Mirror/MarginaliaSearch
1
0
Fork 0
mirror of https://github.com/MarginaliaSearch/MarginaliaSearch.git synced 2025-02-24 05:18:58 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

(valuator) Clean up code

Browse Source
This commit is contained in:
Viktor Lofgren 2023-08-18 11:26:56 +02:00
parent ccf4990add
commit fcfe07fb7d
1 changed files with 10 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
code/processes/converting-process/src/main/java/nu/marginalia/converting/processor/logic/DocumentValuator.java
View File

@ -99,7 +99,7 @@ public class DocumentValuator {
return quality + adjustment; return quality + adjustment;
} }
private static class ScriptVisitor implements NodeVisitor { public static class ScriptVisitor implements NodeVisitor {
boolean hasBadScript = false; boolean hasBadScript = false;
int scriptLength = 0; int scriptLength = 0;
double penalty = 0.; double penalty = 0.;
@ -113,26 +113,23 @@ public class DocumentValuator {
if (node instanceof Element el) { if (node instanceof Element el) {
visitTag(el); visitTag(el);
} }
else if (node instanceof TextNode tn) {
visitScriptText(tn);
}
}
private void visitScriptText(TextNode tn) {
String wholeText = tn.getWholeText();
scriptLength += wholeText.length();
if (!hasBadScript) {
hasBadScript = wholeText.contains(".createElement(");
}
} }
public void visitTag(Element el) { public void visitTag(Element el) {
String srcAttr = el.attr("src"); String srcAttr = el.attr("src");
if (srcAttr.contains("wp-content") || srcAttr.contains("wp-includes") || srcAttr.contains("jquery")) { if (srcAttr.contains("wp-content") || srcAttr.contains("wp-includes") || srcAttr.contains("jquery")) {
penalty += 0.49; penalty += 0.49;
} else if (!Strings.isBlank(srcAttr)) { } else if (!Strings.isBlank(srcAttr)) {
penalty += 1; penalty += 1;
} else {
var wt = el.wholeText();
scriptLength += wt.length();
penalty += 0.25;
if (!hasBadScript) {
hasBadScript = wt.contains(".createElement(");
}
} }
} }
} }