I wish to extracted out elapsedTime attribute values from the file.
Records look
{"realm":"/test","transactionId":"9e26c614","elapsedTime":17,"elapsedTimeUnits":"MILLISECONDS","_id":"9e26c6asdasd"}
The file I am having is in gb's and I want to get the values greater than 10000.
I tried to grep but due to colon grep is not working.
grep -wo --color 'elapsedTime' fileName -> this just prints attribute names
grep -w --color "elapsedTime" fileName -> this just highlights the attribute.
The data is JSON format so it's best to use a parser that understands this format.
This will pick out the elapsedTime value from the JSON in the file /tmp/data
jq .elapsedTime /tmp/data
17
This will pick out only those values larger than 10000
jq '.elapsedTime | select(. > 10000)' /tmp/data
If you really cannot use jq then a sed|awk construct can be considered. However, this requires that there must be only one elapsedTime label and associated value per line. There may be other caveats and I really do not recommend it, but if you're desperate here it is,
sed -En 's/^.*"elapsedTime":\s*"?([0-9]+).*$/\1/p' /tmp/data |
awk '$1 > 10000'
In response to a follow-up question (comment), to pick out two elements you need to filter on a single element from the object, and then display the required elements:
jq -r 'select (.elapsedTime > 10000) | [ .elapsedTime, .transactionId ] | @tsv ' /tmp/data
If jq not available, sed can be used.
ELAP=$(sed 's/.*"elapsedTime":\([0-9]*\).*/\1/' filename)
