2018-08-02 13:33:06 +00:00
|
|
|
package eu.faircode.email;
|
|
|
|
|
|
|
|
/*
|
2018-08-14 05:53:24 +00:00
|
|
|
This file is part of FairEmail.
|
2018-08-02 13:33:06 +00:00
|
|
|
|
2018-08-14 05:53:24 +00:00
|
|
|
FairEmail is free software: you can redistribute it and/or modify
|
2018-08-02 13:33:06 +00:00
|
|
|
it under the terms of the GNU General Public License as published by
|
|
|
|
the Free Software Foundation, either version 3 of the License, or
|
|
|
|
(at your option) any later version.
|
|
|
|
|
2018-10-29 10:46:49 +00:00
|
|
|
FairEmail is distributed in the hope that it will be useful,
|
2018-08-02 13:33:06 +00:00
|
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
GNU General Public License for more details.
|
|
|
|
|
|
|
|
You should have received a copy of the GNU General Public License
|
2018-10-29 10:46:49 +00:00
|
|
|
along with FairEmail. If not, see <http://www.gnu.org/licenses/>.
|
2018-08-02 13:33:06 +00:00
|
|
|
|
|
|
|
Copyright 2018 by Marcel Bokhorst (M66B)
|
|
|
|
*/
|
|
|
|
|
|
|
|
import org.jsoup.Jsoup;
|
|
|
|
import org.jsoup.nodes.Document;
|
2018-08-28 09:10:31 +00:00
|
|
|
import org.jsoup.nodes.Element;
|
2018-08-02 13:33:06 +00:00
|
|
|
import org.jsoup.nodes.Node;
|
|
|
|
import org.jsoup.nodes.TextNode;
|
2018-08-28 12:52:33 +00:00
|
|
|
import org.jsoup.safety.Whitelist;
|
2018-08-02 13:33:06 +00:00
|
|
|
import org.jsoup.select.NodeTraversor;
|
|
|
|
import org.jsoup.select.NodeVisitor;
|
|
|
|
|
2018-08-05 08:53:43 +00:00
|
|
|
import java.util.regex.Matcher;
|
|
|
|
import java.util.regex.Pattern;
|
2018-08-02 13:33:06 +00:00
|
|
|
|
2018-09-12 11:54:16 +00:00
|
|
|
public class HtmlHelper {
|
2018-09-02 11:18:32 +00:00
|
|
|
private static Pattern pattern = Pattern.compile("([http|https]+://[\\w\\S(\\.|:|/)]+)");
|
2018-08-02 13:33:06 +00:00
|
|
|
|
2018-12-09 14:49:43 +00:00
|
|
|
static String sanitize(String html) {
|
2018-12-14 09:05:48 +00:00
|
|
|
Document document = Jsoup.parse(Jsoup.clean(html, Whitelist
|
|
|
|
.relaxed()
|
|
|
|
.addProtocols("img", "src", "cid")
|
|
|
|
.addProtocols("img", "src", "data")));
|
2018-11-24 11:27:44 +00:00
|
|
|
|
2018-09-12 11:54:16 +00:00
|
|
|
for (Element tr : document.select("tr"))
|
|
|
|
tr.after("<br>");
|
2018-11-24 11:27:44 +00:00
|
|
|
|
2018-11-24 11:42:21 +00:00
|
|
|
for (Element img : document.select("img")) {
|
|
|
|
boolean linked = false;
|
|
|
|
for (Element parent : img.parents())
|
|
|
|
if ("a".equals(parent.tagName())) {
|
|
|
|
linked = true;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
if (!linked) {
|
2018-11-24 11:27:44 +00:00
|
|
|
String src = img.attr("src");
|
|
|
|
if (src.startsWith("http://") || src.startsWith("https://")) {
|
|
|
|
Element a = document.createElement("a");
|
|
|
|
a.attr("href", src);
|
|
|
|
img.replaceWith(a);
|
|
|
|
a.appendChild(img);
|
|
|
|
}
|
|
|
|
}
|
2018-11-24 11:42:21 +00:00
|
|
|
}
|
2018-11-24 11:27:44 +00:00
|
|
|
|
2018-09-12 11:54:16 +00:00
|
|
|
NodeTraversor.traverse(new NodeVisitor() {
|
|
|
|
@Override
|
|
|
|
public void head(Node node, int depth) {
|
|
|
|
if (node instanceof TextNode) {
|
|
|
|
String text = ((TextNode) node).text();
|
|
|
|
Matcher matcher = pattern.matcher(text);
|
|
|
|
while (matcher.find()) {
|
|
|
|
String ref = matcher.group();
|
|
|
|
text = text.replace(ref, String.format("<a href=\"%s\">%s</a>", ref, ref));
|
2018-09-02 11:18:32 +00:00
|
|
|
}
|
2018-09-12 11:54:16 +00:00
|
|
|
node.before(text);
|
|
|
|
((TextNode) node).text("");
|
2018-09-02 11:18:32 +00:00
|
|
|
}
|
2018-09-12 11:54:16 +00:00
|
|
|
}
|
2018-09-02 11:18:32 +00:00
|
|
|
|
2018-09-12 11:54:16 +00:00
|
|
|
@Override
|
|
|
|
public void tail(Node node, int depth) {
|
|
|
|
}
|
|
|
|
}, document.body());
|
|
|
|
return document.body().html();
|
2018-08-02 13:33:06 +00:00
|
|
|
}
|
|
|
|
}
|