How to set Z as timezone in SimpleDateFormat

问题

Code sample:

SimpleDateFormat dateFormat = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss z");
System.out.println(dateFormat.getTimeZone());
System.out.println(dateFormat.parse(time));
//  dateFormat.setTimeZone(TimeZone.getTimeZone("Asia/Kolkata"));

I don't want to use the commented section.

Zone should be set based on IST that I am giving in the input string:

String time ="2018-04-06 16:13:00 IST";

Current machine zone is: America/New_York. How should I get zone changes to IST based on z?

SimpleDateFormat dateFormat = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss z");
String time ="2018-04-06 18:40:00 IST";
dateFormat.setTimeZone(TimeZone.getTimeZone("Asia/Kolkata"));

Above is running correctly but I don't want to set zone explicitly. It should be choosen based on IST I am giving in input time string.

回答1:

"I don't want to set zone explicitly"

Sorry to disappoint you, but that's not possible with SimpleDateFormat. Timezone abbreviations like IST are ambiguous - as already said in the comments, IST is used in many places (AFAIK, in India, Ireland and Israel).

Some of those abbreviations might work sometimes, in specific cases, but usually in arbitrary and undocumented ways, and you can't really rely on that. Quoting the javadoc:

For compatibility with JDK 1.1.x, some other three-letter time zone IDs (such as "PST", "CTT", "AST") are also supported. However, their use is deprecated because the same abbreviation is often used for multiple time zones (for example, "CST" could be U.S. "Central Standard Time" and "China Standard Time"), and the Java platform can then only recognize one of them.

Due to the ambiguous and non-standard characteristics of timezones abbreviations, the only way to solve it with SimpleDateFormat is to set a specific timezone on it.

"It should be set to based on Z"

I'm not really sure what this means, but anyway...

Z is the UTC designator. But if the input contains a timezone short-name such as IST, well, it means that it's not in UTC, so you can't parse it as if it was in UTC.

If you want to output the date with Z, then you need another formatter set to UTC:

SimpleDateFormat dateFormat = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss z");
String time = "2018-04-06 18:40:00 IST";
dateFormat.setTimeZone(TimeZone.getTimeZone("Asia/Kolkata"));
// parse the input
Date date = dateFormat.parse(time);

// output format, use UTC
SimpleDateFormat outputFormat = new SimpleDateFormat("yyyy-MM-dd HH:mm:ssX");
outputFormat.setTimeZone(TimeZone.getTimeZone("UTC"));
System.out.println(outputFormat.format(date)); // 2018-04-06 13:10:00Z

Perhaps if you specify exactly the output you're getting (with actual values, some examples of outputs) and what's the expected output, we can help you more.

回答2:

    DateTimeFormatter formatter 
            = DateTimeFormatter.ofPattern("uuuu-MM-dd HH:mm:ss z", Locale.ENGLISH);
    String time ="2018-04-06 16:13:00 IST";
    ZonedDateTime dateTime = ZonedDateTime.parse(time, formatter);
    System.out.println(dateTime.getZone());

On my Java 8 this printed

Asia/Jerusalem

So apparently IST was interpreted as Israel Standard Time. On other computers with other settings you will instead get for instance Europe/Dublin for Irish Summer Time or Asia/Kolkata for India Standard Time. In any case the time zone comes from the abbreviation matching the pattern letter (lowercase) z in the format pattern string, which I suppose was what you meant(?)

If you want to control the choice of time zone in the all too frequent case of ambiguity, you may build your formatter in this way (idea stolen from this answer):

    DateTimeFormatter formatter = new DateTimeFormatterBuilder()
            .appendPattern("uuuu-MM-dd HH:mm:ss ")
            .appendZoneText(TextStyle.SHORT,
                    Collections.singleton(ZoneId.of("Asia/Kolkata")))
            .toFormatter(Locale.ENGLISH);

Now the output is