parse some additional data from a file to another file 1

[amili]

Hi,

i need help on adding additional data from file 1 to file 2 as follows:-

_file1_

No 250-305 Id 1
No 144-146 Id 1
No 189-190 Id 2
No 1904-1909 Id 2
No 1962-1968 Id 2
No 221-222 Id 7
No 248-251 Id 8
No 257-258 Id 8

_file2_

ID 1
START S216
DONE D125
@@


the output should look like this:-

ID 1
START S216
DONE D125
No 250-305
No 144-146
@@
ID 2
START S325
DONE D340
No 189-190
No 1904-1909
No 1962-1968
@@

and so on..until ID 8. Need help on how to do this using awk. Thanks..

 

[CaKiwi]

This assumes that both files contain the same ids in the same order. Change the file in both getlines to your first file and run it against the second file.

{
  print
  j++
  if (j==1) id2 = $NF
  if (j==3)
  {
    j=0
    if (flg==0)
    {
      flg=1
      if ((getline < "file1") < 1) exit
      a=$0
      id1=$NF
    }
    while (id1==id2)
    {
      print a
      if ((getline < "file1") < 1) exit
      a=$0
      id1=$NF
    }
  }
}

CaKiwi

 

[amili]

Hi CaKiWi,

thanks for your prompt response..

the id in my first file does not follow the same order as in my second file. i did try your code but nothing happens, probably because of the order of the id? as of now, i'm still trying to do it..

 

[amili]

Hi CakiWi,

i tried with a small set of actual data with the same order of id and it seems that it only works for the fist id and ignore the rest like this:-

ID 1
START S216
DONE D125
No 250-305
No 144-146
@@
ID 2
START S325
DONE D340
@@
ID 7
START S587
DONE D600
@@

if possible, could you pls explain to me briefly your code there? i don't really understand especially the flg and getline..tried to read in some of the website but still it is not that clear. thanks

 

[CaKiwi]

This might be a better approach. It assumes that the lines for each id are together in each file but does not rely on order. Run with

awk -f file.awk file2 file1

{
  if (FNR==NR)
  {
    if (k==0)
    {
      id = $NF
      l1[id]=$0
    }
    else if (k==1)
      l2[id]=$0
    else
      l3[id]=$0
    k++
    if (k==3) k=0
    next
  }
  id = $NF
  if (id!=svid)
  {
    print l1[id]
    print l2[id]
    print l3[id]
    svid=id
  }
  print
}

CaKiwi

 

[amili]

i tried the 2nd approach.. some of the existing data in file 2 are gone like the @@ symbol, the ID, START and DONE for certain ids. Not only that, i do have some info (comments) that i put in the file 2 before ID 1 and these comments gone too.

 

[CaKiwi]

Sorry, I thought the @@ lines were just something you put in to show there was more data. Try this.

{
  if (FNR==NR)
  {
    if (k==0)
    {
      id = $NF
      l1[id]=$0
    }
    else if (k==1)
      l2[id]=$0
[COLOR=#ff0000]    else if (k==2)[/color]
      l3[id]=$0
    k++
[COLOR=#ff0000]    if (k==4) k=0[/color]
    next
  }
  id = $NF
  if (id!=svid)
  {
[COLOR=#ff0000]    if (flg) print "@@"
    flg=1[/color]
    print l1[id]
    print l2[id]
    print l3[id]
    svid=id
  }
  print
}
[COLOR=#ff0000]END {print "@@"}[/color]

I'll let you add code to handle the comments at the beginning of the second file

CaKiwi

 

[amili]

Hi,

i've been trying to make it works but still failed...the output are a bit messed up but i am still working on this..

 

[CaKiwi]

Post sample input files, expected output and actual output and I (or someone) will help you further.

CaKiwi

 

[amili]

The samples for the actual input for 2 files and expected output as attached. i put them in a file as it is going to be a bit crowded to post directly here even though i already simplified them. Thanks

 

[CaKiwi]

Maybe this time. This assumes a maximum of 12 items per id in file1. Run with
awk -f file.awk file1 file2

/;;/{print;next}
{
  if (FNR==NR)
  {
    id = $NF
    if (id!=svid)
    {
      k=0
      svid=id
    }
    sub(/ *Id_.*$/,"")
    if (k==0)
      l1[id]=$0
    else if (k==1)
      l2[id]=$0
    else if (k==2)
      l3[id]=$0
    else if (k==3)
      l4[id]=$0
    else if (k==4)
      l5[id]=$0
    else if (k==5)
      l6[id]=$0
    else if (k==6)
      l7[id]=$0
    else if (k==7)
      l8[id]=$0
    else if (k==8)
      l9[id]=$0
    else if (k==9)
      l10[id]=$0
    else if (k==10)
      l11[id]=$0
    else if (k==11)
      l12[id]=$0
    k++
    ln[id]=k
    next
  }
  if ($0 ~ /^ID/) id = $NF
  if ($0 ~ /^@@/)
  {
    if (ln[id]>0) print l1[id]
    if (ln[id]>1) print l2[id]
    if (ln[id]>2) print l3[id]
    if (ln[id]>3) print l4[id]
    if (ln[id]>4) print l5[id]
    if (ln[id]>5) print l6[id]
    if (ln[id]>6) print l7[id]
    if (ln[id]>7) print l8[id]
    if (ln[id]>8) print l9[id]
    if (ln[id]>9) print l10[id]
    if (ln[id]>10) print l11[id]
    if (ln[id]>11) print l12[id]
  }
  print
}

CaKiwi

 

[CaKiwi]

Sorry, I forgot awk has 2 dimensional arrays. This is simpler.

/;;/{print;next}
{
  if (FNR==NR)
  {
    id = $NF
    if (id!=svid)
    {
      k=0
      svid=id
    }
    sub(/ *Id_.*$/,"")
    ll[id,k]=$0
    k++
    ln[id]=k
    next
  }
  if ($0 ~ /^ID/) id = $NF
  if ($0 ~ /^@@/)
  {
    for (j=0;j<ln[id];j++) print ll[id,j]
  }
  print
}

CaKiwi

 

[amili]

i run the latest code and it seems that it works for certain data only but i could see that it almost get to the expected output. i will continue working on this and see how it goes..Thanks so much CaKiWi for your big help