This is an automated email from the ASF dual-hosted git repository.
sebastian-nagel pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/nutch.git
The following commit(s) were added to refs/heads/master by this push:
new 5bcb0ab87 NUTCH-3180 BasicURLNormalizer missing catching
ICUInputTooLongException (#923)
5bcb0ab87 is described below
commit 5bcb0ab87d263d3700b835113d6ee7facbe3d7e7
Author: Luca <[email protected]>
AuthorDate: Fri Jun 12 15:24:49 2026 +0200
NUTCH-3180 BasicURLNormalizer missing catching ICUInputTooLongException
(#923)
* fix: wrap the exceptions of host validation and rethrow
MalformedURLException
* test: cover exception chain
---
src/java/org/apache/nutch/util/URLUtil.java | 21 ++++++++++++++++-----
src/test/org/apache/nutch/util/TestURLUtil.java | 16 ++++++++++++++++
2 files changed, 32 insertions(+), 5 deletions(-)
diff --git a/src/java/org/apache/nutch/util/URLUtil.java
b/src/java/org/apache/nutch/util/URLUtil.java
index fd036480a..851d10bef 100644
--- a/src/java/org/apache/nutch/util/URLUtil.java
+++ b/src/java/org/apache/nutch/util/URLUtil.java
@@ -24,6 +24,7 @@ import java.net.URL;
import java.util.Locale;
import java.util.regex.Pattern;
+import com.ibm.icu.util.ICUException;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
@@ -652,7 +653,6 @@ public class URLUtil {
// IndexOutOfBoundsException: thrown (undocumented) if one "label"
// (non-ASCII dot-separated segment) is longer than 256 characters,
// cf. https://bugs.openjdk.java.net/browse/JDK-6806873
- LOG.debug("Failed to convert IDN host {}: ", host, e);
throw (MalformedURLException) new MalformedURLException(
"Invalid IDN " + host + ": " + e.getMessage()).initCause(e);
}
@@ -675,11 +675,22 @@ public class URLUtil {
throws MalformedURLException {
final IDNA.Info idnaInfo = new IDNA.Info();
final StringBuilder hostConverted = new StringBuilder();
- if (toAscii) {
- idna.nameToASCII(host, hostConverted, idnaInfo);
- } else {
- idna.nameToUnicode(host, hostConverted, idnaInfo);
+ try {
+ if (toAscii) {
+ idna.nameToASCII(host, hostConverted, idnaInfo);
+ } else {
+ idna.nameToUnicode(host, hostConverted, idnaInfo);
+ }
+ } catch (ICUException | IllegalStateException e) {
+ // ICU's UTS46 + Punycode conversion throws these unchecked exceptions:
+ // ICUException (incl. ICUInputTooLongException from Punycode.encode on
an
+ // over-long label), IllegalStateException (Punycode).
+ // Convert to MalformedURLException so callers (e.g. BasicURLNormalizer)
+ // reject the URL instead of crashing the task.
+ throw (MalformedURLException) new MalformedURLException(
+ "Invalid IDN host " + host + ": " + e.getMessage()).initCause(e);
}
+
if (idnaInfo.hasErrors()) {
StringBuilder msg = new StringBuilder();
for (IDNA.Error error : idnaInfo.getErrors()) {
diff --git a/src/test/org/apache/nutch/util/TestURLUtil.java
b/src/test/org/apache/nutch/util/TestURLUtil.java
index 200ea59a0..7c7267ce8 100644
--- a/src/test/org/apache/nutch/util/TestURLUtil.java
+++ b/src/test/org/apache/nutch/util/TestURLUtil.java
@@ -444,6 +444,22 @@ public class TestURLUtil {
() -> URLUtil.convertIDNA2008("xn--a-ä.pt", false));
assertThrows(MalformedURLException.class,
() -> URLUtil.convertIDNA2008("xn--a-ä.pt", true));
+
+ assertThrows(MalformedURLException.class, // Thai, 1001 units
+ () -> URLUtil.convertIDNA2008("ก".repeat(1001), true));
+ assertThrows(MalformedURLException.class, // emoji (surrogate pairs), 1002
units
+ () -> URLUtil.convertIDNA2008("😀".repeat(501), true));
+ assertThrows(MalformedURLException.class, // CJK Ext-B, 1200 units
+ () -> URLUtil.convertIDNA2008(
+ new String(Character.toChars(0x20000)).repeat(600), true));
+
+ // boundary: exactly 1000 units does not throw; rejected via LABEL_TOO_LONG
+ assertThrows(MalformedURLException.class,
+ () -> URLUtil.convertIDNA2008("ก".repeat(1000), true));
+
+ // EMPTY_LABEL: a bare dot (empty label) is rejected via IDNA.Info errors
+ assertThrows(MalformedURLException.class,
+ () -> URLUtil.convertIDNA2008(".", true));
}
@Test
