I often had the problem, that I had to process query results and provide them as excel files. Writing them on Unix systems. Usualy I wrote comma separated csv-files. But unfortunately when opening them excel begins to "guess" what datatype a specific value might be. So sometimes a number value 4.09 became a date 4-SEP etc.

Reading AskTom one day I found that you could set MARKUP in SQL*PLUS and excel accepts this as a valid format. Playing around a bit I found some useful features and wrote a package with HTML format as output.

Downside: the file can not be read by OpenOffice Calc. A friend gave me a hint about Microsoft XML format for spreadsheeets so I changed the output a little bit. Not every feature is available ib OpenOffice, but most are.

There is a bug in OpenOffice 2.3.1, it cannot open any XML file!



New: Instead of directly writing into a file you now can choose to write into a LOB

  • Can be read by Open Office and MS Office XP/2002. To import into MS Office 2000 see http://support.microsoft.com/kb/923505
  • Limitation: after version 2.0 I dropped the support for OpenOffice, because despite trying to persuade it, it does not say why opening a file fails.
  • Easy to use (sample script in package header)
  • Column headers
  • Top captions spanning one or more column headers
  • Fix header and/or colums when scrolling
  • Predefined output formatting for
    • date with or without time
    • number as percent, 2 digits, integer, unformated, currency, accounting
    • string with markup for bold/underline
  • any self defined format including colors for text and background, font size and borders
  • Output from a given query or cursor. Begins new worksheet automatically after a give number of rows
  • Comments
  • Autofilter (does not work with OpenOffice)
  • Conditional formatting (does not work with OpenOffice). Set individual text color, background etc.
  • More than one worksheet
  • Print setup including margins and titles
  • ...


Look at submenue at right hand side




The software is not intended for use in productive environments. If you do so be sure to test thoroughly because I will not give support in any case. I don't claim that this software is free of errors (it certainly is not).

Other resources

Perhaps owa_sylk from Tom Kyte will suit you better.

Jason Bennett wrote a similar program. The difference is mainly the ways to retrieve the document:

  • as a CLOB
  • through mod_plsql to a web browser
  • in a PL/SQL table via user defined type called ExcelDocumentLine

Sanjeev Sapre does not support borders, formulas and date format.

oracle@matzberger.de For contacts of any kind including error reports. Impressum