UTF date formatting

[Manoj Raghuwanshi]

Hi,
I am trying to write data to a csv file. It contains a date string.

   OutputStream out = response. getOutputStream();
final byte[] contentBytes = (csvBufff.toString()).getBytes("utf-8");
response. setContentLength(contentBytes.length);
response. setContentType("application/vnd.ms-excel.csv");
out.write(contentBytes);

In the string buffer (cvsBuffer) it is showing the date string in the correct format that I am expecting (mm/dd/YY). But after writing to csv file it is coming in mm/dd/yyyy format. Is it happening due to utf-8 foramtting? If yes please help me how to avoid this date formatting.

Thanks
Manoj

[Jim Yingst]

No, I don't think there's any way that UTF-8 would cause that. It seems more likely that there's some other mix-up. Have you looked at the CSV file yourself, or are you relying on some other program which reads the file and parses that field as a Date?

[Manoj Raghuwanshi]

yes I saw the CSV file.

[Ulf Dittmer]

This is mysterious. You have ascertained that in the StringBuffer the date is in mm/dd, and after it's run through toString and getBytes, it is in dd/mm? You could check whether it's still correct after the toString operation.

[Manoj Raghuwanshi]

No, after toString its is mm/dd/yy but after getBytes(�utf-8�) it becomes mm/dd/yyyy.
If it is 08/05/07 it becomes 08/05/2007

[Manoj Raghuwanshi]

No, after toString its is mm/dd/yy but after getBytes(�utf-8�) it becomes mm/dd/yyyy.
If it is 08/05/07 it becomes 08/05/2007

[Jim Yingst]

Frankly, what you describe does not seem to be possible. I think there's likely been some mixup either before or after the getBytes() method call. To figure this out, I recommend putting some print statements as close as possible to the call, before and after:

OutputStream out = response. getOutputStream();
String csvStr = csvBuff.toString();
 
System.out.println("csvStr=[" + csvStr + "], length=" +  csvStr.length());
 
final byte[] contentBytes = (csvStr).getBytes("utf-8");
 
System.out.println("contentBytes=[" + new String(contentBytes, "utf-8") + "], length=" +  contentBytes.length);
 
response. setContentLength(contentBytes.length);
response. setContentType("application/vnd.ms-excel.csv");
out.write(contentBytes);

This should give a better picture of where the problem really lies.

[Ulf Dittmer]

I just ran the following, and it confirms that all is well in this part of the code (meaning it prints "true").

StringBuffer sb = new StringBuffer("08/05/07");
String str = sb.toString();
byte[] contentBytes = str.getBytes("UTF-8");
System.out.println(contentBytes.length == str.length());

Where are you seeing the "2007"? Some programs -like Excel- format CSV data upon import, thereby turning 2-digit years into 4-digit years, so that would not be a valid test.

[Manoj Raghuwanshi]

Yes Ulf it is indeed Excel issue. Excel is formatting the date. Thanks Ulf,Jim.