Repetitive pattern with different keyword

[AA2K5]

Hi,

I have a repetitive regular expression that I am matching and I am performing the same action over and over.

I don't really want to write out the code every time for each word I want to match. I was wondering if anyone had a tip to reduce code duplication.

Here's a sample of the awk code I am using: -

/[[:graph:]]+[[:space:]]*\<[red]keyword[/red]\>/ {
    gsub(/\<[red]keyword[/red]\>/, "\n" [red]"KEYWORD"[/red])
}

I'm matching the keyword and putting a newline before it and forcing the keyword to uppercase. Basically I have to do this for a long list of keywords.

Any help would be appreciated

Thanks

 

[PHV]

Put all your keywords in an array and then play with the match,gsub and toupper functions inside a for(... in ...) loop.

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ222-2244

 

[AA2K5]

I thought this would be the logical approach but I just wasn't sure whether awk allowed all the match, gsub etc. functions when using array elements.

Thanks for your help.

 

[AA2K5]

Just another quick question.

If I'm in a for loop traversing all the elements of an array, how can I reference an array element in a regular expression?

For example I've tried the following code but I'm pretty sure its wrong because I don't think I'm referencing the array properly in the regular expressions: -

BEGIN {
    IGNORECASE=1
}

array[1] = "keyword1"
array[2] = "keyword2"

{
  for (x in array)
  {
    if($0 ~ /[[:graph:]]+[[:space:]]*\<[red]array[x][/red]\>/) 
    { 
      gsub(/\<[red]array[x][/red]\>/, "\n" toupper([red]array[x][/red]), $0) 
    }
  }

  printf "%s\n",$0

}

END   {}

When I've used this code, the script adds the newlines & capitalisation but also outputs each line several times.

 

[PHV]

And this ?
BEGIN {
IGNORECASE=1
array[1] = "keyword1"
array[2] = "keyword2"
}
{ for (x in array)
if($0 ~ "[[:graph:]]+[[:space:]]*\<"array[x]"\>")
gsub("\<"array[x]"\>", "\n" toupper(array[x]), $0)
print
}

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ222-2244

 

[AA2K5]

I tried the code in the previous post but unfortunately it did not match the keywords in the array. I had a line like this: -

Some text and here is the keyword. Some more text

I expected this to come out like this when "keyword" was in the array: -

Some text and here is the
keyword. Some more text

However there was no newline and the text stayed the same. Also I got the following warnings when running the script:

awk: C:\awk\test.awk:10: warning: escape sequence `\<' treated as plain `<'
awk: C:\awk\test.awk:10: warning: escape sequence `\>' treated as plain `>'

 

[PHV]

And this ?
BEGIN {
IGNORECASE=1
array[1] = "keyword1"
array[2] = "keyword2"
}
{ for (x in array)
if($0 ~ "[[:graph:]]+[[:space:]]*"array[x])
gsub(array[x], "\n" toupper(array[x]), $0)
print
}

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ222-2244

 

[AA2K5]

Yep, thanks for that, it works.

Unfortunately without the <>, it is only a partial match on the keyword so "keywords" or "keyword1" or "keyword100" would all match.

But I need an exact match on the keyword so I need to look for a way round it. (assuming there is a way round it)