Topic: Spam scrambler
+FuckAlms !vX8K53rFBI — 13.6 years ago #18,066
I'm trying to make a function that takes urls that are formatted incorrectly (a href) and turns it into a mailto address using the given domain and dropping any additional address information.
so
becomes
for reference:
corresponds to:
so
a href=http://www.gofuckashitupyourar.se/boners.php
becomes
mailto:abuse@gofuckashitupyourar.se
for reference:
#5 prevent botspam html links '/a href="/', #6 Linkify URLs. '@\b(?<!\[)(https?|ftp)://(www\.)?([A-Z0-9.-]+)(/)?([A-Z0-9/&#+%~=_|?.,!:;-]*[A-Z0-9/&#+%=~_|])?@i', #7 Linkify text in the form of <a href="http://example.org" title="http://example.org" rel="nofollow">text</a>. '@\[(https?|ftps?|mailto)://([A-Z0-9/&#+%~=_|?.,!:;-]+) (.+?)\]@i',corresponds to:
'<a href="mailto:admin@">spam</a>', #5 '<a href="$0" rel="nofollow">$1://$2$3$4$5</a>', #6 '<a href="$1://$2" title="$1://$2" rel="nofollow">$3</a>', #7+Anonymous B — 13.6 years ago, 22 minutes later[T] [B] #220,868
Let me know if you need a hand+Sarcasm.zip !36Q3VF.UPM — 13.6 years ago, 1 minute later, 23 minutes after the original post[T] [B] #220,869
Clever.
+Anonymous D — 13.6 years ago, 5 hours later, 6 hours after the original post[T] [B] #220,945
+Anonymous E — 13.6 years ago, 3 days later, 3 days after the original post[T] [B] #223,334
Depends on your langugage/engine, but here's one using two substitutions with GNU sed in ERE mode.
echo http://www.gofuckashitupyourar.se/boners.php | sed -r -n -e 's#(https?|ftp)://([^/]*).*#\2#;s/.*\.([^\.]*\.[^\.]*$)/\1/p;'
And the same thing in a slightly less robust single subtitution:
echo http://www.gofuckashitupyourar.se/boners.php | sed -r -n -e 's#(https?|ftp)://.*\.([^\./]*\.[^\./]*).*#\2#p'
echo http://www.gofuckashitupyourar.se/boners.php | sed -r -n -e 's#(https?|ftp)://([^/]*).*#\2#;s/.*\.([^\.]*\.[^\.]*$)/\1/p;'
And the same thing in a slightly less robust single subtitution:
echo http://www.gofuckashitupyourar.se/boners.php | sed -r -n -e 's#(https?|ftp)://.*\.([^\./]*\.[^\./]*).*#\2#p'
·FuckAlms !vX8K53rFBI (OP) — 13.6 years ago, 7 hours later, 4 days after the original post[T] [B] #223,451
+Anonymous F — 13.6 years ago, 13 minutes later, 4 days after the original post[T] [B] #223,454
Ask Syntax, he claims to have invented php
+Anonymous G — 13.6 years ago, 6 minutes later, 4 days after the original post[T] [B] #223,458
@previous (F)
You obviously haven't been paying attention, Syntax invented the computer whilst out free climbing the entire grand canyon.
You obviously haven't been paying attention, Syntax invented the computer whilst out free climbing the entire grand canyon.
·FuckAlms !vX8K53rFBI (OP) — 13.6 years ago, 5 minutes later, 4 days after the original post[T] [B] #223,464
@223,454 (F)
@previous (G)
ok, that's enough. this thread isn't about "Syntax" or his life. this thread is about foiling the infernal cuntbots I have to deal with on OTK.
@previous (G)
ok, that's enough. this thread isn't about "Syntax" or his life. this thread is about foiling the infernal cuntbots I have to deal with on OTK.
·Anonymous F — 13.6 years ago, 12 minutes later, 4 days after the original post[T] [B] #223,471
@223,458 (G)
I'm afraid that it's you who hasn't been paying attention. Syntax created the grand canyon back when he was personally advising the Earth on how to form.
I'm afraid that it's you who hasn't been paying attention. Syntax created the grand canyon back when he was personally advising the Earth on how to form.
+Anonymous H — 13.6 years ago, 13 minutes later, 4 days after the original post[T] [B] #223,480
lol regex
·Anonymous G — 13.6 years ago, 34 minutes later, 4 days after the original post[T] [B] #223,502
·Anonymous E — 13.6 years ago, 10 minutes later, 4 days after the original post[T] [B] #223,508
PHP version:
<?php
$url = 'http://www.gofuckashitupyourar.se/boners.php';
$mailto = preg_replace('#(https?|ftp)://.*\.([^\./]*\.[^\./]*).*#', "mailto:abuse@$2", $url);
echo $mailto;
echo "\n";
?>
But I don't have an interpreter on this box to test, so YMMV.
<?php
$url = 'http://www.gofuckashitupyourar.se/boners.php';
$mailto = preg_replace('#(https?|ftp)://.*\.([^\./]*\.[^\./]*).*#', "mailto:abuse@$2", $url);
echo $mailto;
echo "\n";
?>
But I don't have an interpreter on this box to test, so YMMV.
·Anonymous G — 13.6 years ago, 1 minute later, 4 days after the original post[T] [B] #223,509
·Anonymous E — 13.6 years ago, 29 minutes later, 4 days after the original post[T] [B] #223,528
@previous (G)
> What's YMMV?
It's an acronym. It expands to "Ask google and stop wasting other people's time."
> What's YMMV?
It's an acronym. It expands to "Ask google and stop wasting other people's time."
·Anonymous G — 13.6 years ago, 2 minutes later, 4 days after the original post[T] [B] #223,532
·FuckAlms !vX8K53rFBI (OP) — 13.6 years ago, 23 hours later, 5 days after the original post[T] [B] #224,097
@223,508 (E)
Alright, but it doesn't take into account that I can only identify an offending url by the fact that it is input as full html rather than just the url. Namely, the
Also, the parsing is done through an array, so things will need to be structured a bit differently:
Alright, but it doesn't take into account that I can only identify an offending url by the fact that it is input as full html rather than just the url. Namely, the
<a href=
Also, the parsing is done through an array, so things will need to be structured a bit differently:
$markup = array ( # foil bots dumping raw html by breaking links '/a href=#(https?|ftp)://.*\.([^\./]*\.[^\./]*).*#');function parse($text) { global $markup; static $html; $text = htmlspecialchars($text); $text = str_replace("\r", '', $text); if(!$html){ $html = array ( '<a href="mailto:abuse@$2", $url>spam</a>');·Anonymous E — 13.6 years ago, 22 hours later, 6 days after the original post[T] [B] #224,876
@previous (FuckAlms !vX8K53rFBI)
I don't think you understand how this works. The first character of the pattern, #, sets up the delimiter. I chose not to use the usual / because / would be found in the pattern. The "standard" fallbacks are # and |, but I tend to prefer #.
Also, using global is awful and should be avoided.
Also, declaring a static local variable inside a function is horrible and the fact that PHP lets you get away with this is no excuse. Just don't.
Other than that I can't figure out what your sample code is supposed to be doing. A larger context would be useful.
Or, just stop using PHP. It's a horrible language that should not be used by anyone for any reason.
I don't think you understand how this works. The first character of the pattern, #, sets up the delimiter. I chose not to use the usual / because / would be found in the pattern. The "standard" fallbacks are # and |, but I tend to prefer #.
Also, using global is awful and should be avoided.
Also, declaring a static local variable inside a function is horrible and the fact that PHP lets you get away with this is no excuse. Just don't.
Other than that I can't figure out what your sample code is supposed to be doing. A larger context would be useful.
Or, just stop using PHP. It's a horrible language that should not be used by anyone for any reason.
·FuckAlms !vX8K53rFBI (OP) — 13.6 years ago, 1 day later, 1 week after the original post[T] [B] #225,812
@previous (E)
Larger context? TC says it's too much context, so have a pastebin:
http://pastebin.com/s4ZCcNYV
The stuff I'm trying to add to is at line 550.
Larger context? TC says it's too much context, so have a pastebin:
http://pastebin.com/s4ZCcNYV
The stuff I'm trying to add to is at line 550.
(Edited 1 minute later.)
·Anonymous E — 13.5 years ago, 2 days later, 1 week after the original post[T] [B] #228,713
@previous (FuckAlms !vX8K53rFBI)
After line 607 insert the following line:
'#(https?|ftp)://.*\.([^\./]*\.[^\./]*).*#',
After line 633, which is now 634 due to the above inserted line, insert the following line:
'mailto:abuse@$2',
Note that this regex will pick up things which are not inside anchor tags, which you may not want. If this is a problem it could be revised.
After line 607 insert the following line:
'#(https?|ftp)://.*\.([^\./]*\.[^\./]*).*#',
After line 633, which is now 634 due to the above inserted line, insert the following line:
'mailto:abuse@$2',
Note that this regex will pick up things which are not inside anchor tags, which you may not want. If this is a problem it could be revised.
+Anonymous I — 13.5 years ago, 11 hours later, 1 week after the original post[T] [B] #229,214
@previous (E)
If you want it to only slurp anchors of a very specific form make the pattern this:
'#(<a href=["\]?)(https?|ftp)://.*\.([^\./]*\.[^\./]*)[^">'\'']*#'
And make the replacement this:
'$1mailto:abuse@$3'
In all probability you will want it to work on any anchor, not just ones which look exactly like <a href=, so...
'#(<a.*href=["\']?)(https?|ftp)://.*\.([^\./]*\.[^\./]*)[^">\' ]*#'
Which will allow (and preserve) other attributes before and after the href attribute.
If you want it to only slurp anchors of a very specific form make the pattern this:
'#(<a href=["\]?)(https?|ftp)://.*\.([^\./]*\.[^\./]*)[^">'\'']*#'
And make the replacement this:
'$1mailto:abuse@$3'
In all probability you will want it to work on any anchor, not just ones which look exactly like <a href=, so...
'#(<a.*href=["\']?)(https?|ftp)://.*\.([^\./]*\.[^\./]*)[^">\' ]*#'
Which will allow (and preserve) other attributes before and after the href attribute.
·Anonymous G — 13.5 years ago, 5 minutes later, 1 week after the original post[T] [B] #229,215
'mailto:abuse@$2', vs '$1mailto:abuse@$3'
Makes perfect sense really. You do the math
Makes perfect sense really. You do the math
·Anonymous I — 13.5 years ago, 9 hours later, 1 week after the original post[T] [B] #229,391
@previous (G)
Not sure if serious or just confused.
Since I added a capture pattern for the beginning of the anchor (and whatever non-href attributes preceded href) it is necessary to expand that capture in the substitution. Since I added a capture what was formerly the 2nd capture becomes the 3rd.
The only unknown, to me, is whether PHP's parser is smart enough to realize that $1mailto is $1 followed by mailto and not an identifier by itself. Typically identifiers may not start with numbers, but PHP does a lot of things that are not typical (or wise). In addition the $1 inside a PHP substitution stanza is sort-of magic in that it's not a variable an will expand even when interpolation is disabled, so I'm just guessing that it will behave as it should.
Did I mention that testing is only for uncool, ugly people?
Not sure if serious or just confused.
Since I added a capture pattern for the beginning of the anchor (and whatever non-href attributes preceded href) it is necessary to expand that capture in the substitution. Since I added a capture what was formerly the 2nd capture becomes the 3rd.
The only unknown, to me, is whether PHP's parser is smart enough to realize that $1mailto is $1 followed by mailto and not an identifier by itself. Typically identifiers may not start with numbers, but PHP does a lot of things that are not typical (or wise). In addition the $1 inside a PHP substitution stanza is sort-of magic in that it's not a variable an will expand even when interpolation is disabled, so I'm just guessing that it will behave as it should.
Did I mention that testing is only for uncool, ugly people?
·Anonymous G — 13.5 years ago, 15 minutes later, 1 week after the original post[T] [B] #229,401
·Anonymous I — 13.5 years ago, 2 hours later, 1 week after the original post[T] [B] #229,551
Okay, so I had to actually put it in a PHP script to make it work and it now relies on one PCRE feature so it's not portable, but....
'#(<a.*href=["\']?)(https?|ftp)://([^/ ">\']*?)\.?([a-zA-Z0-9-]+\.[a-zA-Z0-9-]+\.?)([/ ">\'][^ ">\']*)#'
'$1mailto:abuse@$4'
Test cases:
<a href=http://one.two>http://one.two</a> [P]
<a href=http://zero.one.two>http://zero.one.two</a> [P]
<a href=http://one.two/relative/path>http://one.two/relative/path</a> [P]
<a href=http://zero.one.two/relative/path>http://zero.one.two/relative/path</a> [P]
[<a href= http://one.two/relative/path]" rel="nofollow">http://one.two/relative/path">http://one.two/relative/path</a></a> [P]
[<a href= http://zero.one.two/relative/path]" rel="nofollow">http://zero.one.two/relative/path">http://zero.one.two/relative/path</a></a> [P]
<a class="something" href="http://one.two/relative/path" rel="nofollow">http://one.two/relative/path">http://one.two/relative/path</a></a> [P]
<a class="something" href="http://zero.one.two/relative/path" rel="nofollow">http://zero.one.two/relative/path">http://zero.one.two/relative/path</a></a> [P]
<a class="something" href="http://one.two/relative/path" target="_blank">http://one.two/relative/path</a> [P]
<a class="something" href="http://zero.one.two/relative/path" target="_blank">http://zero.one.two/relative/path</a> [P]
<a href='http://one.two/relative/path'>http://one.two/relative/path</a> [P]
<a href='http://zero.one.two/relative/path'>http://zero.one.two/relative/path</a> [P]
<a href='http://one.two/relative/path' target="_blank">http://one.two/relative/path</a> [P]
<a href='http://zero.one.two/relative/path' target="_blank">http://zero.one.two/relative/path</a> [P]
<a href=http://one.two/relative/path target="_blank">http://one.two/relative/path</a> [P]
<a href=http://zero.one.two/relative/path target="_blank">http://zero.one.two/relative/path</a> [P]
<a class="something" href=http://one.two/relative/path>http://one.two/relative/path</a> [P]
<a class="something" href=http://zero.one.two/relative/path>http://zero.one.two/relative/path</a> [P]
'#(<a.*href=["\']?)(https?|ftp)://([^/ ">\']*?)\.?([a-zA-Z0-9-]+\.[a-zA-Z0-9-]+\.?)([/ ">\'][^ ">\']*)#'
'$1mailto:abuse@$4'
Test cases:
<a href=http://one.two>http://one.two</a> [P]
<a href=http://zero.one.two>http://zero.one.two</a> [P]
<a href=http://one.two/relative/path>http://one.two/relative/path</a> [P]
<a href=http://zero.one.two/relative/path>http://zero.one.two/relative/path</a> [P]
[<a href= http://one.two/relative/path]" rel="nofollow">http://one.two/relative/path">http://one.two/relative/path</a></a> [P]
[<a href= http://zero.one.two/relative/path]" rel="nofollow">http://zero.one.two/relative/path">http://zero.one.two/relative/path</a></a> [P]
<a class="something" href="http://one.two/relative/path" rel="nofollow">http://one.two/relative/path">http://one.two/relative/path</a></a> [P]
<a class="something" href="http://zero.one.two/relative/path" rel="nofollow">http://zero.one.two/relative/path">http://zero.one.two/relative/path</a></a> [P]
<a class="something" href="http://one.two/relative/path" target="_blank">http://one.two/relative/path</a> [P]
<a class="something" href="http://zero.one.two/relative/path" target="_blank">http://zero.one.two/relative/path</a> [P]
<a href='http://one.two/relative/path'>http://one.two/relative/path</a> [P]
<a href='http://zero.one.two/relative/path'>http://zero.one.two/relative/path</a> [P]
<a href='http://one.two/relative/path' target="_blank">http://one.two/relative/path</a> [P]
<a href='http://zero.one.two/relative/path' target="_blank">http://zero.one.two/relative/path</a> [P]
<a href=http://one.two/relative/path target="_blank">http://one.two/relative/path</a> [P]
<a href=http://zero.one.two/relative/path target="_blank">http://zero.one.two/relative/path</a> [P]
<a class="something" href=http://one.two/relative/path>http://one.two/relative/path</a> [P]
<a class="something" href=http://zero.one.two/relative/path>http://zero.one.two/relative/path</a> [P]
·Anonymous G — 13.5 years ago, 8 minutes later, 1 week after the original post[T] [B] #229,554
·Anonymous I — 13.5 years ago, 18 seconds later, 1 week after the original post[T] [B] #229,555
Since I had to go through the trouble of making a test case to get this fixed, here it is in all its glory
http://pastebin.com/WAd1FVZV
Oooh-aahh.
http://pastebin.com/WAd1FVZV
Oooh-aahh.
·Anonymous I — 13.5 years ago, 35 seconds later, 1 week after the original post[T] [B] #229,556
@229,554 (G)
> >Did I mention that testing is only for uncool, ugly people?
Anon has spotted the joke: I am BOTH of those.
> >Did I mention that testing is only for uncool, ugly people?
Anon has spotted the joke: I am BOTH of those.
(Edited 1 minute later.)
·Anonymous G — 13.5 years ago, 5 minutes later, 1 week after the original post[T] [B] #229,563
·Anonymous I — 13.5 years ago, 24 minutes later, 1 week after the original post[T] [B] #229,598
@previous (G)
> Well, that works. But does it require a proto part?
The pastebin's version does not.
> Well, that works. But does it require a proto part?
The pastebin's version does not.
·Anonymous G — 13.5 years ago, 1 hour later, 1 week after the original post[T] [B] #229,647
·Anonymous I — 13.5 years ago, 3 days later, 2 weeks after the original post[T] [B] #232,340
Bug:
URLs that are domains only that are unquoted that are not followed by any attributes or whitespace before the close tag (>) will fail to match properly and all text up to the next > after the open anchor tag will be improperly replaced.
The fix is this very-slightly-altered regex:
'#(<a.*href=["\']?)((https?|ftp)://)?([^/ ">\']*?)\.?([a-zA-Z0-9-]+\.[a-zA-Z0-9-]+\.?)([/ "\']?[^ ">\']*)#',
Which has been minimally tested and appears not to break any other cases.
URLs that are domains only that are unquoted that are not followed by any attributes or whitespace before the close tag (>) will fail to match properly and all text up to the next > after the open anchor tag will be improperly replaced.
The fix is this very-slightly-altered regex:
'#(<a.*href=["\']?)((https?|ftp)://)?([^/ ">\']*?)\.?([a-zA-Z0-9-]+\.[a-zA-Z0-9-]+\.?)([/ "\']?[^ ">\']*)#',
Which has been minimally tested and appears not to break any other cases.
·Anonymous G — 13.5 years ago, 14 minutes later, 2 weeks after the original post[T] [B] #232,356
·Anonymous I — 13.5 years ago, 34 minutes later, 2 weeks after the original post[T] [B] #232,398
@previous (G)
> Bugs?
> >More testing needed
Release early, release often and let the users test for you. Maximum efficiency.
> Bugs?
> >More testing needed
Release early, release often and let the users test for you. Maximum efficiency.
+Anonymous J — 13.5 years ago, 5 hours later, 2 weeks after the original post[T] [B] #232,643
·FuckAlms !vX8K53rFBI (OP) — 13.5 years ago, 1 day later, 2 weeks after the original post[T] [B] #233,826
now it's just doing the "abuse@" part and nothing else for the hyperlink
http://otakutalk.org/topic/1253
http://otakutalk.org/topic/1253
Start a new topic to continue this conversation.
Or browse the latest topics.