Re^3: splitting lspci output

The x option indicates that most white space should be ignored. That allows me to use spaces to break up the regex a little to make it easier to see where each field is.

The first field is captured by (\S+)\s+ which grabs the first sequence of non-space characters from the start of the line.

The second field is captured by ([^:]+):\s+ which matches and captures a sequence of characters excluding ':', then matches the ':' and any following white space.

The third field is a little more interesting. ((?:(?!\s*\(rev).)+))\s* crawls along the remainder of the line matching and capturing one character at a time until it comes to '(rev' preceeded by any amount of white space. The (?:...) bit groups some stuff without capturing. The (?!...) bit looks ahead to make sure the following stuff doesn't match whatever '...' is. The '...' bit inside of (?:...) matches a single character so long as it's not the start of something that would match '\s*\(rev'.

The fourth field just picks up everything from '(Rev' to the end of the line.

DWIM is Perl's answer to Gödel

Comment on Re^3: splitting lspci output Select or Download Code

Replies are listed 'Best First'.
Re^4: splitting lspci output by HuckinFappy (Pilgrim) on May 18, 2006 at 05:54 UTC
In other words: m/ ^(\S+) # Grab all non white space at the start into $1 \s+ # Skip any white space ([^:]+) # Now grab everything that isn't a colon in $2 : # End the grab for $2 at the colon \s+ # Skip any white space ( # Going to capture into $3 (?: # Start non-capturing parens (?! # Start negative lookahead \s* # Look for some white space \(rev # Followed by (rev ) # Close negative lookahead .) # Close non-capturing parens +) # So if whitespace followed by (rev, grab it $3 \s* # Up until some white space (.*) # Grab everything else after the white space into +$4 /x; [download] I really just wanted to demonstrate to the OP how to powerfully use the /x modifier	[reply] [d/l]

Replies are listed 'Best First'.

Re^4: splitting lspci output
by HuckinFappy (Pilgrim) on May 18, 2006 at 05:54 UTC

    m/ ^(\S+)    # Grab all non white space at the start into $1
         \s+         # Skip any white space
         ([^:]+)   # Now grab everything that isn't a colon in $2
         :            # End the grab for $2 at the colon 
         \s+        # Skip any white space
        (            # Going to capture into $3
        (?:         # Start non-capturing parens
        (?!         # Start negative lookahead
        \s*         # Look for some white space
        \(rev      # Followed by (rev
        )            # Close negative lookahead
       .)            # Close non-capturing parens
       +)           # So if whitespace followed by (rev, grab it $3
       \s*          # Up until some white space
      (.*)          # Grab everything else after the white space into 
+$4
            /x;
[download]

I really just wanted to demonstrate to the OP how to powerfully use the /x modifier

[reply]
[d/l]