Re^6: Pre-process csv files before using

[DrAxeman]

in reply to Re^5: Pre-process csv files before using
in thread Pre-process csv files before using

I'm getting an odd error with this

shift @origcols;
    foreach ( @origcols ) {
       $_ =~ s/\.//;
       push ( @cols , $_ ) unless /Bandwidth.*|MSTCPLoop.*/ ;
    }

It says:

Use of uninitialized value in substitution (s///) at ./runsql3.sql lin
+e 20.
Use of uninitialized value in pattern match (m//) at ./runsql3.sql lin
+e 21.

Line 20 is my "s/\.//" line

[sk]

I would check if there are proper values in your @origcols array.

Do something like this before you get into the loop to modify it

map { print $_,$/ } @origcols;

If you see proper output then need to look at other places. Example of your header data will be helpful!

Actually davidrw's grep solution is much cleaner!

[DrAxeman]

This is a small sample of the header data



PDHCSV40EasternDaylightTime240,ERWWCOMMUNITIES.MemoryPagesPERsec,ERWWC
+OMMUNITIESNetworkInterfaceEthernetAdapterModuleBytesTotalPERsec,ERWWC
+OMMUNITIESNetworkInterfaceEthernetAdapterCurrentBandwidth


In the original data, some of the headers have periods in them. This causes SQL error, so I've been stripping them in a preprocessing program. I'm trying to stream line this, and get rid of the preprocessor.

To test this I've been adding a period in one of the fields. Thats when I get the error. I have inserted a period in the second header value to simulate what I am trying to strip.


20050807 Janitored by Corion: Made page-widening code wrap (only multiline code wraps)