ass big round tit p

[Xwest]

futurama porn

uhh

[nota]

.+@.+(\.com|\.edu|\.org)$


If only.

[Jahi]

.+@.+(\.com|\.edu|\.org)$


If only.

'@' should be escaped too, for most implementations of regexps. I'd say it'll definitely be needed with a scripted language, where '@' defines variables/arrays. Though, I'll beg for '\.br' to be included there.

[Ichron]

This is getting beyond ridiculous.

[Aurelius]

...

[nota]

'@' isn't an operator or pointer in perl/php so it is treated as standard text.


And I suppose we could let the .br in there

[Jahi]

'@' isn't an operator or pointer in perl/php so it is treated as standard text.

Actually, both languages do use '@' to define arrays. You were partially right, though. You don't need to escape the '@' at the regexp (though I'm still not 100% sure about Perl's regexp), but in the text being tested against such regexp, unless you use literal quotes (single quotes):

Code: Select all

cseg@hooligan:~! cat regexp-test.pl
#!/usr/bin/perl

@test = "lololfuckedupthehostname";
$address = "test@test.com";

if ($address =~ /.+@.+(\.com|\.edu|.org)$/i) {
  print("Yeah, '$address' matches .+@.+(\.com|\.edu|\.org)\$\n");
} else {
  print("Nope..\n");
}
cseg@hooligan:~! perl regexp-test.pl
Nope..
cseg@hooligan:~!

Because $address inherited @test's content, becoming "testlololfuckedupthehostnametest.com" and didn't match the regexp. Now with a little change, everything is perfect:

Code: Select all

cseg@hooligan:~! cat regexp-test.pl
#!/usr/bin/perl

@test = "lololfuckedupthehostname";
$address = 'test@test.com';

if ($address =~ /.+@.+(\.com|\.edu|.org)$/i) {
  print("Yeah, '$address' matches .+@.+(\.com|\.edu|\.org)\$\n");
} else {
  print("Nope..\n");
}
cseg@hooligan:~! perl regexp-test.pl
Yeah, 'test@test.com' matches .+@.+(.com|.edu|.org)$
cseg@hooligan:~!

My doubt about Perl's regexp is the fact that you may tell it to search for text that'll be first replaced by a variable/array's content before comparing:

Code: Select all

cseg@hooligan:~! cat regexp-test.pl
#!/usr/bin/perl

@test = "lololfuckedupthehostname";
$address = 'test@test.com';

if ($address =~ /.+@test.+(\.com|\.edu|.org)$/i) {
  print("Yeah, '$address' matches .+@.+(\.com|\.edu|\.org)\$\n");
} else {
  print("Nope..\n");
}
cseg@hooligan:~! perl regexp-test.pl
Nope..
cseg@hooligan:~!

Now it's looking for "<something>lololfuckedupthehostname<something>(.com|.edu.|org)". I just don't know for sure if is there a modifiable "@." array that could be overwritten somehow (for the specific case we were discussing originally), and definitely you'll need to escape that '@' when you have a text following it right away, like in the case above.

To prove my point, however:

Code: Select all

cseg@hooligan:~! cat regexp-test.pl
#!/usr/bin/perl

@test = "lololfuckedupthehostname";
$address = 'test@test.com';

if ($address =~ /.+\@test.+(\.com|\.edu|.org)$/i) {
  print("Yeah, '$address' matches .+@.+(\.com|\.edu|\.org)\$\n");
} else {
  print("Nope..\n");
}
cseg@hooligan:~! perl regexp-test.pl
Nope..
cseg@hooligan:~!

Failed, because it's now looking for "<something>@test<something>(.com|.edu|.org)", and $address doesn't have anything between "@test" and ".com". To finish proving it, then:

Code: Select all

cseg@hooligan:~! cat regexp-test.pl
#!/usr/bin/perl

@test = "lololfuckedupthehostname";
$address = 'test@testlolyessomethinghere.com';

if ($address =~ /.+\@test.+(\.com|\.edu|.org)$/i) {
  print("Yeah, '$address' matches .+@.+(\.com|\.edu|\.org)\$\n");
} else {
  print("Nope..\n");
}
cseg@hooligan:~! perl regexp-test.pl
Yeah, 'test@testlolyessomethinghere.com' matches .+@.+(.com|.edu|.org)$
cseg@hooligan:~!

Works like a charm, since now there IS something there to be matched. I'm too tired to go read pages and pages of manuals to find out about that @. array, though. I'll keep playing safe and escaping every single special character.

And I suppose we could let the .br in there

Yay! <3

[nota]

I usually run with single quotes as it gets parsed faster anyway; due to the fact that the system does parse variables or special characters in to it.

So; $foo='$bar' is literally parsed as $bar.
Whereas $var="$var" is saved to... well; its own value.

Which I'm sure you know (its pretty much what you were saying); but regardless; the variable feat wasn't a problem because i tested it how i usually run code. And trust me; it works. We'd just need to add the Brazilian code to it


Now; if you want to make some crazy code to stop them that'd be great ^_^




TBH though I'm fairly impressed. They get past image confirmation as well as a user-approved activation.

[Yume]

.+@.+(\.com|\.edu|\.org)$


If only.

The spammer that created this thread came through a roadrunner internet account in fairfax, va - probably some zombie machine. He used a .net email address - you can't just filter out that kind of stuff.

TBH though I'm fairly impressed. They get past image confirmation as well as a user-approved activation.

Actually we didn't have a captcha turned on. I installed a 3rd party one a few days ago and we haven't had any bots since.

And (of course) the day after I do that, phpbb3 RC1 is released.

[Jahi]

And (of course) the day after I do that, phpbb3 RC1 is released.

Murphy's Law strikes again.