Re: flattening/rolling up/aggregating a large sorted text file

[EMAIL PROTECTED] writes: Hi, Given a large ascii file (delimited or fixed width) with one ID field and dimensions/measures fields, sorted by dimensions, I'd like to flatten or rollup the file by creating new columns: one for each combination of dimension level, and summing up measures over all

flattening/rolling up/aggregating a large sorted text file

Hi, Given a large ascii file (delimited or fixed width) with one ID field and dimensions/measures fields, sorted by dimensions, I'd like to flatten or rollup the file by creating new columns: one for each combination of dimension level, and summing up measures over all records for a given ID. If

Re: flattening/rolling up/aggregating a large sorted text file

[EMAIL PROTECTED] wrote: Hi, Given a large ascii file (delimited or fixed width) with one ID field and dimensions/measures fields, sorted by dimensions, I'd like to flatten or rollup the file by creating new columns: one for each combination of dimension level, and summing up measures over

Re: flattening/rolling up/aggregating a large sorted text file

Apparently you want to use this data to know how many blue circles, blue squares, red circles and red squares. In other words, I doubt you want to output redundant data columns, you just want this data in a more usable format and that you don't actually need to do multiple passes over it.