AWK time, date, status and sequence from curl output

Question

Need you your help with the following awk syntax. Below is the output from my curl and I need to refine it a little bit:

INPUT:

        RSYNCA-BACKUP
        RCYNCA 20140517 0021 2182097 2082097
        2014820905820917 10:03:54
        2014820905820917 10:37:43
        0:33:49


        RSYNCB-COPY
        20140517 0020 2082097 1982097 7 6 20
        2014820905820917 09:32:20
        2014820905820917 10:59:20
        1:27:00


        RSYNCC
        RCYNCE 20140517 0021 2182097 2082097
        2014820905820917 10:03:54
        2014820905820917 10:37:43
        0:33:49

        RSYNCD
        20140517 0020 2082097 1982097 7 6 20
        2014820905820917 09:32:20
        2014820905820917 10:59:20
        1:27:00

THE OUTPUT I RECEIVE USING AWK:

RSYNCA-BACKUP|20140502|RCYNCA|10:02:15|10:56:42|0:54:27|FINISHED
RSYNCB-COPY|0022||15:31:06|        |0:06:04|INITIATED

---

Job Name|sequence|date|start time|end time|runtime|status

For job with initiated status there is no end time so the field can be empty

Thats what I am running and getting messed up awk output

awk -v RS='FINISHED|INITIATED' -v OFS='|' '$0 { print $1, $3, $2, $8, RS }'

RSYNCJOBNA|0021|20140502|2014820905820902|FINISHED|INITIATED
RSYNCJOBNA|0022|20140502|2014820905820902|FINISHED|INITIATED

---

My input from curl has additional spaces I guess, that might be the issue, here is a real example:

INITIATED
            RSYNCA
            20140502 0036 3682096 3582096 6 5
            2014820905820902 17:31:08
                0:17:16 ce eque
            INITIATED
            RSYNCA
            20140502 0035 3582096 3482096 6 5
            2014820905820902 17:01:10
                0:47:14 ce eque
            FINISHED
            RSYNCA
            20140502 0034 3482096 3382096 6 5
            2014820905820902 16:31:03
            2014820905820902 17:24:45
            0:53:42
            FINISHED
            RSYNCA
            20140502 0033 3382096 3282096 6 5
            2014820905820902 16:01:09
            2014820905820902 16:47:12
            0:46:03

Answer 1

curl "URL" |
    awk -v OFS='|' '/FINISHED|INITIATED/ {
        status = $1; getline;
        jobname = $1; getline;
        sequence = $2; date = $1; getline;
        start = $2; getline;
        if (status == "FINISHED") { end = $2; getline } else { end = "        " }
        runtime = $1;
        print jobname, sequence, date, start, end, runtime, status;
    }'

The output with your input is:

RSYNCA|0036|20140502|17:31:08|        |0:17:16|INITIATED
RSYNCA|0035|20140502|17:01:10|        |0:47:14|INITIATED
RSYNCA|0034|20140502|16:31:03|17:24:45|0:53:42|FINISHED
RSYNCA|0033|20140502|16:01:09|16:47:12|0:46:03|FINISHED

Answer 2

Here's one way using GNU AWK. Run like:

curl "$URL" | awk -f script.awk

Contents of script.awk:

BEGIN {

    RS="FINISHED|INITIATED"
    OFS="|"
}

s {
    print ( \
        $1, \
        $3, \
        $2, \
        $9, \
        (s == "FINISHED" ? $11 : "        "), \
        ($NF ~ /:/ ? $NF : $(NF-2)), \
        s \
    )
}

{
    s = RT
}

Results:

RSYNCA|0036|20140502|17:31:08|        |0:17:16|INITIATED
RSYNCA|0035|20140502|17:01:10|        |0:47:14|INITIATED
RSYNCA|0034|20140502|16:31:03|17:24:45|0:53:42|FINISHED
RSYNCA|0033|20140502|16:01:09|16:47:12|0:46:03|FINISHED

Alternatively, here's the one-liner:

curl "$URL" | awk 'BEGIN { RS="FINISHED|INITIATED"; OFS="|" } s { print $1, $3, $2, $9, (s == "FINISHED" ? $11 : "        "), ($NF ~ /:/ ? $NF : $(NF-2)), s } { s = RT }'