Understanding PHP multi_curl_exec loops

August 29, 2012
php
curl
php
understanding

In the PHP curl_multi_exec manual, a strange 3 loops are given as an example:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12


<?php
do {
    $mrc = curl_multi_exec($mh, $active);
} while ($mrc == CURLM_CALL_MULTI_PERFORM);

while ($active && $mrc == CURLM_OK) {
    if (curl_multi_select($mh) != -1) {
        do {
            $mrc = curl_multi_exec($mh, $active);
        } while ($mrc == CURLM_CALL_MULTI_PERFORM);
    }
}

I was wondering why we needed so many while loops as opposed to a simple one such as:

1
2
3
4


<?php
do {
    curl_multi_exec($master,$running);
} while($running &gt; 0);

What I have found is mostly due to prevent making unnecessary calls to curl_multi_exec. Here the same first snippet with comments:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18


<?php
// Launch the requests. This is non blocking,
// and returns CURLM_CALL_MULTI_PERFORM as long as there is something to do
// (in this case, launch the remaining requests)
do {
    $mrc = curl_multi_exec($mh, $active);
} while ($mrc == CURLM_CALL_MULTI_PERFORM);

// Actually do the fetching job
while ($active && $mrc == CURLM_OK) {
    // This function is blocking and waits until there is activity
    if (curl_multi_select($mh) != -1) {
        do {
            // Run the connections
            $mrc = curl_multi_exec($mh, $active);
        } while ($mrc == CURLM_CALL_MULTI_PERFORM);
    }
}

With those loops, we only run curl_multi_exec when something changed. The second snipped would make unnecessary calls to this function and shoot up our CPU usage.