Got a String coming in with this format: YYYY-MM-DD-HH.MM.SS.NNNNNN The Timestamp is coming from a DB2 database. I need to parse it into a java.sql.Timestamp and NOT lose any precison. So far I've been unable to find existing code to parse that far out to microseconds. SimpleDateFormat returns a Date and only parses to milliseconds. Looked at JodaTime briefly and didn't see that would work either.
问题:
回答1:
Have you tried using Timestamp.valueOf(String)
? It looks like it should do almost exactly what you want - you just need to change the separator between your date and time to a space, and the ones between hours and minutes, and minutes and hours, to colons:
import java.sql.*;
public class Test {
public static void main(String[] args) {
String text = "2011-10-02 18:48:05.123456";
Timestamp ts = Timestamp.valueOf(text);
System.out.println(ts.getNanos());
}
}
Assuming you've already validated the string length, this will convert to the right format:
static String convertSeparators(String input) {
char[] chars = input.toCharArray();
chars[10] = ' ';
chars[13] = ':';
chars[16] = ':';
return new String(chars);
}
Alternatively, parse down to milliseconds by taking a substring and using Joda Time or SimpleDateFormat
(I vastly prefer Joda Time, but your mileage may vary). Then take the remainder of the string as another string and parse it with Integer.parseInt
. You can then combine the values pretty easily:
Date date = parseDateFromFirstPart();
int micros = parseJustLastThreeDigits();
Timestamp ts = new Timestamp(date.getTime());
ts.setNanos(ts.getNanos() + micros * 1000);
回答2:
You could use Timestamp.valueOf(String)
. The documentation states that it understands timestamps in the format yyyy-mm-dd hh:mm:ss[.f...]
, so you might need to change the field separators in your incoming string.
Then again, if you're going to do that then you could just parse it yourself and use the setNanos
method to store the microseconds.
回答3:
Here's the intended way to do it:
String timestamp = "2011-10-02-18.48.05.123";
DateFormat df = new SimpleDateFormat("yyyy-MM-dd-kk.mm.ss.SSS");
Date parsedDate = df.parse(timestamp);
Admittedly, it only has millisecond resolution, but in all services slower than Twitter, that's all you'll need, especially since most machines don't even track down to the actual nanoseconds.
回答4:
If you get time as string in format such as 1441963946053 you simply could do something as following:
//String timestamp;
Long miliseconds = Long.valueOf(timestamp);
Timestamp ti = new Timestamp(miliseconds);
回答5:
I believe you need to do this:
- Convert everythingButNano using SimpleDateFormat or the like to everythingDate.
- Convert nano using Long.valueof(nano)
- Convert everythingDate to a Timestamp with new Timestamp(everythingDate.getTime())
- Set the Timestamp nanos with Timestamp.setNano()
Option 2 Convert to the date format pointed out in Jon Skeet's answer and use that.