How would you write this RegExp?

harangzsolt33 has asked for the wisdom of the Perl Monks concerning the following question:

I am so frustrated, I have been reading the RegExp help for awhile, and I just don't get it. All I want to do is find out if a string is made up of a set of characters, and print either 1 or 0 (true or false).

Here is the character set:


0-9 - all numbers
a-zA-Z - all letters
. - the decimal point
\ - backslash
/ - forward slash
- - the dash
_ - the underline
% - percent sign
$ - dollar sign
' - single quote
() - parenthesis
{} - opening and closing brackets
& - the and sign
! - the exclamation point
~ - wave
` - tick or whatever this is
@ - the at sign
# - the comment character
^ - this guy
[download]

Any other character is NOT allowed. SO, if the string contains a character that is not allowed, then I want the match to evaluate to 1, otherwise 0. This way I can tell whether the string has any illegal characters in it.

Here is the code I have written, which doesn't do anything. :-P :-(


my $STR1 = '///?///';
my $STR2 = 'ABC_ABC';

print "\n" . ($STR1 =~ /(\d\w.\/~-_!@#$%\^&{}\(\).'`)/);
print "\n" . ($STR2 =~ /(\d\w.\/~-_!@#$%\^&{}\(\).'`)/);
[download]

Comment on How would you write this RegExp? Select or Download Code

Replies are listed 'Best First'.
Re: How would you write this RegExp? by BrowserUk (Patriarch) on Jul 26, 2016 at 21:26 UTC
Test this: my $re = qr[^[0-9a-zA-Z.\\/_%$'(){}&!~`@#^-]+$];; print $string !~ $re ? 1 : 0;; [download] With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday' Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error. "Science is about questioning the status quo. Questioning authority". I knew I was on the right track :) In the absence of evidence, opinion is indistinguishable from prejudice.	[reply] [d/l]
Re^2: How would you write this RegExp? by harangzsolt33 (Deacon) on Jul 26, 2016 at 21:49 UTC
Wow. This is awesome! Thank you very much!!!!! All I had to do was put a backslash before the $ sign, and it worked beautifully! $re = qr[^0-9a-zA-Z.\\/_%\$'(){}&!~`@#^-+$];	[reply]
Re: How would you write this RegExp? by AnomalousMonk (Archbishop) on Jul 27, 2016 at 00:28 UTC
I might approach the problem slightly differently, with an inverted character class: c:\@Work\Perl\monks>perl -wMstrict -le "use constant NOT_ALLOWED => qr{ [^-\x27\x23\$\@0-9a-zA-Z.\\/_%(){}&!~ +`^] }xms; ;; for my $str ('', qw(ABC_ABC 'A' ///?/// =+\| ABC=), ' ', qq{\t +\r\n}) { my $status = 0 + $str =~ NOT_ALLOWED; print qq{[$str] $status}; } " [] 0 [ABC_ABC] 0 ['A'] 0 [///?///] 1 [=+\|] 1 [ABC=] 1 [ ] 1 [ ] 1 [download] Note that: A `^` (caret) in the first position in a character class inverts the contents of the class. A `-` (hyphen) in the first position in a character class (or in the first position after an initial caret, if there is one), or in the last position in the class is a literal hyphen character. Everywhere else it is a character range operator as in `A-Z` and `0-9` elsewhere in the example class. The `$ @` characters should always be escaped, as otherwise they will often be interpreted as interpolable variables. Finally, `\x27 \x23` are how I have to represent the `' #` characters, respectively, because my REPL doesn't like these literal characters in `qr//` and similar operators. (Have to fix that someday.) (And a lot more test cases wouldn't be a bad idea.) Give a man a fish: `<%-{-{-{-<`	[reply] [d/l] [select]
Re: How would you write this RegExp? by Anonymous Monk on Jul 26, 2016 at 21:32 UTC
This guy is called a caret. :-) You will need the following: `0+` arithmetic to make the result a 0/1 boolean Anchors `^$` at the regex pattern, to force a full match A character class `[]` in the regex for your accepted set A repetition `*` (or `+`) after the character class Did you read the documentation?	[reply] [d/l] [select]
Re^2: How would you write this RegExp? by harangzsolt33 (Deacon) on Jul 26, 2016 at 22:57 UTC
Well, I have read the documentation, but it's so long, by the time I get to the end, I forget everything else. lol I downloaded an Android app called RegEx GUI, which asks a bunch of questions and is supposed to help you build a regex, but it didn't help me.. :-( I think, the most complicated part of learning the Perl language is understanding how to write and interpret regular expressions. I still don't understand how to do it. It's a big mess. and everytime somebody writes a long !$,\\.m./(/&/\/%(\/_\/!/@Y\/%/\@\\\\$H&(P!//(/@&/P/@/#!/$/^/-//-^xqfa34s*(67)_w/0//P@\/UI\.\.\ I am like oh, wow!!! what is that!? :-O Haha :D	[reply]
Re^3: How would you write this RegExp? by MidLifeXis (Monsignor) on Jul 27, 2016 at 11:21 UTC
And don't forget Damian Conway's Regexp::Debugger. It does a very good job of showing, in full, glorious, mind-numbing detail, every step of the regexp processing sequence. --MidLifeXis	[reply]
Re^3: How would you write this RegExp? by Anonymous Monk on Jul 27, 2016 at 00:12 UTC
Luckily :) other regex tools, ppixregexplain.pl Also perlrequick and Regexp::English	[reply]
Re: How would you write this RegExp? by ww (Archbishop) on Jul 27, 2016 at 12:42 UTC
This may give you a (slightly golfed) alternate start (on the letters; you have a fine solution for including non-a-zA-Z characters): `C:>perl -E "print 'Enter a char: '; my $char = <>; print $char; print +$char =~ /[AEIOU]/i ? 'char is a vowel' : 'char is not a vowel';" Enter a char: v v char is not a vowel C:>perl -E "print 'Enter a char: '; my $char = <>; print $char; print +$char =~ /[AEIOU]/i ? 'char is a vowel' : 'char is not a vowel';" Enter a char: E E char is a vowel C:>` [download] Spirit of the Monastery	[reply] [d/l] [select]