update trailing character cleanup regex in FlexibleDateTimeParser (#763)

NationalSecurityAgency · May 10, 2024 · af83fd0 · af83fd0
1 parent 65add6d
commit af83fd0
Show file tree

Hide file tree

Showing 2 changed files with 6 additions and 2 deletions.
diff --git a/src/main/java/emissary/util/FlexibleDateTimeParser.java b/src/main/java/emissary/util/FlexibleDateTimeParser.java
@@ -51,7 +51,7 @@ public class FlexibleDateTimeParser {
     private static final Pattern REPLACE = Pattern.compile("\t+|[ ]+", Pattern.DOTALL);
 
     /* Remove other junk */
-    private static final Pattern REMOVE = Pattern.compile("<.+?>|=0D$", Pattern.DOTALL);
+    private static final Pattern REMOVE = Pattern.compile("<.+?>$|=0D$", Pattern.DOTALL);
 
     /* timezone - config var: TIMEZONE */
     private static ZoneId timezone = ZoneId.of(DEFAULT_TIMEZONE);
@@ -225,7 +225,8 @@ private static String cleanDateString(final String date) {
             return date;
         }
 
-        String cleanedDateString = date;
+        // date strings over 100 characters are more than likely invalid
+        String cleanedDateString = StringUtils.substring(date, 0, 100);
         cleanedDateString = REPLACE.matcher(cleanedDateString).replaceAll(SPACE);
         cleanedDateString = REMOVE.matcher(cleanedDateString).replaceAll(EMPTY);
         return StringUtils.trimToNull(cleanedDateString);

diff --git a/src/test/java/emissary/util/FlexibleDateTimeParserTest.java b/src/test/java/emissary/util/FlexibleDateTimeParserTest.java
@@ -2,6 +2,7 @@
 
 import emissary.test.core.junit5.UnitTest;
 
+import org.apache.commons.lang3.RandomStringUtils;
 import org.junit.jupiter.api.Assertions;
 import org.junit.jupiter.api.BeforeAll;
 import org.junit.jupiter.api.Test;
@@ -420,6 +421,7 @@ void parse_yyyy_DDD() {
     @Test
     void testCleanDateString() {
         test("2016-01-04 18:20<br>", EXPECTED_NO_SECS, "HTML");
+        test("2016-01-04 18:20<br>br>", EXPECTED_NO_SECS, "HTML");
         test("2016-01-04\t\t18:20", EXPECTED_NO_SECS, "TABS");
         test("2016-01-04        18:20", EXPECTED_NO_SECS, "SPACES");
         test("2016-01-04 18:20=0D", EXPECTED_NO_SECS, "qp'ified ending");
@@ -433,5 +435,6 @@ void testBad() {
         assertNull(FlexibleDateTimeParser.parse("1234", Collections.singletonList(null)));
         test("17.Mar.2016", 0L, "UNKNOWN");
         test("Mon, 2 Feb 2017 06:20:30 PM +0000", 0L, "UNKNOWN");
+        test("2016:01:04 18:20:30 GMT+0000<" + RandomStringUtils.randomAlphanumeric(75) + ">", 0L, "UNKNOWN");
     }
 }