I want to create a URL shortener service where you can write a long URL into an input field and the service shortens the URL to \"http://www.example.org/abcdef
\
// simple approach
$original_id = 56789;
$shortened_id = base_convert($original_id, 10, 36);
$un_shortened_id = base_convert($shortened_id, 36, 10);
I have a variant of the problem, in that I store web pages from many different authors and need to prevent discovery of pages by guesswork. So my short URLs add a couple of extra digits to the Base-62 string for the page number. These extra digits are generated from information in the page record itself and they ensure that only 1 in 3844 URLs are valid (assuming 2-digit Base-62). You can see an outline description at http://mgscan.com/MBWL.
Did you omit O, 0, and i on purpose?
I just created a PHP class based on Ryan's solution.
<?php
$shorty = new App_Shorty();
echo 'ID: ' . 1000;
echo '<br/> Short link: ' . $shorty->encode(1000);
echo '<br/> Decoded Short Link: ' . $shorty->decode($shorty->encode(1000));
/**
* A nice shorting class based on Ryan Charmley's suggestion see the link on Stack Overflow below.
* @author Svetoslav Marinov (Slavi) | http://WebWeb.ca
* @see http://stackoverflow.com/questions/742013/how-to-code-a-url-shortener/10386945#10386945
*/
class App_Shorty {
/**
* Explicitly omitted: i, o, 1, 0 because they are confusing. Also use only lowercase ... as
* dictating this over the phone might be tough.
* @var string
*/
private $dictionary = "abcdfghjklmnpqrstvwxyz23456789";
private $dictionary_array = array();
public function __construct() {
$this->dictionary_array = str_split($this->dictionary);
}
/**
* Gets ID and converts it into a string.
* @param int $id
*/
public function encode($id) {
$str_id = '';
$base = count($this->dictionary_array);
while ($id > 0) {
$rem = $id % $base;
$id = ($id - $rem) / $base;
$str_id .= $this->dictionary_array[$rem];
}
return $str_id;
}
/**
* Converts /abc into an integer ID
* @param string
* @return int $id
*/
public function decode($str_id) {
$id = 0;
$id_ar = str_split($str_id);
$base = count($this->dictionary_array);
for ($i = count($id_ar); $i > 0; $i--) {
$id += array_search($id_ar[$i - 1], $this->dictionary_array) * pow($base, $i - 1);
}
return $id;
}
}
?>
You could hash the entire URL, but if you just want to shorten the id, do as marcel suggested. I wrote this Python implementation:
https://gist.github.com/778542
Very good answer, I have created a Golang implementation of the bjf:
package bjf
import (
"math"
"strings"
"strconv"
)
const alphabet = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789"
func Encode(num string) string {
n, _ := strconv.ParseUint(num, 10, 64)
t := make([]byte, 0)
/* Special case */
if n == 0 {
return string(alphabet[0])
}
/* Map */
for n > 0 {
r := n % uint64(len(alphabet))
t = append(t, alphabet[r])
n = n / uint64(len(alphabet))
}
/* Reverse */
for i, j := 0, len(t) - 1; i < j; i, j = i + 1, j - 1 {
t[i], t[j] = t[j], t[i]
}
return string(t)
}
func Decode(token string) int {
r := int(0)
p := float64(len(token)) - 1
for i := 0; i < len(token); i++ {
r += strings.Index(alphabet, string(token[i])) * int(math.Pow(float64(len(alphabet)), p))
p--
}
return r
}
Hosted at github: https://github.com/xor-gate/go-bjf
Here is my PHP 5 class.
<?php
class Bijective
{
public $dictionary = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789";
public function __construct()
{
$this->dictionary = str_split($this->dictionary);
}
public function encode($i)
{
if ($i == 0)
return $this->dictionary[0];
$result = '';
$base = count($this->dictionary);
while ($i > 0)
{
$result[] = $this->dictionary[($i % $base)];
$i = floor($i / $base);
}
$result = array_reverse($result);
return join("", $result);
}
public function decode($input)
{
$i = 0;
$base = count($this->dictionary);
$input = str_split($input);
foreach($input as $char)
{
$pos = array_search($char, $this->dictionary);
$i = $i * $base + $pos;
}
return $i;
}
}