output the occurence number in sed or grep results in every regex match

mbaste2 · 04-05-2011, 03:50 AM

i have a large plain text kml file (example below) :

<Polygon>
<coordinates>
120,14
121,14
...
...
...
121,13
</coordinates>
</Polygon>
<Polygon>
<coordinates>
130,15
131,15
...
...
...
131,14
</coordinates>
</Polygon>

where i need to replace every line containing <Polygon> with
<Polygon>
where n is the nth occurence of the match in the entire file.

search results i have encountered teach only how to replace the (say) fifth occurence of a match in the file. what i need is to display the nth occurence everytime there is a match.

is this possible? the file is very large.

any help, suggestion, recommendation welcome. thanks.

kurumi · 04-05-2011, 04:43 AM

Ruby(1.9+)

Code:

$ ruby -pne 'BEGIN{c=0}; c+=1 if /<Polygon>/; $_.gsub!(/(<Polygon>)/,"<!-- #{c} --> \\1" ) if /<Polygon>/ ' file
<!-- 1 --> <Polygon>
<coordinates>
120,14
121,14
...
...
...
121,13
</coordinates>
</Polygon>
<!-- 2 --> <Polygon>
<coordinates>
130,15
131,15
...
...
...
131,14
</coordinates>
</Polygon>

Kenhelm · 04-05-2011, 10:50 AM

Code:

awk '/<Polygon>/{printf "<!-- "++n" -->"}1' infile > outfile

mbaste2 · 04-06-2011, 01:58 AM

thanks to both replies.
awk worked as expected.
i'm sure that will work in ruby too, but i don't know ruby yet.