AWK. generate limited files from xml

frambau · 02-10-2012, 06:41 AM

Hi,

I need to extract the tag content in a xml to a diferent secuentials files. One for each ocurrence of this tag

i use this sentence

H="try_"
awk -v name=$H 'BEGIN {RS = "[<>]"}{file = (name)(++i)".out"; ORS = /description/?"\n":"|"}/item/,/\/item/{if(/^(description)$/ && getline)print > file; close(file) }' inputfile

but i have two problems.
The first is the limit of files created. I can only generate 1000 files.
My second problem in this case is the name. This is not secuential.

ANY HELP PLEASE!!!!

Cedrik · 02-10-2012, 06:59 AM

I don't think you need to close(file)

What is the problem with "only generate 1000 files", are you sure there is more than 1000 <item>.*</item>
in the inputfile ?

grail · 02-10-2012, 07:00 AM

Quote:

The first is the limit of files created. I can only generate 1000 files.

Why do you believe this? As an example the following works without issue:

Code:

awk 'BEGIN{while(i++ < 1050)print "in the file" > "file"i}'

I see no reason why you would be limited by awk to create more than 1000 files.

Also, you seem to be creating multiple questions all based on the same information. Is there a reason you are not sticking with your original question?

Cedrik · 02-10-2012, 07:02 AM

Quote:

Originally Posted by grail

Also, you seem to be creating multiple questions all based on the same information. Is there a reason you are not sticking with your original question?

Yes, it gets tedious to follow

colucix · 02-10-2012, 07:09 AM

Quote:

Originally Posted by Cedrik

Yes, it gets tedious to follow

I agree. Furthermore the input file is not posted in this thread, therefore we have to look at your posting history to fully understand the question. For this reason I report this thread for merging with your previous one (leaving your very first one alone). Please, refrain from splitting your issue into multiple threads in the future. Thank you.

frambau · 02-10-2012, 07:11 AM

Sorry so much. This error is mine

THANKS again.