LinuxQuestions.org - extracting elements

I'm stuck between a rock and a hard place...hopefully you can help.

I have a text file filled with all sorts of junk....sprinkled throughout are "proprietary" markup tags. Here's an example:

Code:

COM blah 

COM blah

{{foo}}

  {{bar}}1{{/bar}} {{baz}}<MNFRAME_VAR_HERE>{{/baz}}

  {{blah}}<...>{{/blah}}    {{tag5}}off{{/tag5}}

COM junk I don't care about

COM junk I don't care about

{{foo}}

  {{bar}}2{{/bar}} {{baz}}<MNFRAME_VAR_HERE>{{/baz}}

  {{blah}}<...>{{/blah}}    {{tag5}}on{{/tag5}}

none of the "elements" have attributes....they are simply markup for variables coming from the mainframe
there can be multiple "elements" on a line
the white space between elements on the same line is unknown. you will see
{{...}}{{....}} {{....}}
on one line then see
{{...}} {{....}} {{....}}
on the next
there is no closeing {{foo}}. That's in a different file

what I need is a script that will extract all the {{???}} elements. My plan (at this point) is to pipe that through "sort -u" and create a DB table with a column for each unique element.

I'm still early in my research on the best way to accomplish my task, but in any case, a bash/sed/awk/perl script that would return my unique elements would be very helpful.

basically....a non-greedy 'grep -e "{{.*}}" filename'

any help is appreciated....

thanks,

-sky