Agfdhyk

I have a text file named as file1.txt .and its need to be aligned like below expected output .
Note:The input file file1.txt is TAB delemitted and its always aligned.

                      DISK OK - free space:          CRITICAL

01-08-2018 07:05:05   Service Required     Critical  CPU:loadaverage 6.0%                    



01-08-2018 07:10:25   Service Alert        Critical  memoryUsage

                                                     :critical alert

                       DISK OK - free space:

02-08-2018 01:05:20    Service Alert       Warning   memoryUsage

                                                     :1.0,2.0,5.0





                                                     CRITICAl:outstanding alert attention 

02-08-2018 02:05:20    Service Alert       Critical  required 

Expected Output:

 01-08-2018 07:05:05    DISK OK - free space:Service Required Critical    CRITICALservice requiredCPU:loadaverage 6.0%



 01-08-2018 07:10:25    Service Alert                         Critical    memoryUsage:critical alert



 02-08-2018 01:05:20    DISK OK - free space:Service Alert    Warning     memoryUsage:1.0,2.0,3.0



 02-08-2018 02:05:20    Service Alert                         Critical     CRITICAL:outstanding alert attention required

Any suggestions are much appreciated.

You can use awk to pull this off:

 awk -F"t" '!$1{for(i=1;i<=NF;i++){line[i]=$i};next} {for(i=1;i<=NF;i++){printf i<NF?"%s%s%s":"%s%s%sn",line[i],$i,FS}}' yourfile

If you need the output all aligned like you have it here (not tab delimited, but more like fixed width) you can pipe to column

awk -F"t" '!$1{for(i=1;i<=NF;i++){line[i]=$i}} $1{for(i=1;i<=NF;i++){printf i<NF?"%s%s%s":"%s%s%sn",line[i],$i,FS}}' test.log | column -t -s $'t'

Basically what this is doing is:

1. Splitting each line by tab -F"t"
   


2. If the first column is empty, then collect each column for this line into an array by the column number as index and proceed to the next record!$1{for(i=1;i<=NF;i++){line[i]=$i};next}
   


3. If we are still processing the line (that first condition didn't trip) then go through each column {for(i=1;i<=NF;i++)
   


4. and print out what's stored in the array, the contents of the column of the current line, the Field Separator (tab), and a line feed if this is the last column {printf i<NF?"%s%s%s":"%s%s%sn",line[i],$i,FS}}
   

Example of this in use:

$ cat test.log

        DISK OK - free space:           CRITICAL

1/8/2018 7:05:05        Service Required        Critical        CPU:loadaverage 6%



1/8/2018 7:10:25        Service Alert   Critical        memoryUsage

                        :critical alert

        DISK OK - free space:

2/8/2018 1:05:20        Service Alert   Warning memoryUsage

                        :1.0,2.0,5.0





                        CRITICAl:outstanding alert attention

2/8/2018 2:05:20        Service Alert   Critical        required



$ awk -F"t" '!$1{for(i=1;i<=NF;i++){line[i]=$i};next} {for(i=1;i<=NF;i++){printf i<NF?"%s%s%s":"%s%s%sn",line[i],$i,FS}}' test.log | column -t -s $'t'

1/8/2018 7:05:05  DISK OK - free space:Service Required  Critical  CRITICALCPU:loadaverage 6%

1/8/2018 7:10:25  Service Alert                          Critical  memoryUsage

2/8/2018 1:05:20  DISK OK - free space:Service Alert     Warning   memoryUsage

2/8/2018 2:05:20  Service Alert                          Critical  CRITICAl:outstanding alert attentionrequired