-------------------------------------------------------------------------------
FLIRT: "First Letter Indicates Row Type"

From http://c2.com/cgi/wiki?FlirtDataTextFormat

A simple one item per line data file format, for saving multiple tables into a
single file.

  # comment (though usually not required)
  @ - section marker   (table,columns,record)
  $ - data in that section  (defining or as defined by 'columns' section)
  space -- line continuation (variation)

This is simplier than CSV as only one real separator (followed by a marker) is
used, thus requiring only one escape sequence (for return) to be used.
specifically the literal string, " \n\ ",  which is typically long enough that
it is unlikely to occur naturally in data.

If a line does not start with one of these it is likely somebody forgot to
escape a line-feed. Format also allows for empty records ($ on its own)

The @ marker seperates three types of sections
  @table     one data item follows defining the table name
  @columns   defining the column headers
             (and in a variant: type, length, format)
  @records   the data for one record (entry or row) in the table
             The number of records should match number of columns'
             unless record ends in a variable length array of values

Example FLIRT data file...

  @table
  $product_log
  @columns
  $date
  $price
  $shortdescription
  $longdescription
  @record
  $12-DEC-2002
  $27.93
  $red t-shirt
  $This red t-shirt is a perfect gift \n\ It also comes with a free tie.
  @record
  $01-JAN-2003
  $15.79
  $blue pants
  $These pants are for the winter \n\ They also can be cleaned with bleach.
  @record
  $Etc..

Variation: types and maximum lengths for each column (for databases)
Though this is not intended use.

  @columns
  $date,date
  $price,number,12,2
  $shortdescription,char,20
  $longdescription,char,100
  @records

-------------------------------------------------------------------------------
Variation (personal use):

This is not strictly FLIRT, but uses a FLIRT data style

I used this for my Account Distribution Data File, which first lists
various file 'sets' and then has a list of various accounts I have
and how I can log into that account, be it ssh, rsh, telnet, or ftp,
or just a specific 'file distribution set' for copying to USB devices.

   # comment
   = assigmment of file groups and exclusions
   ] ssh distribution, automatically distributed overnight.
   ) rsh distribution (depreciated)
   [ ssh login but no automatic file distribution
   ( rsh login but no automatic file distribution (depreciated)
   { special file distribution set (no specific destination)

The exact format of each line type forms a multi-column white space seperated
items, such as login name, machine name and what file sets should be
distributed, or other special labels.

-------------------------------------------------------------------------------