Shell Programming and Scripting

Splitting the file for every 10000 lines will work perfectly. But i still have know how many files have been created? And have to get the filenames from the directory to write them out as a xml.

It would be good if can have all of them in the original file and create a xml file for each 10000 lines.

---------- Post updated at 10:19 AM ---------- Previous update was at 09:41 AM ----------

Filenames generated by the awk script looks like

Code:

file.txt.0
file.txt.1
file.txt.2

How to get the number of files generated after this? Or it would be simple if we can able to append a temporary string in front of the filename to capture the temp files and write them as xmls

Say
tempfile1.txt
tempfile2.txt
tempfile3.txt

Code:

maxline=10000
awk -vm="$maxline" 'END{print "The number of file generated will be : " int(NR/m)+!!(NR%m)}' yourinitialfile.txt

borrowing from stsgnb:

Code:

awk '!(NR%10000){close(f);c++;=FILENAME"."c}{print $0 > f }END{print "created " c " files"}' yourfile

Thanks a lot.. How to assign this value to a shell variable?

Oooops lol, thx versh99

By the way i just tested the code and got an error message (even adding the missing f=FILENAME...)

Code:

awk: cannot open "" for output (No such file or directory)

And this code can also lead to bad file calculation if $maxline is an exact multiple of the number of line of the initial file because c would be incremented once too more.

so maybe there are some more simple way to write it but at least, the following should work:

Code:

maxline=10000
awk -vm="$maxline" 'BEGIN{c=1}{f=FILENAME"."c}NR%m==0{print $0 > f;close(f);c++;next}{print $0 > f }END{print int(NR/m)+!!(NR%m) " files generated" }' yourfile

We have declared the variable 'c' here but how it got assigned to 0.

Also what exactly the below code snippet will do? Just trying to understand.

Code:

{print $0 > f }

Code:

{print $0 > f}

does just print the current scanned line into the file f

The trick it that we make f refering to a name which varies depending the Number of Records (NR) (which -in our case- is the same as the line number since the record separator (RS) is by default the newline (\n) character)