We have to use an extra worktime-repoting Tool.
This tool can produce csv-files (we can do xml from csv ) with Issue-Key, User, and amount-of-time spent, date/time start working and date/time end-working (and a lot of others to).

It would be much greater, quicker an less errorous to import these data instead of enter them manually a second time for each ISSUE in JIRA.

.. or it would even be easyer to enter them in a table and "SAVE" several line at once, instead of search each issue individual, and add the Log work done seperate. ( compare to JRA-5034 )