\s*=\\s*['\\\"]" +
"([^'\\\"]+)['\\\"]\\s*/>");
- private static final Pattern CONTENT_TYPE_PATTERN =
- Pattern.compile("(?i);\\s*charset\\s*=\\s*(.*)");
+ // TIKA-350: handle charset as first element in content-type
+ private static final Pattern CONTENT_TYPE_PATTERN = Pattern.compile(
+ "(?i)(?:;|)\\s*charset\\s*=\\s*([^\r;\\s]*)");
/**
* TIKA-332: Check for meta http-equiv tag with charset info in
@@ -86,7 +87,8 @@ public class HtmlParser implements Parser {
}
}
- // No charset in a meta http-equiv tag, so detect from actual content bytes.
+ // No charset in a meta http-equiv tag, see if it's in the passed content-encoding
+ // hint, or the passed content-type hint.
CharsetDetector detector = new CharsetDetector();
String incomingCharset = metadata.get(Metadata.CONTENT_ENCODING);
if (incomingCharset == null) {
diff --git a/tika-parsers/src/test/java/org/apache/tika/parser/html/HtmlParserTest.java b/tika-parsers/src/test/java/org/apache/tika/parser/html/HtmlParserTest.java
index da90393..18bc5af 100644
--- a/tika-parsers/src/test/java/org/apache/tika/parser/html/HtmlParserTest.java
+++ b/tika-parsers/src/test/java/org/apache/tika/parser/html/HtmlParserTest.java
@@ -336,5 +336,28 @@ public class HtmlParserTest extends TestCase {
assertEquals("ISO-8859-1", metadata.get(Metadata.CONTENT_ENCODING));
}
+ /**
+ * Test case for TIKA-350
+ * @see TIKA-350
+ */
+ public void testUsingFunkyCharsetInContentTypeHeader() throws Exception {
+ final String test =
+ "the name is \u00e1ndre"
+ + "";
+
+ Metadata metadata = new Metadata();
+ new HtmlParser().parse (
+ new ByteArrayInputStream(test.getBytes("UTF-8")),
+ new BodyContentHandler(), metadata, new ParseContext());
+ assertEquals("UTF-8", metadata.get(Metadata.CONTENT_ENCODING));
+
+ metadata = new Metadata();
+ metadata.set(Metadata.CONTENT_TYPE, "charset=ISO-8859-1;text/html");
+ new HtmlParser().parse (
+ new ByteArrayInputStream(test.getBytes("UTF-8")),
+ new BodyContentHandler(), metadata, new ParseContext());
+ assertEquals("ISO-8859-1", metadata.get(Metadata.CONTENT_ENCODING));
+ }
+
}