Beefy Boxes and Bandwidth Generously Provided by pair Networks
Your skill will accomplish
what the force of many cannot
 
PerlMonks  

Re^3: regex return true instead of false (precedence)

by Marshall (Canon)
on Aug 26, 2019 at 18:00 UTC ( [id://11105079]=note: print w/replies, xml ) Need Help??


in reply to Re^2: regex return true instead of false (precedence)
in thread regex return true instead of false

I prefer to use indenting to make all the parens more clear. Also, this "unless not" syntax can be confusing.
I think I got this right, but it is easy to make a mistake. Consider:
# X is the "default action" # do X with an exception that when A,B and C are all false # then don't do X unless (!A and !B and !C) {X} # if anybody is true then they all can't be false # so we do X unless anybody is true (meaning they # can't all be false) unless (A or B or C) {X} ------- unless ( (!($args =~ /\-rs(\s+)(\S+)/)) && (!($args =~ /\-p(\s+)(\S+)/)) && (!($args =~ /\-P(\s+)(\S+)/)) ) {..... # I figure this is easier to understand: unless ( $args =~ /\-rs(\s+)(\S+)/ or $args =~ /\-p(\s+)(\S+)/ or $args =~ /\-P(\s+)(\S+)/ ) {....

Replies are listed 'Best First'.
Re^4: regex return true instead of false (precedence)
by GrandFather (Saint) on Aug 26, 2019 at 21:10 UTC

    If you break a condition like that over multiple lines (and for a complicated or long condition I recommend it) put the operators at the front of the lines so they don't get "lost" at the back:

    if ( $args =~ /\-rs(\s+)(\S+)/ || $args =~ /\-p(\s+)(\S+)/ || $args =~ /\-P(\s+)(\S+)/ )
    Optimising for fewest key strokes only makes sense transmitting to Pluto or beyond
      Actually, I like that better.
      As I remember, a senior Monk yelled at me once when I posted something like that. But yes this is the way that I would do it in my code. I didn't want to get into an argument that detracts from the main point.
      Update: As I remember the argument against this was like "injecting spacing like that is bad because it is manually done and IDE's don't do it that way". BS. The compiler doesn't give a darn about spaces. Whitespace is added to help the humans. If it helps the humans, it is a good idea.

        I agree with your and GrandFather's leftist leaning when it comes to formatting complex logical expressions, but I think it's tangential to the main point made by haukex here: the statements
            if     (    condition  ) {} else {code }
            unless (    condition  )         {code }
            if     ( ! (condition) )         {code }
        are all exactly equivalent, so the condition expression within them does not need to change at all.


        Give a man a fish:  <%-{-{-{-<

      Also you can waste one line of code for making first conditional to have the same line-format as others:
      if( not 0 || $args =~ /\-rs(\s+)(\S+)/x # || $args =~ /\-? (\s+)(\S+)/x || $args =~ /\-p (\s+)(\S+)/x || $args =~ /\-P (\s+)(\S+)/x ){ ... }
      ...or equivalently inside regex:
      if( $args !~ / (*FAIL) | \-rs(\s+)(\S+) # | \-? (\s+)(\S+) | \-P (\s+)(\S+) | \-p (\s+)(\S+) /x ){ ... }
      edit: not equivalently, because there are 8 capturing groups instead of 2 :)

      Of course regex can be simplyfied as AnomalousMonk have shown here Re: regex return true instead of false.
        if( $args !~ / (*FAIL) | \-rs(\s+)(\S+) # | \-? (\s+)(\S+) | \-P (\s+)(\S+) | \-p (\s+)(\S+) /x ){ ... }

        I don't understand the purpose of the  (*FAIL) operator (see Special Backtracking Control Verbs in perlre in Perl versions 5.10+ (update: see also Backtracking control verbs in perlretut - but that's not actually as extensive as the discussion in perlre)) in the quoted regex. At any position in the  | alternation,  (*FAIL) will simply force the RE to try the next alternation; if it's at the last position, all the preceding pattern matches would have failed and the alternation would fail anyway without (*FAIL).

        c:\@Work\Perl\monks>perl -wMstrict -le "use 5.010; ;; for my $s (qw(XXX ppp)) { print qq{for '$s' string}; if ($s !~ m{ (*F) | p | q }xms) { print ' with (*F): no match, !~ +true' } if ($s !~ m{ Z | p | q }xms) { print ' with Z: no match, !~ +true' } if ($s !~ m{ p | q }xms) { print ' no (*F): no match, !~ +true' } print ' -------'; } " for 'XXX' string with (*F): no match, !~ true with Z: no match, !~ true no (*F): no match, !~ true ------- for 'ppp' string -------
        Note that for all these variations, for string 'XXX' there is never a match (!~ is always true); string 'ppp' always matches (!~ always false).

        ... not equivalently, because there are 8 capturing groups instead of 2 ...

        If you're using  (*FAIL) you must be using Perl version 5.10+, so you also have the  (?|pattern) "branch reset" operator (see Extended Patterns in perlre). This allows you to go back to having two capture groups again!

        Update: Changed example code slightly to better reflect discussion.


        Give a man a fish:  <%-{-{-{-<

Log In?
Username:
Password:

What's my password?
Create A New User
Domain Nodelet?
Node Status?
node history
Node Type: note [id://11105079]
help
Chatterbox?
and the web crawler heard nothing...

How do I use this?Last hourOther CB clients
Other Users?
Others goofing around in the Monastery: (4)
As of 2024-04-19 16:16 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?

    No recent polls found