shellscript: add (upcounting) number value between matching strings

Question

I have more than 100 files with with over 9000 lines of text. A preview extract of one of the files looks like this:

<productname>kidscar1</productname>
<productid>98</productid>
<productname>kidscar2</productname>
<productcolor>yellow</productcolor>
<productid>101</productid>
<productname>kidscar3</productname>
<productsize>xxl</productsize>
<productcolor>green</productcolor>
<productid>104</productid>
<productname>kidscar4</productname>
<productcolor>bleu</productcolor>
<productsize>xl</productsize>
<producttype>electric</producttype>
<productid>103</productid>

I'm trying to change (re-arange) the product ID's starting from a different product ID and automatically count this up for the next lines containing an product ID.

I was thinking of a shell script solution which I could use in a for loop.

**sh idscript.sh oldfile.txt 1000 productid > newfile.txt**

Result:

<productname>kidscar1</productname>
<productid>1000</productid>
<productname>kidscar2</productname>
<productcolor>yellow</productcolor>
<productid>1001</productid>
<productname>kidscar3</productname>
<productsize>xxl</productsize>
<productcolor>green</productcolor>
<productid>1002</productid>
<productname>kidscar4</productname>
<productcolor>bleu</productcolor>
<productsize>xl</productsize>
<producttype>electric</producttype>
<productid>1003</productid>`

I know that it's possible to replace a whole line in sed with the next command:

**sed "s/<productid>100</productid>=.*/<productid>=<productid>1000</productid>/g"**

But how can I make this work to get the above result? If there are other (simpler) ways to achieve this, i like to hear it also!

Ed Morton · Accepted Answer · 2017-12-12 14:06:38Z

2

sed is for s/old/new that is all. In this case you need a numeric variable to keep track of the incrementing product and sed doesn't support variables at all so just use awk:

$ awk -v pid=1000 '/<productid>/{sub(/[0-9]+/,pid++)}1' file
<productname>kidscar1</productname>
<productid>1000</productid>
<productname>kidscar2</productname>
<productcolor>yellow</productcolor>
<productid>1001</productid>
<productname>kidscar3</productname>
<productsize>xxl</productsize>
<productcolor>green</productcolor>
<productid>1002</productid>
<productname>kidscar4</productname>
<productcolor>bleu</productcolor>
<productsize>xl</productsize>
<producttype>electric</producttype>
<productid>1003</productid>

edited Dec 12, 2017 at 14:06

answered Dec 12, 2017 at 14:03

Ed Morton

209k18 gold badges90 silver badges212 bronze badges

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

3 Comments

Srini V Over a year ago

Nice. productid string is used as RS and removing all numbers and writing the incremented pid startig from 1000?

Ed Morton Over a year ago

No, pid is a variable containing a number and I'm just replacing the existing number with its value and post-incrementing it. I'm not doing anything with RS.

Zacks Delga Over a year ago

Thanx for the infor Ed, exactly what I was looking for!

RavinderSingh13 · Accepted Answer · 2017-12-12 14:27:22Z

0

Following awk may also help you on same too.

awk -v val=1000 '/productid/{sub(/>[0-9]+</,">"val++"<")} 1'  Input_file

edited Dec 12, 2017 at 14:27

answered Dec 12, 2017 at 14:17

RavinderSingh13

135k14 gold badges61 silver badges100 bronze badges

1 Comment

RavinderSingh13 Over a year ago

@EdMorton, done the change sir, yes I thought if in case null value there for <productid></productid>, though OP didn't mention it but thought to cover it in it, thanks for letting me know.

Collectives™ on Stack Overflow

shellscript: add (upcounting) number value between matching strings

2 Answers 2

3 Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related