# putting the end tag /</idsDoc> in the right place | |
BEGIN {DOCCOUNT = 0;} | |
$0 ~ "</idsDoc>" {} | |
$0 ~ "<idsDoc" { | |
if(DOCCOUNT++ > 0){ | |
print " </idsDoc>"; | |
}; | |
print $0; | |
} | |
$0 ~ "</idsCorpus>" {print " </idsDoc>\n</idsCorpus>"} | |
$0 !~ "</idsDoc>" && $0 !~ "</idsCorpus>" && $0 !~ "<idsDoc" {print} | |