[Sitemaps] Unit tests depend on system timezone, fixes #238 (#239)

- fix unit test to format data in time zone UTC
- improve documentation of `convertToZonedDateTime`:
  add note that UTC is assumed if no time zone is contained in
  date string

CHANGES.txt

@@ -1,6 +1,7 @@
 Crawler-Commons Change Log
 
 Current Development 0.11-SNAPSHOT (yyyy-mm-dd)
+- [Sitemaps] Unit tests depend on system timezone (kkrugler, sebastian-nagel) #238
 - EffectiveTldFinder: upgrade public suffix list (sebastian-nagel) #219
 - [Sitemaps] Detection and parsing of XML sitemaps fails with whitespace before XML declaration (sebastian-nagel, jnioche) #144
 - [Sitemaps] XMLHandler needs to append text in characters() vs. immediately processing (kkrugler, sebastian-nagel) #226

src/main/java/crawlercommons/sitemaps/AbstractSiteMap.java

@@ -146,13 +146,13 @@ public abstract class AbstractSiteMap {
     }
 
     /**
-     * Convert the given date (given in an acceptable DateFormat), null if the
-     * date is not in the correct format.
+     * Convert the given date (given in an acceptable DateFormat), return null
+     * if the date is not in the correct format.
      * 
      * <p>
      * Dates must follow the <a href="https://www.w3.org/TR/NOTE-datetime">W3C
-     * Datetime format</a> which is similar to <a
-     * href="https://en.wikipedia.org/wiki/ISO_8601">ISO-8601</a> but allows
+     * Datetime format</a> which is similar to
+     * <a href="https://en.wikipedia.org/wiki/ISO_8601">ISO-8601</a> but allows
      * dates with different precisions:
      * </p>
      * 
@@ -171,10 +171,13 @@ public abstract class AbstractSiteMap {
      *      YYYY-MM-DDThh:mm:ss.sTZD (eg 1997-07-16T19:20:30.45+01:00)
      * </pre>
      * 
+     * Note: Greenwich time (UTC) is assumed if the date string does not specify
+     * a time zone.
+     * 
      * @param date
      *            - the date to be parsed
-     * @return the zoned date time equivalent to the date string or NULL parsing
-     *         failed
+     * @return the zoned date time equivalent to the date string or NULL if
+     *         parsing failed
      */
     public static ZonedDateTime convertToZonedDateTime(String date) {
 

src/test/java/crawlercommons/sitemaps/AbstractSiteMapTest.java

@@ -23,7 +23,6 @@ import java.time.Instant;
 import java.time.ZoneId;
 import java.time.ZonedDateTime;
 import java.time.format.DateTimeFormatter;
-import java.util.Date;
 import java.util.Locale;
 import java.util.TimeZone;
 
@@ -36,14 +35,14 @@ public class AbstractSiteMapTest {
         assertNull(AbstractSiteMap.convertToDate("blah"));
         assertNull(AbstractSiteMap.convertToDate(null));
 
-        SimpleDateFormat isoFormatNoTimezone = new SimpleDateFormat("yyyyMMdd", Locale.ROOT);
+        SimpleDateFormat isoFormatShortDate = new SimpleDateFormat("yyyyMMdd", Locale.ROOT);
+        isoFormatShortDate.setTimeZone(TimeZone.getTimeZone("UTC"));
 
-        // For formats where there's no time zone information, the time zone is
-        // undefined, so we can only check on the year/month/day portion of the
-        // result.
-        assertEquals("20140101", isoFormatNoTimezone.format(AbstractSiteMap.convertToDate("2014")));
-        assertEquals("20140601", isoFormatNoTimezone.format(AbstractSiteMap.convertToDate("2014-06")));
-        assertEquals("20140603", isoFormatNoTimezone.format(AbstractSiteMap.convertToDate("2014-06-03")));
+        // For short dates we only check on the year/month/day portion of the result.
+        // Time zone UTC is assumed because short dates do not contain a time zone.
+        assertEquals("20140101", isoFormatShortDate.format(AbstractSiteMap.convertToDate("2014")));
+        assertEquals("20140601", isoFormatShortDate.format(AbstractSiteMap.convertToDate("2014-06")));
+        assertEquals("20140603", isoFormatShortDate.format(AbstractSiteMap.convertToDate("2014-06-03")));
 
         SimpleDateFormat isoFormat = new SimpleDateFormat("yyyyMMdd'T'HHmmss", Locale.ROOT);
         isoFormat.setTimeZone(TimeZone.getTimeZone("UTC"));