From: <rb...@us...> - 2018-07-15 15:06:51
|
Revision: 15464 http://sourceforge.net/p/htmlunit/code/15464 Author: rbri Date: 2018-07-15 15:06:46 +0000 (Sun, 15 Jul 2018) Log Message: ----------- set default charset for the parser if not specified Modified Paths: -------------- trunk/htmlunit/src/main/java/com/gargoylesoftware/htmlunit/html/HTMLParser.java Modified: trunk/htmlunit/src/main/java/com/gargoylesoftware/htmlunit/html/HTMLParser.java =================================================================== --- trunk/htmlunit/src/main/java/com/gargoylesoftware/htmlunit/html/HTMLParser.java 2018-07-15 10:53:41 UTC (rev 15463) +++ trunk/htmlunit/src/main/java/com/gargoylesoftware/htmlunit/html/HTMLParser.java 2018-07-15 15:06:46 UTC (rev 15464) @@ -28,6 +28,7 @@ import java.lang.reflect.InvocationTargetException; import java.net.URL; import java.nio.charset.Charset; +import java.nio.charset.StandardCharsets; import java.util.ArrayDeque; import java.util.ArrayList; import java.util.Deque; @@ -219,9 +220,12 @@ final URL url = webResponse.getWebRequest().getUrl(); final HtmlUnitDOMBuilder domBuilder = new HtmlUnitDOMBuilder(page, url, null); - final Charset charset = webResponse.getContentCharsetOrNull(); + Charset charset = webResponse.getContentCharsetOrNull(); try { - if (charset != null) { + if (charset == null) { + charset = StandardCharsets.ISO_8859_1; + } + else { domBuilder.setFeature(HTMLScanner.IGNORE_SPECIFIED_CHARSET, true); } |