awk: Pattern substitution in one file based on input from another file

Question

I have an XML file that has a set of entries like this:

<attr name="trajectory" value="nodo2"/>

What I want to do is to substitute the value field in each entry of the file by an entry of the form "nodoX", where X should be read from a second file that contains a list of numbers, e.g "file2" contains:

In that case my processed XML file should have substituted the previous entries by:

<attr name="trajectory" value="nodo4"/>
...
<attr name="trajectory" value="nodo8"/>
...
<attr name="trajectory" value="nodo67"/>
...
<attr name="trajectory" value="nodo52"/>
...

Currently I am trying with the following awk command but it is not working:

awk '/"trajectory"/ {if (getline < "file2") {x=$1; sub(/"nodo2"/,"\"nodo"x"\"")}}1' $XML_INPUT_FILE > $XML_OUTPUT_FILE

Could someone advise the correct way to get this done in awk ?

Thanks

Daniel

glenn jackman · Accepted Answer · 2012-03-30 15:39:40Z

2

getline < "file2" overwrites $0, so the sub command will not do what you want. Try:

awk '/"trajectory"/ {if (getline x < "file2") {sub(/"nodo2"/,"\"nodo"x"\"")}}1'

See http://www.gnu.org/software/gawk/manual/gawk.html#Getline

answered Mar 30, 2012 at 15:39

glenn jackman

249k42 gold badges233 silver badges362 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Zsolt Botykai · Accepted Answer · 2012-03-30 15:40:58Z

1

Well, you can do that with awk...

awk 'BEGIN { c=1 ; while ((getline line < "OTHER_NUMBERS") > 0) { a[c]=line ; c++ } }
     NR == 1 { c=1 }
     /<attr name="trajectory" value="nodo[0-9]+".>/ {
          print gensub("^(.*=.nodo)[0-9]+(.*)","\\1" a[c] "\\2","g",$0)
          c++
     }
     ! /<attr name="trajectory" value="nodo[0-9]+".>/ { print }' XMLFILE

Generally it reads your number files into an indexed array, then on every (matching) trajectory lines prints the replacement. Note that it can fail you, if the number of lines in the numbers file is less than the number of nodo lines...

answered Mar 30, 2012 at 15:40

Zsolt Botykai

52k14 gold badges90 silver badges111 bronze badges

Comments

potong · Accepted Answer · 2012-03-30 17:40:42Z

0

This might work for you:

 awk 'FNR==NR{n[++i]=$1;next};/"trajectory"/{sub(/[0-9]+/,n[++j],$3)}1' file2 XML

answered Mar 30, 2012 at 17:40

potong

59.3k6 gold badges55 silver badges92 bronze badges

Comments

jaypal singh · Accepted Answer · 2014-03-13 23:39:46Z

0

solution below

awk '
BEGIN{i = 0; while(getline num < "file2"){a[i] = num;i++}}
/"trajectory"/{sub(/nodo2/,"node"a[NR-1], $0);print $0}' file1

edited Mar 13, 2014 at 23:39

jaypal singh

77.6k24 gold badges108 silver badges147 bronze badges

answered Mar 30, 2012 at 17:42

Vishal

2831 gold badge5 silver badges15 bronze badges

Collectives™ on Stack Overflow

awk: Pattern substitution in one file based on input from another file

4 Answers 4

Comments

Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

Comments

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related