In the first load script SET TimestampFormat='DD/MM/YYYY h:mm:ss[.fff] TT' is used. In the second load script the timestamp format is changed to SET TimestampFormat='MM/DD/YYYY hh:mm:ss[.fff]'. The different results show how the SET TimeFormat statement works with different time data formats.
The table below shows the data set that is used in the load scripts that follow. The second column of the table shows the format of each timestamp in the data set. The first five timestamps follow ISO 8601 rules but the sixth does not.
Data set
Table showing the time data used and the format for each timestamp in the data set.
| transaction_timestamp |
time data format |
| 2018-08-30 |
YYYY-MM-DD |
| 20180830T193614.857 |
YYYYMMDDhhmmss.sss |
| 20180830T193614.857+0200 |
YYYYMMDDhhmmss.sss±hhmm |
| 2018-09-16T12:30-02:00 |
YYYY-MM-DDhh:mm±hh:mm |
| 2018-09-16T13:15:30Z |
YYYY-MM-DDhh:mmZ |
| 9/30/18 19:36:14 |
M/D/YY hh:mm:ss |
In the Data load editor, create a new section, and then add the example script and run it. Then add, at least, the fields listed in the results column to a sheet in your app to see the result.
Load script
SET FirstWeekDay=0;
SET BrokenWeeks=1;
SET ReferenceDay=0;
SET DayNames='Mon;Tue;Wed;Thu;Fri;Sat;Sun';
SET LongDayNames='Monday;Tuesday;Wednesday;Thursday;Friday;Saturday;Sunday';
SET DateFormat='YYYYMMDD';
SET TimestampFormat='DD/MM/YYYY h:mm:ss[.fff] TT';
Transactions:
Load
*,
Timestamp(transaction_timestamp, 'YYYY-MM-DD hh:mm:ss[.fff]') as LogTimestamp
;
Load * Inline [
transaction_id, transaction_timestamp, transaction_amount, transaction_quantity, discount, customer_id, size, color_code
3750, 2018-08-30, 12423.56, 23, 0,2038593, L, Red
3751, 20180830T193614.857, 5356.31, 6, 0.1, 203521, m, orange
3752, 20180830T193614.857+0200, 15.75, 1, 0.22, 5646471, S, blue
3753, 2018-09-16T12:30-02:00, 1251, 7, 0, 3036491, l, Black
3754, 2018-09-16T13:15:30Z, 21484.21, 1356, 75, 049681, xs, Red
3755, 9/30/18 19:36:14, -59.18, 2, 0.333333333333333, 2038593, M, Blue
];
Results
Qlik Sense table showing results of the TimestampFormat interpretation variable being used in the load script. The last timestamp in the data set does not return a correct date.
| transaction_id |
transaction_timestamp |
LogTimeStamp |
| 3750 |
2018-08-30 |
2018-08-30 00:00:00 |
| 3751 |
20180830T193614.857 |
2018-08-30 19:36:14 |
| 3752 |
20180830T193614.857+0200 |
2018-08-30 17:36:14 |
| 3753 |
2018-09-16T12:30-02:00 |
2018-09-16 14:30:00 |
| 3754 |
2018-09-16T13:15:30Z |
2018-09-16 13:15:30 |
| 3755 |
9/30/18 19:36:14 |
- |
The next load script uses the same data set. However, it uses SET TimestampFormat='MM/DD/YYYY hh:mm:ss[.fff]' to match the non-ISO 8601 format of the sixth timestamp.
In the Data load editor, replace the previous example script with the one below and run it. Then add, at least, the fields listed in the results column to a sheet in your app to see the result.
Load script
SET FirstWeekDay=0;
SET BrokenWeeks=1;
SET ReferenceDay=0;
SET DayNames='Mon;Tue;Wed;Thu;Fri;Sat;Sun';
SET LongDayNames='Monday;Tuesday;Wednesday;Thursday;Friday;Saturday;Sunday';
SET DateFormat='YYYYMMDD';
SET TimestampFormat='MM/DD/YYYY hh:mm:ss[.fff]';
Transactions:
Load
*,
Timestamp(transaction_timestamp, 'YYYY-MM-DD hh:mm:ss[.fff]') as LogTimestamp
;
Load * Inline [
transaction_id, transaction_timestamp, transaction_amount, transaction_quantity, discount, customer_id, size, color_code
3750, 2018-08-30, 12423.56, 23, 0,2038593, L, Red
3751, 20180830T193614.857, 5356.31, 6, 0.1, 203521, m, orange
3752, 20180830T193614.857+0200, 15.75, 1, 0.22, 5646471, S, blue
3753, 2018-09-16T12:30-02:00, 1251, 7, 0, 3036491, l, Black
3754, 2018-09-16T13:15:30Z, 21484.21, 1356, 75, 049681, xs, Red
3755, 9/30/18 19:36:14, -59.18, 2, 0.333333333333333, 2038593, M, Blue
];
Results
Qlik Sense table showing results of the TimestampFormat interpretation variable being used in the load script.
| transaction_id |
transaction_timestamp |
LogTimeStamp |
| 3750 |
2018-08-30 |
2018-08-30 00:00:00 |
| 3751 |
20180830T193614.857 |
2018-08-30 19:36:14 |
| 3752 |
20180830T193614.857+0200 |
2018-08-30 17:36:14 |
| 3753 |
2018-09-16T12:30-02:00 |
2018-09-16 14:30:00 |
| 3754 |
2018-09-16T13:15:30Z |
2018-09-16 13:15:30 |
| 3755 |
9/30/18 19:36:14 |
2018-09-16 19:36:14 |
