DateFormat parses 2-digit years other than "00" through "99". Fix restricts parsing to those 2-digit year strings.

X-Trac-URL: https://ssl.icu-project.org/trac/ticket/28



git-svn-id: http://source.icu-project.org/repos/icu/icu/trunk@64 0a8b766c-62c9-45f9-954d-7f2943c9ead0

source/i18n/smpdtfmt.cpp

@@ -24,6 +24,8 @@
 *                             Removed subParseLong
 *                             Removed chk
 *  02/22/99     stephen     Removed character literals for EBCDIC safety
+*   10/14/99    aliu        Updated 2-digit year parsing so that only "00" thru
+*                           "99" are recognized. {j28 4182066}
 ********************************************************************************
 */
 
@@ -1117,12 +1119,15 @@ int32_t SimpleDateFormat::subParse(const UnicodeString& text, int32_t& start, UC
     case kEraField:
         return matchString(text, start, Calendar::ERA, fSymbols->fEras, fSymbols->fErasCount);
     case kYearField:
-        // If there are 4 or more YEAR pattern characters, this indicates
+        // If there are 3 or more YEAR pattern characters, this indicates
         // that the year value is to be treated literally, without any
         // two-digit year adjustments (e.g., from "01" to 2001).  Otherwise
         // we made adjustments to place the 2-digit year in the proper
-        // century -- unless the given year itself is more than two characters.
-        if (count <= 2 && (pos.getIndex() - start) <= 2)
+        // century, for parsed strings from "00" to "99".  Any other string
+        // is treated literally:  "2250", "-1", "1", "002".
+        if (count <= 2 && (pos.getIndex() - start) == 2
+            && Unicode::isDigit(text.charAt(start))
+            && Unicode::isDigit(text.charAt(start+1)))
         {
             // Assume for example that the defaultCenturyStart is 6/18/1903.
             // This means that two-digit years will be forced into the range

source/i18n/smpdtfmt.h

@@ -23,6 +23,8 @@
 *                            Removed subParseLong
 *                            Removed getZoneIndex (added in DateFormatSymbols)
 *   06/14/99    stephen     Removed fgTimeZoneDataSuffix
+*   10/14/99    aliu        Updated class doc to describe 2-digit year parsing
+*                           {j28 4182066}.
 *******************************************************************************
 */
 
@@ -135,6 +137,28 @@ class DateFormatSymbols;
  * marker 'a' is left out from the format pattern while the "hour in am/pm"
  * pattern symbol is used. This information loss can happen when formatting the
  * time in PM.
+ *
+ * <p>
+ * When parsing a date string using the abbreviated year pattern ("y" or "yy"),
+ * SimpleDateFormat must interpret the abbreviated year
+ * relative to some century.  It does this by adjusting dates to be
+ * within 80 years before and 20 years after the time the SimpleDateFormat
+ * instance is created. For example, using a pattern of "MM/dd/yy" and a
+ * SimpleDateFormat instance created on Jan 1, 1997,  the string
+ * "01/11/12" would be interpreted as Jan 11, 2012 while the string "05/04/64"
+ * would be interpreted as May 4, 1964.
+ * During parsing, only strings consisting of exactly two digits, as defined by
+ * <code>Unicode::isDigit()</code>, will be parsed into the default century.
+ * Any other numeric string, such as a one digit string, a three or more digit
+ * string, or a two digit string that isn't all digits (for example, "-1"), is
+ * interpreted literally.  So "01/02/3" or "01/02/003" are parsed, using the
+ * same pattern, as Jan 2, 3 AD.  Likewise, "01/02/-3" is parsed as Jan 2, 4 BC.
+ *
+ * <p>
+ * If the year pattern has more than two 'y' characters, the year is
+ * interpreted literally, regardless of the number of digits.  So using the
+ * pattern "MM/dd/yyyy", "01/11/12" parses to Jan 11, 12 A.D.
+ *
  * <P>
  * For time zones that have no names, SimpleDateFormat uses strings GMT+hours:minutes or
  * GMT-hours:minutes.

source/test/intltest/calregts.cpp

@@ -70,6 +70,7 @@ CalendarRegressionTest::runIndexedTest( int32_t index, bool_t exec, char* &name,
         CASE(32,Test4165343) 
         CASE(33,Test4166109) 
         CASE(34,Test4167060) 
+        CASE(35,Test4197699)
 
     default: name = ""; break;
     }
@@ -1580,8 +1581,53 @@ CalendarRegressionTest::Test4167060()
     delete calendars[2];
 }
 
+/**
+ * Week of year is wrong at the start and end of the year.
+ */
+void CalendarRegressionTest::Test4197699() {
+    UErrorCode status = U_ZERO_ERROR;
+    GregorianCalendar cal(status);
+    cal.setFirstDayOfWeek(Calendar::MONDAY);
+    cal.setMinimalDaysInFirstWeek(4);
+    SimpleDateFormat fmt("E dd MMM yyyy  'DOY='D 'WOY='w",
+                         Locale::US, status);
+    fmt.setCalendar(cal);
+    if (FAILURE(status)) {
+        errln("Couldn't initialize test");
+        return;
+    }
 
+    int32_t DATA[] = {
+        2000,  Calendar::JANUARY,   1,   52,
+        2001,  Calendar::DECEMBER,  31,  1,
+    };
+    int32_t DATA_length = sizeof(DATA) / sizeof(DATA[0]);
 
+    UnicodeString str;
+    DateFormat& dfmt = *(DateFormat*)&fmt;
+    for (int32_t i=0; i<DATA_length; ) {
+        cal.clear();
+        cal.set(DATA[i], DATA[i+1], DATA[i+2]);
+        i += 3;
+        int32_t expWOY = DATA[i++];
+        int32_t actWOY = cal.get(Calendar::WEEK_OF_YEAR, status);
+        if (expWOY == actWOY) {
+            logln(UnicodeString("Ok: ") + dfmt.format(cal.getTime(status), str.remove()));
+        } else {
+            errln(UnicodeString("FAIL: ") + dfmt.format(cal.getTime(status), str.remove())
+                  + ", expected WOY=" + expWOY);
+            cal.add(Calendar::DATE, -8, status);
+            for (int j=0; j<14; ++j) {
+                cal.add(Calendar::DATE, 1, status);
+                logln(dfmt.format(cal.getTime(status), str.remove()));
+            }
+        }
+        if (FAILURE(status)) {
+            errln("FAIL: Unexpected error from Calendar");
+            return;
+        }
+    }
+}
 
 UDate
 CalendarRegressionTest::makeDate(int32_t y, int32_t m, int32_t d,

source/test/intltest/calregts.h

@@ -65,6 +65,7 @@ public:
     void Test4165343(void) ;
     void Test4166109(void) ;
     void Test4167060(void) ;
+    void Test4197699();
 
     void printdate(GregorianCalendar *cal, char *string);
     void dowTest(bool_t lenient) ;