Issue Summary

When importing a .ics Calendar that was Exported from Confluence site, attendees are not matched properly when Importing the Calendar on a different Confluence instance.

This is reproducible on Data Center: yes

Steps to Reproduce

1. Confluence Source
   1. Browse to your Calendar tab, and use Export to iCalendar to export a calendar with at least one Attendee in the events
2. Confluence Destination
   1. Ensure the Attendee is a user from the new Confluence instance
   2. Browse to your Calendars tab, and use Import Calendar
   3. In the new prompt, select the ics created before in field Upload iCal

Expected Results

Team Calendars will match the username during import and add the user as attendee to the event.

Actual Results

Imported Calendar doesn't show attendees for the events.

The import is only mapping the "X-CONFLUENCE-USER-KEY" part of the Attendee in the .ics:

ATTENDEE;X-CONFLUENCE-USER-KEY=2cb280858d9d05c1018d9d096fef0000;CN=test.username;CUTYPE=INDIVIDUAL:mailto:noreplay@example.com

If it doesn't exist, it doesn't attempt to match with the CN.

Workaround

Manually change the X-CONFLUENCE-USER-KEY in the Exported Calendar to match the new site values (requires DB access). 

Instructions when migrating from SOURCE to DESTINATION

1. Create database dump of the list of users of each site. Using Postgres, this would be as follows: 

   psql confluence 
   
   COPY user_mapping(username, user_key) TO '/tmp/usermappingSOURCE.csv' DELIMITERS E'\t' CSV;  
   

   Repeat the same in the DESTINATION database. 
   Files will look like:

   username1        ff80808142c21f330142c2215dd20004
   username2        ff80808142c21f330142c2215dd20013
   ...
   

2. Download them to the same location and name them accordingly: "usermappingSOURCE.csv" "usermappingDEST.csv"
3. Create a replacement list: (command explained: here)

   awk 'NR==FNR {h[$1] = $2; next} {print $2,$3,h[$1]}' usermappingDEST.csv usermappingSOURCE.csv  > replaceUSERS.csv
   

4. Remove lines from replaceUSERS.csv that don't have a second value. (That means that the user doesn't exist on DESTINATION)

   awk 'NF > 1' replaceUSERS.csv > replaceMATCHEDUSERS.csv
   

5. Export the calendar/s from SOURCE and replace all keys dynamically on the .ics using the replaceMATCHEDUSERS.csv file: (command explained here)

   while read n k; do 
           sed -i "s/$n/$k/g" calendar.ics; 
   done < replaceMATCHEDUSERS.csv
   

6. Import the modified calendar.ics in DESTINATION