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

experimental implementation of base parsing in html parser

Browse Source 
git-svn-id: https://svn.berlios.de/svnroot/repos/yacy/trunk@355 6c8d7289-2bf4-0310-a012-ef5d649a1542
pull/1/head
orbiter 20 years ago
parent 86f2aa8478
commit 1d3cfb380a
1 changed files with 3 additions and 2 deletions
Show Stats Download Patch File Download Diff File

5
source/de/anomic/htmlFilter/htmlFilterContentScraper.java
Unescape View File

@ -123,8 +123,9 @@ public class htmlFilterContentScraper extends htmlFilterAbstractScraper implemen
public void scrapeTag1(String tagname, Properties tagopts, byte[] text) { public void scrapeTag1(String tagname, Properties tagopts, byte[] text) {
//System.out.println("ScrapeTag1: tagname=" + tagname + ", opts=" + tagopts.toString() + ", text=" + new String(text)); //System.out.println("ScrapeTag1: tagname=" + tagname + ", opts=" + tagopts.toString() + ", text=" + new String(text));
if ((tagname.equals("a")) && (text.length < 2048)) anchors.put(absolutePath(tagopts.getProperty("href", "")), super.stripAll(new serverByteBuffer(text)).trim().toString()); if ((tagname.equals("a")) && (text.length < 2048)) anchors.put(absolutePath(tagopts.getProperty("href", "")), super.stripAll(new serverByteBuffer(text)).trim().toString());
if ((tagname.equals("h1")) && (text.length < 512)) headline = super.stripAll(new serverByteBuffer(text)).toString(); if ((tagname.equals("h1")) && (text.length < 1024)) headline = super.stripAll(new serverByteBuffer(text)).toString();
if ((tagname.equals("title")) && (text.length < 512)) title = super.stripAll(new serverByteBuffer(text)).toString(); if ((tagname.equals("title")) && (text.length < 1024)) title = super.stripAll(new serverByteBuffer(text)).toString();
if ((tagname.equals("base")) && (text.length < 512)) try {root = new URL(tagopts.getProperty("href", ""));} catch (MalformedURLException e) {}
} }

Write Preview
Loading…
Cancel
Save