Created attachment 81607 [details]
contains 2 files: an OO base database containing the above query and the required tables and an OO calc spreadsheet with the query result sdr_report imported, showing the bug.

When importing into calc from a query result in a registered base database, the first 26 columns are interspersed with blank columns, which have to be deleted by hand before the calc worksheet can be saved as a .CSV file.  In the calc worksheet, row 1 contains the column headings, with the first 26 spanning a double column (two letters) in which the first column contains the data and the column name and the second column is blank.  

The query is as follows:  
SELECT "A"."email", "A"."list", "A"."excluded","A"."date","A"."comment", "A"."softbounces", "B".*  FROM "main" AS "A" LEFT OUTER JOIN "sdr-all" AS "B" ON "A"."email" = "B"."Email_PGM" WHERE "A"."date" > {D '2013-09-20' } ORDER BY "A"."list", "A"."email"

In the above query, table B has 64 columns of information on a person.  The query result contains 6 columns from table A followed by all 64 columns of table B.

The query result is imported by hitting the F4 key in an open, empty calc worksheet, selecting the query from the registered database, right-clicking and selecting "Copy", then clicking on the upper-left cell of the worksheet and pasting.