*/
package jalview.util;
+import java.util.Locale;
+
import java.util.ArrayList;
+import java.util.List;
import java.util.StringTokenizer;
import java.util.regex.Pattern;
*/
public class ParseHtmlBodyAndLinks
{
+ private static final Pattern LEFT_ANGLE_BRACKET_PATTERN = Pattern
+ .compile("<");
+
String orig = null;
public String getOrig()
return htmlContent;
}
- ArrayList<String> links = new ArrayList<String>();
+ List<String> links = new ArrayList<String>();
- StringBuffer sb = new StringBuffer();
+ String content;
/**
* result of parsing description - with or without HTML tags
public String getContent()
{
- return sb.toString();
+ return content;
}
/**
*
* @return
*/
- public ArrayList<String> getLinks()
+ public List<String> getLinks()
{
return links;
}
/**
+ * Parses the given html and
+ * <ul>
+ * <li>extracts any 'href' links to a list of "displayName|url" strings,
+ * retrievable by #getLinks</li>
+ * <li>extracts the remaining text (with %LINK% placeholders replacing hrefs),
+ * retrievable by #getContent</li>
+ * </ul>
*
* @param description
* - html or text content to be parsed
htmlContent = false;
return;
}
- if (description.toUpperCase().indexOf("<HTML>") == -1)
+ StringBuilder sb = new StringBuilder(description.length());
+ if (description.toUpperCase(Locale.ROOT).indexOf("<HTML>") == -1)
{
htmlContent = false;
}
String tag = null;
while (st.hasMoreElements())
{
- token = st.nextToken("&>");
+ token = st.nextToken(">");
if (token.equalsIgnoreCase("html") || token.startsWith("/"))
{
continue;
token = token.substring(0, startTag);
}
- if (tag != null && tag.toUpperCase().startsWith("A HREF="))
+ if (tag != null && tag.toUpperCase(Locale.ROOT).startsWith("A HREF="))
{
if (token.length() > 0)
{
{
sb.append(newline);
}
- else if (token.startsWith("lt;"))
- {
- sb.append("<" + token.substring(3));
- }
- else if (token.startsWith("gt;"))
- {
- sb.append(">" + token.substring(3));
- }
- else if (token.startsWith("amp;"))
- {
- sb.append("&" + token.substring(4));
- }
else
{
sb.append(token);
{
// instead of parsing the html into plaintext
// clean the description ready for embedding in html
- sb = new StringBuffer(Pattern.compile("<").matcher(description)
+ sb = new StringBuilder(LEFT_ANGLE_BRACKET_PATTERN.matcher(description)
.replaceAll("<"));
-
}
+ content = translateEntities(sb.toString());
+ }
+ private String translateEntities(String s)
+ {
+ s = s.replaceAll("&", "&");
+ s = s.replaceAll("<", "<");
+ s = s.replaceAll(">", ">");
+ return s;
}
/**
*/
public String getNonHtmlContent()
{
- return isHtmlContent() ? sb.toString() : orig;
+ return isHtmlContent() ? content : orig;
}
}