i've mixed up some xml files and now have something like
<Schema>
stuff
</Schema><Schema>
stuff
</Schema><Schema>
..
i need to split them all so to have from <Schema> to </Schema> in each file
Add a comment
|
1 Answer
One way using awk. It splits registers with end tag and if there are characters before it print all it:
awk -c '
BEGIN { RS = "</Schema>" }
$0 ~ /[^[:blank:]\n]/ {
printf "%s\n", $0 RS >> FILENAME "_" ++i ".xml"
}
' infile
Assuming infile with content:
<Schema>
stuff
</Schema><Schema>
more stuff
</Schema><Schema>
and more stuff
</Schema>
It yields:
==> infile_1.xml <==
<Schema>
stuff
</Schema>
==> infile_2.xml <==
<Schema>
more stuff
</Schema>
==> infile_3.xml <==
<Schema>
and more stuff
</Schema>
6 Comments
mfirry
amazing. one cool thing would be to have the output like this "infile_1.xml" "infile_2.xml" and so on
Birei
@mfirry: Simply add
FILENAME variable to the output redirection of the printf. I've updated the answer with it.mfirry
awesome! one more thing before declaring you my saviour is i'm gonna have to put it in a sh file and inside a 'for file in .' kinda thing. it'll hopefully work, right?
Birei
@mfirry: Why should not work? I can't see much problem. That is not a problem of
awk.mfirry
#!/bin/bash for file in $(find . "*.xml") do awk -c ' BEGIN { RS = "</Schema>" } $0 ~ /[^[:blank:]\n]/ { printf "%s\n", $0 RS >> FILENAME "_" ++i } ' $file done with that i get some gawk error Usage: gawk [POSIX or GNU style options] -f progfile [--] file ...
|