I have to sort before I remove the lines at the top because the lines that have the zeros in column 5 are not at the top. The whole point of the task is not to sort the data, but to filter unneeded data. Some zeroes in column 5 are okay, but the redundant ones are the ones at the top after sorting by col 4 then by col 3.
The columns are tab delimited, then the entries are right justified using spaces. TIA. - Bryan __________________ Bryan R Harris wrote: > > I suppose it does look a little bizarre. Actually, my goal is a little > more complex. We have a simulation that outputs data files, but often up > to 90% of the data is redundant. So I'm trying to write a filter for the > data. I have to: > > 1. open and load the file > 2. strip all comments (marked with a #) and blank lines off the top > 3. sort by column 4, then by column 3 > 4. remove all lines at the top that have a 0 in column 5 > 5. write the comments + sorted lines back out to a new file I would change the order to: 1. open and load the file 2. strip all comments (marked with a #) and blank lines off the top 3. remove all lines at the top that have a 0 in column 5 4. sort by column 4, then by column 3 5. write the comments + sorted lines back out to a new file Why sort something that you are going to remove anyway? > Unfortunately I'm new enough at perl that I've only got steps 1 and 2 to > work so far... I got some great help here on the list for step 3, though > the code at (URL for analysis of Schwartzian Transform by tchrist) > seems much more concise. I just haven't gotten it to work on my array > instead of a string. How are your columns defined? Fixed width? Space separated? John -- use Perl; program fulfillment -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
