Conversion

String conversion and/or encoding is an important part of exploitation and firewall bypasses.

Convert String/Binary to Hex

If no prefix is needed, you just do the following

  1. "Rubyfu".unpack("H*")    #=> ["527562796675"]

Otherwise, see the below ways

For a single character

  1. '\x%02x' % "A".ord    #=> "\\x41"

Note: the symbols *"" are equal of .join

  1. "ABCD".unpack('H*')[0].scan(/../).map {|h| '\x'+h }.join    #=> "\\x41\\x42\\x43\\x44"

or

  1. "ABCD".unpack('C*').map { |c| '\x%02x' % c }.join    #=> "\\x41\\x42\\x43\\x44"

or

  1. "ABCD".split("").map {|h| '\x'+h.unpack('H*')[0] }*""    #=> "\\x41\\x42\\x43\\x44"

or

  1. "ABCD".split("").map {|c|'\x' + c.ord.to_s(16)}.join    #=> "\\x41\\x42\\x43\\x44"

or

  1. "ABCD".split("").map {|c|'\x' + c.ord.to_s(16)}*""    #=> "\\x41\\x42\\x43\\x44"

or

  1. "ABCD".chars.map {|c| '\x' + c.ord.to_s(16)}*""    #=> "\\x41\\x42\\x43\\x44"

or

  1. "ABCD".each_byte.map {|b| b.to_s(16)}.join    #=> "41424344"

or

  1. "ABCD".each_char.map {|c| '\x'+(c.unpack('H*')[0])}.join    #=> "\\x41\\x42\\x43\\x44"

or

  1. "ABCD".chars.map {|c| '\x%x' % c.ord}.join    #=> "\\x41\\x42\\x43\\x44"

Convert Hex to String/Binary

  1. ["41424344"].pack('H*')    #=> ABCD

or

  1. "41424344".scan(/../).map { |x| x.hex.chr }.join    #=> ABCD

or for raw socket

  1. "41424344".scan(/../).map(&:hex).pack("C*")    #=> ABCD

in-case of binary that is out of .chr range. For example you may need to convert an IP-address to hex raw then send it through the socket. The case of just converting it to hex would not work for you

  1. >> ip = "192.168.100.10"
  2. => "192.168.100.10"
  3. >> ip.split(".").map {|c| '\x%02x' % c.to_i}.join 
  4. => "\\xc0\\xa8\\x64\\x0a"

As you can see, Ruby reads returns "\\xc0\\xa8\\x64\\x0a" which doesn’t equal "\xc0\xa8\x64\x0a". Try to enter this value (with double-quotes) "\xc0\xa8\x64\x0a" into your irb directly and you’ll notice that the return is "\xC0\xA8d\n" which is what should be passed to the raw socket, not the "\\xc0\\xa8\\x64\\x0a". The main cause is ruby escapes the backslash(\).

To solve this issue, use pack to convert integers to 8-bit unsigned (unsigned char)

  1. ip.split(".").map(&:to_i).pack("C*")    #=> "\xC0\xA8d\n"

Note about hex: Sometimes you might face non-printable characters, especially when dealing with binary raw. In this case, append (# -*- coding: binary -*-) at the top of your file to fix any interpretation issues.

Convert Hex (Return address) to Little-Endian format

Little-endian format is simply reversing the string such as reversing/backwarding “Rubyfu” to “ufybuR” which can be done by calling the reverse method of the String class

  1. "Rubyfu".reverse

In exploitation, this is not as simple as that since we’re dealing with hex values that may not represent printable characters.

So assume we have 0x77d6b141 as the return address which we want to convert to Little-Endian format to allow the CPU to read it correctly.

Generally speaking, it’s really a trivial task to convert 0x77d6b141 to \x41\xb1\xd6\x77 since it’s a one time process, but this is not the case if you have a ROP chain that has to be staged in your exploit. To do so simply pack it as an array

  1. [0x77d6b141].pack('V')

It happens that sometimes you get an error because of a non-Unicode string issue. To solve this issue, just force encoding to UTF-8, but most of the time you will not face this issue

  1. [0x77d6b141].pack('V').force_encoding("UTF-8")

If you have a ROP chain, then it’s not decent to apply this each time - so you can use the first way and append (# -*- coding: binary -*-) at top of your exploit file.

Convert to Unicode Escape

Hexadecimal unicode escape

  1. "Rubyfu".each_char.map {|c| '\u' + c.ord.to_s(16).rjust(4, '0')}.join

Or using unpack

  1. "Rubyfu".unpack('U*').map{ |i| '\u' + i.to_s(16).rjust(4, '0') }.join

A shorter way

  1. "Rubyfu".unpack('U*').map{ |i| "\\u00%x" % i }.join

Octal unicode escape

An octal escape is exactly the same, except we convert the string to octal instead of hexadecimal

  1. "Rubyfu".each_char.map {|c| '\u' + c.ord.to_s(8).rjust(4, '0')}.join

Escape Sequences in Double-Quoted Strings

  1. "\u{52 75 62 79 66 75}"

En/Decode base-64 String

We’ll present this in a few ways.

Encode string

  1. ["RubyFu"].pack('m0')

or

  1. require 'base64'
  2. Base64.encode64 "RubyFu"

Decode

  1. "UnVieUZ1".unpack('m0')

or

  1.  Base64.decode64 "UnVieUZ1"

TIP:
The string unpack method is incredibly useful for converting data we read as strings back to their original form. To read more, visit the String class reference at www.ruby-doc.org/core/classes/String.html.

En/Decode URL String

URL encoding/decoding is well known. From a hacker’s point of view, we need it often for client-side vulnerabilities.

Encoding string

  1. require 'uri'
  2. puts URI.encode 'http://vulnerable.site/search.aspx?txt="><script>alert(/Rubyfu/.source)</script>'

Decoding string

  1. require 'uri'
  2. puts URI.decode "http://vulnerable.site/search.aspx?txt=%22%3E%3Cscript%3Ealert(/Rubyfu/.source)%3C/script%3E"

You can encode/decode any non-URL string, of-course.

The above way will encode any non-URL standard strings only (ex. <>"{}) however if you want to encode the full string use URI.encode_www_form_component

  1. puts URI.encode_www_form_component 'http://vulnerable.site/search.aspx?txt="><script>alert(/Rubyfu/.source)</script>'

HTML En/Decode

Encoding HTML

  1. require 'cgi'
  2. CGI.escapeHTML('"><script>alert("Rubyfu!")</script>')

Returns

  1. &quot;&gt;&lt;script&gt;alert(&quot;Rubyfu!&quot;)&lt;/script&gt;

Decoding HTML

  1. require 'cgi'
  2. CGI.unescapeHTML("&quot;&gt;&lt;script&gt;alert(&quot;Rubyfu!&quot;)&lt;/script&gt;")

Returns

  1. "><script>alert("Rubyfu!")</script>

En/Decode SAML String

Decoding SAML

  1. # SAML Request 
  2. saml = "fZJNT%2BMwEIbvSPwHy%2Fd8tMvHympSdUGISuwS0cCBm%2BtMUwfbk%2FU4zfLvSVMq2Euv45n3fd7xzOb%2FrGE78KTRZXwSp5yBU1hpV2f8ubyLfvJ5fn42I2lNKxZd2Lon%2BNsBBTZMOhLjQ8Y77wRK0iSctEAiKLFa%2FH4Q0zgVrceACg1ny9uMy7rCdaM2%2Bs0BWrtppK2UAdeoVjW2ruq1bevGImcvR6zpHmtJ1MHSUZAuDKU0vY7Si2h6VU5%2BiMuJuLx65az4dPql3SHBKaz1oYnEfVkWUfG4KkeBna7A%2Fxm6M14j1gZihZazBRH4MODcoKPOgl%2BB32kFz08PGd%2BG0JJIkr7v46%2BhRCaEpod17DCRivYZCkmkd4N28B3wfNyrGKP5bws9DS6PKDz%2FMpsl36Tyz%2F%2Fax1jeFmi0emcLY7C%2F8SDD0Z7dobcynHbbV3QVbcZW0TlqQemNhoqzJD%2B4%2Fn8Yw7l8AA%3D%3D"
  4. require 'cgi'
  5. require 'base64'
  6. require 'zlib'
  8. inflated = Base64::decode64(CGI.unescape(saml))
  9. # You don't need below code if it's not deflated/compressed
  10. zlib = Zlib::Inflate.new(-Zlib::MAX_WBITS)
  11. zlib.inflate(inflated)

Returns

  1. "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\r\n<samlp:AuthnRequest xmlns:samlp=\"urn:oasis:names:tc:SAML:2.0:protocol\" ID=\"agdobjcfikneommfjamdclenjcpcjmgdgbmpgjmo\" Version=\"2.0\" IssueInstant=\"2007-04-26T13:51:56Z\" ProtocolBinding=\"urn:oasis:names:tc:SAML:2.0:bindings:HTTP-POST\" ProviderName=\"google.com\" AssertionConsumerServiceURL=\"https://www.google.com/a/solweb.no/acs\" IsPassive=\"true\"><saml:Issuer xmlns:saml=\"urn:oasis:names:tc:SAML:2.0:assertion\">google.com</saml:Issuer><samlp:NameIDPolicy AllowCreate=\"true\" Format=\"urn:oasis:names:tc:SAML:2.0:nameid-format:unspecified\" /></samlp:AuthnRequest>\r\n"

Source
[More about SAML][3]

[3]: http://dev.gettinderbox.com/2013/12/16/introduction-to-saml/