On Nov 29, 2007 10:59 AM, Peter Smit <peter@xxxxxxxxxxxx> wrote:
> If nobody has a solution I think I'll report it as a bug tomorrow.
did you ever report a bug on this ?
i was messing around with it today, and i discovered that some urls,
partially work. for example the google translate 'api', and php.net
note, i set the user agent for the google site to work. and i say,
partially, because even when data does come back, its not the complete
page, which you can easily realize by navigating to the page and comparing
the source w/ the output from the test script.
i dont know of any ini setting that would influence the amount of space
available to the memory buffer, aside from memory_limit, which i have set
to 128M.
<?php
#$c = curl_init("
http://google.com/translate_t?langpair=en%7Cfr&text=newspaper";);
$c = curl_init("http://php.net";);
$st = fopen('php://memory', 'r');
curl_setopt($c, CURLOPT_FILE, $st);
curl_setopt($c, CURLOPT_USERAGENT, 'Mozilla/5.0 (X11; U; Linux i686; en-US;
rv:1.8.1.11) Gecko/20080115 Firefox/2.0.0.11');
if(!curl_exec($c)) die ("error: ".curl_error($c));
curl_close($c);
rewind($st);
echo stream_get_contents($st);
fclose($st);
?>
-nathan
