PHP Script to extract all email addresses from a mail box over IMAP protocol Usage: ./imap-addressess-extractor.php INBOX 0 1000 first parameter is the mailbox name second is the offset third is the quantity message to scan

imap-addressess-extractor.php

#!/usr/bin/php
<?php

class Contact {
  public $name;
	public $email;

	public function __construct($name = "", $email = "") {
		$this->name = $name;
		$this->email = $email;
	}

	public function toArray() {
		return array($this->name, $this->email);
	}
}

$fp = fopen('file.csv', 'w');

$mailboxName = "INBOX";
if(isset($argv[1]) && $argv[1] != "list") {
	$mailboxName = $argv[1];
}

$server = "server"
$username = "user";
$password = "password";
$port = 143
$mailboxPath = "{$server:$port/novalidate-cert/norsh/notls}$mailboxName";

$mbox = @imap_open($mailboxPath, $username, $password) 
	or die("can't connect: " . imap_last_error());

if(isset($argv[1]) && $argv[1] == "list") {
	$list = imap_list($mbox, $mailboxPath, "*");

	is_array($list) or die("imap_list failed: " . imap_last_error());

	foreach ($list as $val) {
		echo imap_utf7_decode($val) . "\n";
	}
	exit(0);
}

$offset = 0;
if(isset($argv[2]) && is_int($argv[2])) {
	$offset = $argv[2];
}
$quantity = 0;
if(isset($argv[3]) && $argv[3] > 0) {
        $quantity = $argv[3];
}

$percent = 0;
$numMessages = imap_num_msg($mbox);
for ($i = $numMessages - $offset; $i > ($numMessages - $offset - $quantity); $i--) {
	$header = @imap_headerinfo($mbox, $i);
	if(!$header instanceof StdClass) {
		$retries = 0;
		while($retries < 5) {
			echo "Re-openning $retries\n";
			if(imap_reopen($mbox, $mailboxPath)) {
	                	$header = @imap_headerinfo($mbox, $i);
				break;
			}
			$retries++;
		}
		if(!$header instanceof StdClass) continue;
	}
	if(isset($header->from[0])) {
		$from = $header->from[0];
		$contact = new Contact(isset($from->personal) ? $from->personal : "", $from->mailbox . "@" . $from->host);
		fputcsv($fp, $contact->toArray());
	}
        if(isset($header->to)) {
		$toContacts = $header->to;
		foreach($toContacts as $to) {
			$contact = new Contact(isset($to->personal) ? $to->personal : "", $to->mailbox . "@" . $to->host);
			fputcsv($fp, $contact->toArray());
		}
	}
        if(isset($header->cc)) {
                $ccContacts = $header->cc;
                foreach($ccContacts as $cc) {
                        $contact = new Contact(isset($cc->personal) ? $cc->personal : "", $cc->mailbox . "@" . $cc->host);
                        fputcsv($fp, $contact->toArray());
                }
        }
	if( $percent != (int)((($numMessages - $offset - $i) / $quantity) * 100) && (($percent = (int)((($numMessages - $offset - $i) / $quantity) * 100)) % 1) == 0) echo "$percent% mails: " . ($numMessages - $i - $offset) . "\n";
}
echo "Mailbox: $mailboxName\n";
echo "Number of messages: $numMessages\n";
echo "Offset: $offset\n";
echo "Quantity: $quantity\n";

imap_close($mbox);
fclose($fp);

The you could use something like

$ cat file.csv | awk -F "\"*,\"*" '{print $2}' | sort | uniq

to extract unique addresses