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

Performance optimizations in EdgeDomain's parsing, reduce the number of unguarded regular expressions

Browse Source
This commit is contained in:
Viktor Lofgren 2023-01-30 09:23:11 +01:00
parent 4854f40447
commit 618582dc74
1 changed files with 31 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

35
marginalia_nu/src/main/java/nu/marginalia/wmsa/edge/model/EdgeDomain.java
View File

@ -11,8 +11,6 @@ import java.util.regex.Pattern;
@Getter @Setter @Builder @Getter @Setter @Builder
public class EdgeDomain { public class EdgeDomain {
private static final Predicate<String> ipPatternTest = Pattern.compile("[\\d]{1,3}\\.[\\d]{1,3}\\.[\\d]{1,3}\\.[\\d]{1,3}").asMatchPredicate();
private static final Predicate<String> govListTest = Pattern.compile(".*\\.(ac|co|org|gov|edu|com)\\.[a-z]{2}").asMatchPredicate();
@Nonnull @Nonnull
public final String subDomain; public final String subDomain;
@ -27,7 +25,7 @@ public class EdgeDomain {
var dot = host.lastIndexOf('.'); var dot = host.lastIndexOf('.');
if (dot < 0 || ipPatternTest.test(host)) { // IPV6 >.> if (dot < 0 || looksLikeAnIp(host)) { // IPV6 >.>
subDomain = ""; subDomain = "";
domain = host; domain = host;
} }
@ -38,7 +36,7 @@ public class EdgeDomain {
domain = host; domain = host;
} }
else { else {
if (govListTest.test(host)) if (looksLikeGovTld(host))
{ // Capture .ac.jp, .co.uk { // Capture .ac.jp, .co.uk
int dot3 = host.substring(0, dot2).lastIndexOf('.'); int dot3 = host.substring(0, dot2).lastIndexOf('.');
if (dot3 >= 0) { if (dot3 >= 0) {
@ -59,6 +57,35 @@ public class EdgeDomain {
} }
} }
private static final Predicate<String> govListTest = Pattern.compile(".*\\.(ac|co|org|gov|edu|com)\\.[a-z]{2}").asMatchPredicate();
private boolean looksLikeGovTld(String host) {
if (host.length() < 8)
return false;
int cnt = 0;
for (int i = host.length() - 7; i < host.length(); i++) {
if (host.charAt(i) == '.')
cnt++;
}
return cnt >= 2 && govListTest.test(host);
}
private static final Predicate<String> ipPatternTest = Pattern.compile("[\\d]{1,3}\\.[\\d]{1,3}\\.[\\d]{1,3}\\.[\\d]{1,3}").asMatchPredicate();
private boolean looksLikeAnIp(String host) {
if (host.length() < 7)
return false;
char firstChar = host.charAt(0);
int lastChar = host.charAt(host.length() - 1);
return Character.isDigit(firstChar)
&& Character.isDigit(lastChar)
&& ipPatternTest.test(host);
}
public EdgeUrl toRootUrl() { public EdgeUrl toRootUrl() {
// Set default protocol to http, as most https websites redirect http->https, but few http websites redirect https->http // Set default protocol to http, as most https websites redirect http->https, but few http websites redirect https->http
return new EdgeUrl("http", this, null, "/", null); return new EdgeUrl("http", this, null, "/", null);