gawk asort to sort record groups based on one subfield


 
Thread Tools Search this Thread
Top Forums UNIX for Dummies Questions & Answers gawk asort to sort record groups based on one subfield
# 1  
Old 10-07-2012
gawk asort to sort record groups based on one subfield

input ("/" delimited fields):
Code:
style1/book1 (author_C)/editor1/2000
style1/book2 (author_A)/editor2/2004
style1/book3 (author_B)/editor3/2001
style2/book8 (author_B)/editor4/2010
style2/book5 (author_A)/editor2/1998

Records with same field 1 belong to the same group.
Using asort (not sort), in each group I need to sort the records in ascending order based on the string between braces in field 2, to obtain:
Code:
style1/book2 (author_A)/editor2/2004
style1/book3 (author_B)/editor3/2001
style1/book1 (author_C)/editor1/2000
style2/book5 (author_A)/editor2/1998
style2/book8 (author_B)/editor4/2010

I tried to sort the records by field1 and then by subfield2 in field2, but it didn't work:
Code:
BEGIN{FS=OFS="/"}

{
    array[$1] = $0

    split ($2, aut, " ")

    asort(array)

    o = asort(aut)

    for (o in aut)
        print array[aut[o]]

}

# 2  
Old 10-07-2012
The versions of awk that I use (on OS X) don't have the asort() and asorti() functions, but I have read the gawk man page. Unlike the sort utility, there is no way to specify a sort key for these functions; they always sort the array using the entire contents of the string as the sort key. If you want to use asort() in gawk to sort with field 1 as your primary sort key and the second part of field 2 as your secondary key; you need to prepend each line in your array with primary and secondary sort fields, use asort() or asorti() to sort the modified records, and then strip off the added sort fields when you print (or otherwise process) the results.
# 3  
Old 10-08-2012
Quote:
ou need to prepend each line in your array with primary and secondary sort fields, use asort() or asorti() to sort the modified records, and then strip off the added sort fields when you print (or otherwise process) the results.
So you mean I should
1st) sort by field1 and generate a first output
2nd) use this output to sort by subfield 2 and generate the final output.

I tried things like below but still doesn't work.
Code:
BEGIN{FS=OFS="/"}

{
# sort by field1
    array[$1] = $0

    asort(array)

# first output
    for (i in array)
        $0 = array[i]

# redefine fields in first output    
        split($0, rec, FS)
        rec[$2] = $0

        split($0, sub, " ")
        aut[++a] = sub[2]

# sort by subfield2 
        n = asort(aut)

# print final output
        for (j=1; j<=n; j++)
            print array[aut[j]]
    
}

# 4  
Old 10-08-2012
Quote:
Originally Posted by lucasvs
So you mean I should
1st) sort by field1 and generate a first output
2nd) use this output to sort by subfield 2 and generate the final output.

I tried things like below but still doesn't work.
Code:
BEGIN{FS=OFS="/"}

{
# sort by field1
    array[$1] = $0

    asort(array)

# first output
    for (i in array)
        $0 = array[i]

# redefine fields in first output    
        split($0, rec, FS)
        rec[$2] = $0

        split($0, sub, " ")
        aut[++a] = sub[2]

# sort by subfield2 
        n = asort(aut)

# print final output
        for (j=1; j<=n; j++)
            print array[aut[j]]
    
}

What you have above sorts the accumulated input twice each time you read a line from your input file. Clearly that isn't what you want. Even if you were doing the sorts in an END clause instead of in a clause that is executed every time you read an input line, performing a sort on the entire line and then performing a sort on the second sort key is not the same as performing a single sort with a primary and secondary sort key.

I don't have access to a system running gawk, but just using standard interfaces, I get the output:
Quote:
style1/book2 (author_A)/editor2/2004
style1/book3 (author_B)/editor3/2001
style1/book1 (author_C)/editor1/2000
style2/book5 (author_A)/editor2/1998
style2/book8 (author_B)/editor4/2010
which I think is what you're trying to get, when I use the script:
Code:
#!/bin/ksh
awk 'BEGIN{FS=OFS="/"
    tmpfile = "asorti.out"
    sortcommand = "sort -t/ -o " tmpfile
    cleanup = "rm " tmpfile
}
{   split ($2, sub, " ")
    array[$1 "/" sub[2] "/" $0] = $0
}
END{for (i in array) print i | sortcommand
    close(sortcommand)
    while(getline i < tmpfile) print array[i]
    close(tmpfile)
    system(cleanup)
}' in

where in contains the data listed in your first posting on this thread. If I read the gawk man page correctly, this should be roughly equivalent to:
Code:
#!/bin/ksh
gawk 'BEGIN{FS = OFS = "/" }
{   split ($2, sub, " ")
    array[$1 "/" sub[2] "/" $0] = $0
}   
END{n = asorti(array)
    for(i = 1; i <= n; print array[i++]);
}' in

 
Login or Register to Ask a Question

Previous Thread | Next Thread

10 More Discussions You Might Find Interesting

1. Shell Programming and Scripting

Best way to sort file with groups of text of 4-5 lines by the first one

Hi, I have some data I have taken from the internet in the following scheme: name direction webpage phone number open hours menu url book url name ... Of course the only line that is mandatory is the name wich is the one I want to sort by. I have the following sed & awk script that... (3 Replies)
Discussion started by: devmsv
3 Replies

2. UNIX for Beginners Questions & Answers

Gawk: PROCINFO["sorted_in"] does not sort my numerical array values

Hi, PROCINFO seems to be a great function but I don't manage to make it works. input: B,A,C B B,B As an example, just want to count the occurence of each letter across the input and sort them by decreased order. Wanted output: B 4 A 1 C 1 When I use this command, the PROCINFO... (4 Replies)
Discussion started by: beca123456
4 Replies

3. Shell Programming and Scripting

Sort html based on .jar, .war file names and still keep text within three groups.

Output from zipdiff GNU EAR comparison tool produces output in html divided into three sections "Added, Removed, Changed". I want the output to be sorted by jar or war file. <html> <body> <table> <tr> <td class="diffs" colspan="2">Added </td> </tr> <tr><td> <ul>... (5 Replies)
Discussion started by: kchinnam
5 Replies

4. Shell Programming and Scripting

Match groups of capital words using gawk

Hi I'd like to extract from a text file, using gawk, the groups of words beginning with a capital letter, that are not at the begining of a sentence (i.e. Not after a full stop and a pace ". "), including special characters like registered or trademark (® or ™ ). For example I would like to... (1 Reply)
Discussion started by: louisJ
1 Replies

5. Shell Programming and Scripting

Don't understand AWK asort behaviour

Hello, I have the following script : BEGIN { print "1 ***"; split("abc",T,""); T="e"; T="z"; T="y"; for (i in T) printf("%i:%s ",i,T); print ""; for (i=1; i<=length(T); i++) printf(T); print "" print "2 ***"; asort(T,U); for (i in U) printf("%i:%s ",i,U); ... (3 Replies)
Discussion started by: jgilot
3 Replies

6. Shell Programming and Scripting

AWK exclude first and last record, sort and print

Hi everyone, I've really searched for a solution to this and this is what I found so far: I need to sort a command output (here represented as a "cat file" command) and from the second down to the second-last line based on the second row and then print ALL the output with the specified section... (7 Replies)
Discussion started by: dentex
7 Replies

7. Shell Programming and Scripting

sort file specifying record length

I've been searching high and low for this...but, maybe I'm just missing something. I have a file to be sorted that, unfortunately, contains binary data at the end of the line. As you may guess, this binary data may contain a newline character, which messes up the sort. I think I could resolve this... (5 Replies)
Discussion started by: jcagle
5 Replies

8. Programming

Help with sort and keep data record to calculate N50 in c

Input_file_1 #content_1 A #content_2 AF #content_3 AAR #content_4 ASEI #content_5 AS #content_6 ADFSFGS Rules: 1. Based on c program to calculate content of each "#". Result getting from the above Input_file_1 are 1,2,3,4,2,7; 2. Sort length on reverse order (descending order).... (5 Replies)
Discussion started by: cpp_beginner
5 Replies

9. Shell Programming and Scripting

Gawk / Awk Merge Lines based on Key

Hi Guys, After windows died on my netbook I installed Lubuntu and discovered Gawk about a month ago. After using Excel for 10+ years I'm amazed how quick and easily Gawk can process data but I'm stuck with a little problem merging data from multiple lines. I'm an SEO Consultant and provide... (9 Replies)
Discussion started by: Jamesfirst
9 Replies

10. Shell Programming and Scripting

Removing \n within a record (awk/gawk)

I am using a solution that was provided by a member: awk '{s=$0;if(length(s) < 700){getline; s=s " " $0}printf("%s\n",s)}' This scans through a file and removes '\n' within a record but not the record delimiter. However, there are instances where there are MULTIPLE instances of '\n'... (10 Replies)
Discussion started by: CKT_newbie88
10 Replies
Login or Register to Ask a Question