Regexp Matching Not Blank And Not A Specific Word?

Ruby
1

Hi,

Please could somebody show me how to write a regular expression that
matches when the input is not blank and it’s not a specific word,
e.g. dog.

For example, I would hope for these results:

‘’ =~ regexp # nil
‘dog’ =~ regexp # nil
‘cat’ =~ regexp # not nil

I have tried various permutations and combinations involving (?!re)
but I just can’t find the answer. In fact I’m beginning to doubt
that (?!re) works at all on my system (Ruby 1.8.6)…but on balance
it’s probably me rather than my Ruby installation!

Thanks and regards,
Andy S.

2

Andrew S. wrote:

In fact I’m beginning to doubt
that (?!re) works at all on my system (Ruby 1.8.6)…but on balance
it’s probably me rather than my Ruby installation!

It’s not you. The ruby 1.8 regexen don’t support negative lookahead.
You’ll
get that in ruby 1.9 or by installing and using the oniguruma
regexp-engine
instead of the default ruby 1.8 one.

HTH,
Sebastian

3

On Dec 13, 11:47 am, Sebastian H. [email protected]
wrote:

Andrew S. wrote:

In fact I’m beginning to doubt
that (?!re) works at all on my system (Ruby 1.8.6)…but on balance
it’s probably me rather than my Ruby installation!

It’s not you. The ruby 1.8 regexen don’t support negative lookahead. You’ll
get that in ruby 1.9 or by installing and using the oniguruma regexp-engine
instead of the default ruby 1.8 one.

Not true:
C:>ruby -v
ruby 1.8.6 (2007-03-13 patchlevel 0) [i386-mswin32]

C:>irb
irb(main):001:0> s = “hello world”
=> “hello world”
irb(main):002:0> s.scan /[eo]./
=> [“el”, "o ", “or”]
irb(main):003:0> s.scan /eo./
=> [“el”, "o "]

It just doesn’t support negative or positive lookbehinds.

4

On Dec 13, 2007 11:09 AM, Phrogz [email protected] wrote:

“doggone it” => 0
Just to shill for a moment:

I was doing that kind of irb stuff so often for regexps in Ruby, I wrote
this:

http://rubyforge.org/projects/regexpbench/
http://regexpbench.rubyforge.org/

Hope it’s useful.

Judson

5

On Dec 13, 11:16 am, Andrew S. [email protected] wrote:

‘cat’ =~ regexp # not nil
irb(main):006:0> tests = [ “”, “dog”, “cat”, “doggone it” ]
=> [“”, “dog”, “cat”, “doggone it”]

irb(main):007:0> tests.each do |str|
irb(main):008:1* hit = (str =~ /^(?!dog).+$|^dog.+$/)
irb(main):009:1> puts “#{str.inspect} => #{hit.inspect}”
irb(main):010:1> end
“” => nil
“dog” => nil
“cat” => 0
“doggone it” => 0

6

On 13 Dec 2007, at 20:06, Judson L. wrote:

Hope it’s useful.
That looks very useful. I’m often twiddling regexps (time-
consumingly) so I’ll give it a try.

Thanks!

Regards,
Andy S.

7

On 13 Dec 2007, at 19:09, Phrogz wrote:

‘dog’ =~ regexp # nil
“” => nil
“dog” => nil
“cat” => 0
“doggone it” => 0

That’s fantastic, thank you!

When I was trying this yesterday, I think I failed to appreciate that
(?!re) doesn’t consume the match. Now I know.

Thanks also for the tip in your previous email about Ruby 1.8 and
lookbehinds.

Regards,
Andy S.