Understanding aRegexp === aString


#1

i’m experiencing with regexp and ruby and follo the page
http://www.rubycentral.com/book/ref_c_regexp.html

at “===” i’ve found an example :

a = “HELLO”
case a
when /^a-z*$/; print “Lower case\n”
when /^A-Z*$/; print “Upper case\n”
else; print “Mixed case\n”
end

saying it produces :

=> Upper case

i got :

Mixed case

why this discrapency, if any ???

is that due to the fact my script file is UTF-8 encoded ???

also with :
truc = ‘toto’
rgx=Regexp.new(’^toto$’)
flag=(truc =~ rgx)? true : false
p “flag = #{flag}”

=> “flag = true”

flag=(truc === rgx) /// this one i don’t understand the result
p “flag = #{flag}”

=> “flag = false”

flag=(truc == rgx)
p “flag = #{flag}”

=> “flag = false”

flag=(truc =~ rgx)
p “flag = #{flag}”

=> “flag = 0”

why, when aString = ‘toto’ and ARegexp = Regexp.new(’^toto$’)

the equality “===” returns false ???


#2

Ross B. wrote:

On Thu, 2006-03-23 at 23:13 +0900, Une bévue wrote:

the equality “===” returns false ???

Adding to Ross’ excellent explanation: it helps to not view “===” as
an equality operator. Rather it’s a general matching operator, whatever
matching means in a certain context. For RX it will do an RX match, for
class objects it will do the kind_of? check, for a range it will check
include? etc. Btw, has anyone an idea why this does not apply to
Enumerable?

[1,2,3].include? 2
=> true

[1,2,3] === 2
=> false

Kind regards

robert

#3

On Thu, 2006-03-23 at 23:13 +0900, Une bévue wrote:

end

saying it produces :

=> Upper case

i got :

Mixed case

why this discrapency, if any ???

I guess that’s a typo on the site, it’s correct in my printed 2nd ed:

a = “HELLO”
case a
when /^[a-z]$/; print “Lower case\n”
when /^[A-Z]
$/; print “Upper case\n”
else; print “Mixed case\n”
end

-> Upper Case

(Notice the [] brackets that denote a character set)

p “flag = #{flag}”

=> “flag = false”

flag=(truc =~ rgx)
p “flag = #{flag}”

=> “flag = 0”

why, when aString = ‘toto’ and ARegexp = Regexp.new(’^toto$’)

the equality “===” returns false ???

Try this:

truc = ‘toto’
rgx=Regexp.new(’^toto$’)
flag=(truc =~ rgx)? true : false
p “flag = #{flag}”

=> “flag = true”

flag=(rgx === truc) # /// switch these around
p “flag = #{flag}”

=> “flag = true”

flag=(truc == rgx)
p “flag = #{flag}”

# => “flag = false”

flag=(truc =~ rgx)
p “flag = #{flag}”

# => “flag = 0”

See also the ‘what on earth…’ thread that’s been on the list recently
(yesterday/today I think).


#4

Robert K. wrote:

Enumerable?

[1,2,3].include? 2
=> true

[1,2,3] === 2
=> false

It would be nice, but then how would this work:

irb(main):001:0> case [1,2,3]
irb(main):002:1> when [1,2,3]
irb(main):003:1> puts “same array!”
irb(main):004:1> end
same array!
=> nil


#5

Ross B. removed_email_address@domain.invalid wrote:

(Notice the [] brackets that denote a character set)

fine, thanks, difficult to catch out typos particularly in regexps…

Try this:

[…]

flag=(rgx === truc) # /// switch these around
p “flag = #{flag}”

=> “flag = true”

i did both :
flag=(truc === rgx)
p “flag = #{flag} for flag=(truc === rgx)”

=> “flag = false for flag=(truc === rgx)”"

flag=(rgx === truc)
p “flag = #{flag} for flag=(rgx === truc)”

=> “flag = true for flag=(rgx === truc)”

that’s a really strange behaviour, to me, of ruby, a “symetrical”
operator symbol being not commutative ???

better to know :wink:

See also the ‘what on earth…’ thread that’s been on the list recently
(yesterday/today I think).

ok, thanks very much !


#6

I’m trying to use builder to construct jnlp files. I need to create
elements with “-” in the name and can’t. For example I need to output
the following xml fragment:

Here’s what I tried with builder:

x = Builder::XmlMarkup.new(:target => $stdout, :indent => 1)

x.security {
x.all-permissions
}

but the “-” is parsed in the name and this produces a NameError

NameError: undefined local variable or method `permissions’ for
main:Object

Is there a way to do this?


#7

Robert K. wrote:

Enumerable?

[1,2,3].include? 2
=> true

[1,2,3] === 2
=> false

It would be nice, but then how would this work:

irb(main):001:0> case [1,2,3]
irb(main):002:1> when [1,2,3]
irb(main):003:1> puts “same array!”
irb(main):004:1> end
same array!
=> nil


#8

YANAGAWA Kazuhisa wrote:

prints “foo”

Good point! Thanks to you and Joel!

Kind regards

robert

#9

In Message-Id: removed_email_address@domain.invalid
Robert K. removed_email_address@domain.invalid writes:

[1,2,3].include? 2
=> true

[1,2,3] === 2
=> false

Once that can be but “fixed” later since… it can be a seed of
confusion and we can use “*” for array on when clause.

n = 2
array = [2, 4, 6]

case n
when *array
puts “foo”
else
puts “bar”
end

prints “foo”