yam has asked for the wisdom of the Perl Monks concerning the following question:
The first question I have is concerning LWP::UserAgent and enabling HTTP/1.0. From what I can tell, you enable the $ENV{PERL_LWP_USE_HTTP_10} by setting it to TRUE, so I set it equal to 1, thusly:
$ENV{PERL_LWP_USE_HTTP_10} = 1;
Is this right? If so, why are my requests still HTTP/1.1?
Second, the client is sending non-canonicalized headers (as determined by a sniffer trace), and I want to replicate them. I'm using HTTP::Headers but the headers are still "right." Is this the right way to do this?
Code follows:
use LWP; use HTTP::Headers; use Crypt::SSLeay; $ENV{HTTPS_CERT_FILE} = '/c/stuff/client-certs/companyX/companyX.crt'; $ENV{HTTPS_KEY_FILE} = '/c/stuff/client-certs/companyX/companyX.key'; $ENV{PERL_LWP_USE_HTTP_10} = 1; $ua = LWP::UserAgent->new(); $ua->agent("FooBar-POST/0.0.7"); #attempt to replicate mangled header # content-type: text/xml # host: securepost.stage.mystical.org # proxy-authorization: Basic VVVpWWW2bDpXXX9YYYRlZZZ # content-length: 3593 $h = HTTP::Headers->new( ":content-type" => 'text/xml', ":host" => 'securepost.stage.mystical.org', ":proxy-authorization" => "Basic VVVpWWW2bDpXXX9YYYRlZZZ", ":content-length" => '3593' ); my $req = HTTP::Request->new(POST => 'https://securepost.stage.mystica +l.org/companyX/getPOST.asp',$h); $req->content(<<EOT); ### ### the data goes here ### deleted for clarity and brevity ### EOT my $res = $ua->request($req); if ($res->is_success) { print $res->content; } else { print $res->status_line, "\n"; }
|
|---|
| Replies are listed 'Best First'. | |
|---|---|
|
Re: Forcing HTTP/1.0 and mangling headers
by Anonymous Monk on Jan 16, 2007 at 15:11 UTC | |
by yam (Novice) on Jan 16, 2007 at 15:23 UTC | |
by jhourcle (Prior) on Jan 16, 2007 at 16:19 UTC | |
by yam (Novice) on Jan 16, 2007 at 16:59 UTC | |
|
Re: Forcing HTTP/1.0 and mangling headers
by shmem (Chancellor) on Jan 16, 2007 at 15:59 UTC | |
by yam (Novice) on Jan 16, 2007 at 17:03 UTC | |
by shmem (Chancellor) on Jan 16, 2007 at 17:42 UTC | |
|
Re: Forcing HTTP/1.0 and mangling headers
by tobyink (Canon) on Mar 05, 2012 at 22:40 UTC |