boxtec
/
yacy_search_server
mirror of https://github.com/yacy/yacy_search_server
4
1
Fork 0
Code Issues Projects Releases Wiki Activity

bugfix for last commit

Browse Source 
git-svn-id: https://svn.berlios.de/svnroot/repos/yacy/trunk@1028 6c8d7289-2bf4-0310-a012-ef5d649a1542
pull/1/head
borg-0300 19 years ago
parent 544e4ea90e
commit 60e869f236
1 changed files with 12 additions and 13 deletions
Show Stats Download Patch File Download Diff File

25
source/de/anomic/htmlFilter/htmlFilterContentScraper.java
Unescape View File

@ -122,19 +122,7 @@ public class htmlFilterContentScraper extends htmlFilterAbstractScraper implemen
// cutting of everything behind #
int cpos = us.indexOf("#");
if (cpos >= 0) { us = us.substring(0, cpos); }
if (us.startsWith("http")) {
if (us.endsWith(":80")) {
us = us.substring(0, us.length() - 3);
serverLog.logFinest("htmlFilter", "urlNormalform: :80=" + us);
} else {
cpos = us.indexOf(":80/");
if (cpos >= 0) {
us = us.substring(0, cpos).concat(us.substring(cpos + 3));
serverLog.logFinest("htmlFilter", "urlNormalform: :80/=" + us);
}
}
} else if (us.startsWith("https")) {
if (us.startsWith("https")) {
if (us.endsWith(":443")) {
us = us.substring(0, us.length() - 4);
serverLog.logFinest("htmlFilter", "urlNormalform: :443=" + us);
@ -145,6 +133,17 @@ public class htmlFilterContentScraper extends htmlFilterAbstractScraper implemen
serverLog.logFinest("htmlFilter", "urlNormalform: :443/=" + us);
}
}
} else if (us.startsWith("http")) {
if (us.endsWith(":80")) {
us = us.substring(0, us.length() - 3);
serverLog.logFinest("htmlFilter", "urlNormalform: :80=" + us);
} else {
cpos = us.indexOf(":80/");
if (cpos >= 0) {
us = us.substring(0, cpos).concat(us.substring(cpos + 3));
serverLog.logFinest("htmlFilter", "urlNormalform: :80/=" + us);
}
}
}
if (((us.endsWith("/")) && (us.lastIndexOf('/', us.length() - 2) < 8))) us = us.substring(0, us.length() - 1);
serverLog.logFiner("htmlFilter", "urlNormalform: OUT=" + us);

Write Preview
Loading…
Cancel
Save