How can I append field names in a file in linux at run time

Question

i want to do something like this:

I have a script :

latestdate=$(ec2-describe-snapshots | grep ^SNAPSHOT | sort -rk 5 | awk '{print substr($5, 1, 10); exit}')
ec2-describe-snapshots | grep "^SNAPSHOT.*$latestdate" > "$EC2_HOME/Working/SnapshotsLatest_$today_date"

This code is writing the content of ec2-describe-snapshots command in a file. The content of the file is

SNAPSHOT    snap-1062e  vol-aef8    completed   2013-12-12T05:38:28+0000    100%    109030037527    20  2013-12-12: Daily Backup for i-39 (VolID:vol-aef8 InstID:i-3e2)
SNAPSHOT    snap-1c422  vol-f66a0   completed   2013-12-12T05:38:27+0000    100%    109030037527    10  2013-12-12: Daily Backup for i-211 (VolID:vol-9a0 InstID:i-211)

I want to add headings to the file at run time like field names Snapshot, snapshot id,volume id etc with correct formatting. How can i achieve that?

output after @janos answer :

Label       SnapshotID  VolID       Status      Date                                              
SNAPSHOT    snap-626fee5cvol-aecbbcf8completed   2013-12-13T04:53:18+0000  100%  109030037527  20  2013-12-13: Daily Backup for Jst with i-3d09 (Vol ID:vol-af8 Inst ID:i-3e209) 
SNAPSHOT    snap-686fee56vol-f66409a0completed   2013-12-13T04:53:16+0000  100%  109030037527  10  2013-12-13: Daily Backup for_Test_Machine with i-26011 (Vol ID:vol-f66a0 Inst ID:i-260111) 

Thanks

Answer 1

Assuming you want to get an output like this:

Label     SnapshotID  VolID      Status     Date
SNAPSHOT  snap-1062e  vol-aef8   completed  2013-12-12T05:38:28+0000  100%  109030037527  20
SNAPSHOT  snap-1c422  vol-f66a0  completed  2013-12-12T05:38:27+0000  100%  109030037527  10

You can use printf to format your header columns with the right amount of spaces like this:

printf "%-12s%-12s%-12s%-12s%-28s%-8s\n" Label SnapshotID VolID Status Date

For example the formatting placeholder %-12s means: format as a 12-character wide string, aligned to the left.

You could create your file with such header followed by the list of snapshots by first redirecting the header line with > and then the list of snapshots by >> to append to the same file:

out="$EC2_HOME/Working/SnapshotsLatest_$today_date"
printf "%-12s%-12s%-12s%-12s%-28s%-8s\n" Label SnapshotID VolID Status Date > "$out"
latestdate=$(ec2-describe-snapshots | grep ^SNAPSHOT | sort -rk 5 | awk '{print substr($5, 1, 10); exit}')
ec2-describe-snapshots | grep "^SNAPSHOT.*$latestdate" >> "$out"

But even better to put all the printing in one block and redirect the whole thing in one swift move:

latestdate=$(ec2-describe-snapshots | grep ^SNAPSHOT | sort -rk 5 | awk '{print substr($5, 1, 10); exit}')
out="$EC2_HOME/Working/SnapshotsLatest_$today_date"
{
printf "%-12s%-12s%-12s%-12s%-28s%-8s\n" Label SnapshotID VolID Status Date
ec2-describe-snapshots | grep "^SNAPSHOT.*$latestdate"
} > "$out"

Ok I see the problem. The columns in the output of ec2-describe-snapshots are separated by a tab. You could print a header where the labels are separated by a tab like this:

printf "Label\tSnapshotID\tVolID\tStatus\tDate\n"

But I'm not sure if that will be good enough for you.

If that's not good enough, then I think you'll have to reformat the output of ec2-describe-snapshots. Since you are in Linux, the column command can help you here. Here's an imperfect but easily readable solution:

latestdate=$(ec2-describe-snapshots | grep ^SNAPSHOT | sort -rk 5 | awk '{print substr($5, 1, 10); exit}')
out="$EC2_HOME/Working/SnapshotsLatest_$today_date"
{
echo Label SnapshotID VolID Status Date
ec2-describe-snapshots | grep "^SNAPSHOT.*$latestdate"
} | column -t > "$out"

UPDATE

Although you say the above is not working, I cannot reproduce what you're getting. I'm getting good columnized formatting here, like in the sample at the top of my answer. So I think you're doing something wrong.

Anyway, here's an alternative, uglier solution:

latestdate=$(ec2-describe-snapshots | grep ^SNAPSHOT | sort -rk 5 | awk '{print substr($5, 1, 10); exit}')
out="$EC2_HOME/Working/SnapshotsLatest_$today_date"
{
echo Label SnapshotID VolID Status Date
ec2-describe-snapshots | grep "^SNAPSHOT.*$latestdate"
} | awk '{printf "%-12s%-16s%-16s%-12s%-26s%-6s%-14s%-4s", $1, $2, $3, $4, $5, $6, $7, $8; for (i=9; i<=NF; ++i) printf $i " "; print ""; }' > "$out"

This may not work well with all possible outputs of ec2-describe-snapshots. For example this last script assumes that the SnapshotID column is always less than 16 characters. If you notice that some columns are printed without a space between them, then adjust the format string in the printf statement there to increase the width of columns.

Answer 2

Yo can do like this with awk with redirecting the output to file. I have put your output of command in a file aa.

latestdate=$(ec2-describe-snapshots | grep ^SNAPSHOT | sort -rk 5 | awk '{print substr($5, 1, 10); exit}') > "$EC2_HOME/Working/SnapshotsLatest_$today_date"

> cat aa | grep  "^SNAPSHOT.*$latestdate" | awk ' BEGIN { printf "%-12s%-12s%-12s%-12s%-28s%-8s\n", "SNAPSHOT",  "SnapShotID",  "VolumeID"," etc"," etc"," etc"," etc"} 1

'

SNAPSHOT    SnapShotID  VolumeID     etc         etc                         etc
SNAPSHOT    snap-1062e  vol-aef8    completed   2013-12-12T05:38:28+0000    100%    109030037527    20  2013-12-12: Daily Backup for i-39 (VolID:vol-aef8 InstID:i-3e2)
SNAPSHOT    snap-1c422  vol-f66a0   completed   2013-12-12T05:38:27+0000    100%    109030037527    10  2013-12-12: Daily Backup for i-211 (VolID:vol-9a0 InstID:i-211)

Answer 3

echo "headers go here" > "$EC2_HOME/Working/SnapshotsLatest_$today_date"
ec2-describe-snapshots | grep "^SNAPSHOT.*$latestdate" >> "$EC2_HOME/Working/SnapshotsLatest_$today_date"

Note the change to the redirection operator.