Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

Register Log in
  • Congratulations strongm on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Merge using AWK 2

Status
Not open for further replies.
stu78

stu78

Programmer
May 29, 2002
121
GB
Hi,

I'm not an expert in AWK, but I've been advised awk could help with this problem...

I have the following file I'd been trying to merge with SQL within a database, basically, I've had to merge AD users and group information, which in SQL is not an issue. However, it's producing me output (for many users, that looks like this;


| AccountDisabled | CanonicalNameUser | City | Comment | CommonName | Company | Country | CreationDate | Department | Description | DisplayName | DistinguishedName | GroupName |

| FALSE | client.local/client/Accounts/user1 | Town | | user1 | company name | | 02/12/2009 17:02 | Department1 | | User 1 | CN=user1,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G1 |
| FALSE | client.local/client/Accounts/user1 | Town | | user1 | company name | | 02/12/2009 17:02 | Department1 | | User 1 | CN=user1,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G2 |
| FALSE | client.local/client/Accounts/user1 | Town | | user1 | company name | | 02/12/2009 17:02 | Department1 | | User 1 | CN=user1,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G3 |
| FALSE | client.local/client/Accounts/user1 | Town | | user1 | company name | | 02/12/2009 17:02 | Department1 | | User 1 | CN=user1,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G4 |
| FALSE | client.local/client/Accounts/user2 | Town | | user2 | company name | | 02/12/2009 17:02 | Department1 | | User 2 | CN=user2,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G7 |
| FALSE | client.local/client/Accounts/user3 | Town | | user3 | company name | | 02/12/2009 17:02 | Department1 | | User 3 | CN=user3,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G7 |
| FALSE | client.local/client/Accounts/user3 | Town | | user2 | company name | | 02/12/2009 17:02 | Department1 | | User 2 | CN=user2,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G8 |

So you can see, that it's listing me all all groups, with the users also.

What I wanted to do in SQL was produce output like;

FALSE | client.local/client/Accounts/user1 | Town | | user1 | company name | | 02/12/2009 17:02 | Department1 | | User 1 | CN=user1,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G1,Grp_G2,Grp_G3,Grp_G4 |
| FALSE | client.local/client/Accounts/user2 | Town | | user2 | company name | | 02/12/2009 17:02 | Department1 | | User 2 | CN=user2,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G7 |
| FALSE | client.local/client/Accounts/user3 | Town | | user3 | company name | | 02/12/2009 17:02 | Department1 | | User 3 | CN=user3,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G7,Grp_G8 |

So basically, I need to scan the file for the column DistinguishedName e.g. CN=user1,OU=Accounts,OU=Client,DC=client,DC=local
If it appears in the subsequent rows, merge the attribute contained in GroupName into the first Groupname Value
If it doesn't appear in the subsequent rows, then just continue, but also put this into the output
 
Sort by date Sort by votes
I assumed that the last line in your input data should have been for user3 and that lines to be merged always appear consecutively in the file
Code: 
BEGIN {FS="|"}
NR<3{print;next}
{
  if (cn!=$13) {
    if (NR>3) print a,b,"|"
    cn=$13
    b=$14
    a=$0
    sub(/[^\|]*\|$/,"",a)
  } else {
    sub(/ *$/,"",b)
    sub(/^ */,"",$14)
    b=b "," $14
  }
}
END {
  print a,b,"|"
}

CaKiwi
 
Upvote 0 Downvote
Hi CaKiwi,

This is very helpful, however, produces the following output;

| FALSE | client.local/client/Accounts/user1 | Town | | user1 | company name | | 02/12/2009 17:02 | Department1 | | User 1 | CN=user1,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G1 |
| FALSE | client.local/client/Accounts/user1 | Town | | user1 | company name | | 02/12/2009 17:02 | Department1 | | User 1 |CN=user1,OU=Accounts,OU=Client
,DC=client,DC=local | Grp_G2 |
| FALSE | client.local/client/Accounts/user1 | Town | | user1 | company name | | 02/12/2009 17:02 | Department1 | | User 1 | CN=user1,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G3,Grp_G4 |
| FALSE | client.local/client/Accounts/user2 | Town | | user2 | company name | | 02/12/2009 17:02 | Department1 | | User 2 | CN=user2,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G7 |
| FALSE | client.local/client/Accounts/user3 | Town | | user3 | company name | | 02/12/2009 17:02 | Department1 | | User 3 |CN=user3,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G7 |
| FALSE | client.local/client/Accounts/user3 | Town | | user2 | company name | | 02/12/2009 17:02 | Department1 | | User 2 | CN=user2,OU=Accounts,OU=Client,DC=client,DC=local | Grp_G8 |

Meaning it doesn't merge all columns... e.g. user 2 and user 1. Also, you're right, the data won't appear in consecutive lines; more likely it won't :)

How would you fix these?

You're correct in your assumption about user 3 :)
 
Upvote 0 Downvote
Try this
merge.awk
Code: 
[COLOR=#6a5acd]BEGIN[/color] {
  [COLOR=#6a5acd]FS[/color]=[COLOR=#ff00ff]"|"[/color][COLOR=#6a5acd];[/color]
}

[COLOR=#6a5acd]NR[/color] > [COLOR=#ff00ff]1[/color] {
  [COLOR=#0000ff]# strip out all spaces[/color]
  [COLOR=#008080]gsub[/color]([COLOR=#ff00ff]/[[/color][COLOR=#6a5acd]" "[/color][COLOR=#ff00ff]][/color][COLOR=#6a5acd]+[/color][COLOR=#ff00ff]/[/color][COLOR=#6a5acd],[/color] [COLOR=#ff00ff]""[/color][COLOR=#6a5acd],[/color] [COLOR=#6a5acd]$0[/color])
  user = [COLOR=#6a5acd]$6[/color][COLOR=#6a5acd];[/color]
  [COLOR=#804040][b]if[/b][/color] (user_array[[COLOR=#6a5acd]user[/color]]) {
    [COLOR=#804040][b]if[/b][/color] ([COLOR=#008080]index[/color](user_array[[COLOR=#6a5acd]user[/color]][COLOR=#6a5acd],[/color] [COLOR=#6a5acd]$14[/color]) == [COLOR=#ff00ff]0[/color]) {
      user_array[[COLOR=#6a5acd]user[/color]] = user_array[[COLOR=#6a5acd]user[/color]] [COLOR=#ff00ff]","[/color] [COLOR=#6a5acd]$14[/color]
    }
  }
  [COLOR=#804040][b]else[/b][/color] {
    user_array[[COLOR=#6a5acd]user[/color]] = [COLOR=#6a5acd]$1[/color] [COLOR=#ff00ff]"|"[/color] [COLOR=#6a5acd]$2[/color] [COLOR=#ff00ff]"|"[/color] [COLOR=#6a5acd]$3[/color] [COLOR=#ff00ff]"|"[/color] [COLOR=#6a5acd]$4[/color] [COLOR=#ff00ff]"|"[/color] [COLOR=#6a5acd]$5[/color] [COLOR=#ff00ff]"|"[/color] [COLOR=#6a5acd]$6[/color] [COLOR=#ff00ff]"|"[/color] [COLOR=#6a5acd]$7[/color] [COLOR=#ff00ff]"|"[/color] [COLOR=#6a5acd]\[/color]
                       [COLOR=#6a5acd]$8[/color] [COLOR=#ff00ff]"|"[/color] [COLOR=#6a5acd]$9[/color] [COLOR=#ff00ff]"|"[/color] [COLOR=#6a5acd]$10[/color] [COLOR=#ff00ff]"|"[/color] [COLOR=#6a5acd]$11[/color] [COLOR=#ff00ff]"|"[/color] [COLOR=#6a5acd]$12[/color] [COLOR=#ff00ff]"|"[/color] [COLOR=#6a5acd]$13[/color] [COLOR=#ff00ff]"|"[/color] [COLOR=#6a5acd]$14[/color]
  }
}

[COLOR=#6a5acd]END[/color] {
  [COLOR=#804040][b]for[/b][/color] (user in user_array) {
    [COLOR=#804040][b]printf[/b][/color]([COLOR=#ff00ff]"[/color][COLOR=#6a5acd]%s[/color][COLOR=#ff00ff]|[/color][COLOR=#6a5acd]\n[/color][COLOR=#ff00ff]"[/color][COLOR=#6a5acd],[/color] user_array[[COLOR=#6a5acd]user[/color]])
  }
}
Running it with
Code: 
awk -f merge.awk select.dat > select_merge.dat
where select.dat is your file produces select_merge.dat, i.e. what you want
Code: 
|FALSE|client.local/client/Accounts/user1|Town||user1|companyname||02/12/200917:02|Department1||User1|CN=user1,OU=Accounts,OU=Client,DC=client,DC=local|Grp_G1,Grp_G2,Grp_G3,Grp_G4|
|FALSE|client.local/client/Accounts/user2|Town||user2|companyname||02/12/200917:02|Department1||User2|CN=user2,OU=Accounts,OU=Client,DC=client,DC=local|Grp_G7,Grp_G8|
|FALSE|client.local/client/Accounts/user3|Town||user3|companyname||02/12/200917:02|Department1||User3|CN=user3,OU=Accounts,OU=Client,DC=client,DC=local|Grp_G7|
 
Upvote 0 Downvote
Thanks guys - that's fantastic!
 
Upvote 0 Downvote
Status
Not open for further replies.

Similar threads

MotorCycleDude
  • Locked
  • Question
awk Question
Replies
12
Views
240
mikrom
mikrom
FrancisMMM
  • Locked
  • Question
awk parsing is getting slower and stalled :(
Replies
1
Views
115
FrancisMMM
FrancisMMM
johnegracejr
  • Locked
  • Question
AWK suffix overwriting original text.
Replies
1
Views
141
feherke
feherke
V_M
  • Locked
  • Question
Awk Command not able to read French/Spanish characters from a file
Replies
1
Views
128
mikrom
mikrom
dlicheri
  • Locked
  • Question
awk - transposing output of program 1
Replies
4
Views
159
mikrom
mikrom

Part and Inventory Search

Sponsor

Back
Top