Re: ap_get_server_port differences

2005-11-14 Thread Jim Jagielski
Ruediger Pluem wrote: > > On 11/14/2005 10:07 PM, Jim Jagielski wrote: > > Jim Jagielski wrote: > > > >>Yes, the _UNSET usage is very inconsistent, but it's something > >>that's been inherited from older versions :) > > So fixing it, breaks backward compatibility ;-) > Well, we are going from

Re: ap_get_server_port differences

2005-11-14 Thread Ruediger Pluem
On 11/14/2005 10:07 PM, Jim Jagielski wrote: > Jim Jagielski wrote: > >>Yes, the _UNSET usage is very inconsistent, but it's something >>that's been inherited from older versions :) So fixing it, breaks backward compatibility ;-) >> >>I think that either: >> >>1. _UNSET should be removed >

Re: ap_get_server_port differences

2005-11-14 Thread Jim Jagielski
Jim Jagielski wrote: > > Yes, the _UNSET usage is very inconsistent, but it's something > that's been inherited from older versions :) > > I think that either: > > 1. _UNSET should be removed > or > 2. _UNSET should be considered _OFF > > My pers. pref is #1... Again, it's t

Re: ap_get_server_port differences

2005-11-14 Thread Jim Jagielski
On Nov 14, 2005, at 3:51 PM, Ruediger Pluem wrote: On 11/14/2005 03:49 PM, Jim Jagielski wrote: I am thinking about something like the below: As far as I understand the patch the default value will be UseCanonicalPhysicalPort off which is the 2.0 behaviour, correct? If yes, +1 from my

Re: ap_get_server_port differences

2005-11-14 Thread Ruediger Pluem
On 11/14/2005 03:49 PM, Jim Jagielski wrote: > I am thinking about something like the below: As far as I understand the patch the default value will be UseCanonicalPhysicalPort off which is the 2.0 behaviour, correct? If yes, +1 from my side. This leaves only open the inconsistency between ap

Re: ap_get_server_port differences

2005-11-14 Thread Jim Jagielski
I am thinking about something like the below: Index: server/core.c === --- server/core.c (revision 344120) +++ server/core.c (working copy) @@ -115,6 +115,7 @@ conf->accept_path_info = 3; conf->use_canonical_nam

Re: ap_get_server_port differences

2005-11-14 Thread Brian Akins
Sander Temme wrote: On Nov 11, 2005, at 12:33 PM, Brian Akins wrote: in 2.1.9 ap_get_server_port uses the following: port = r->parsed_uri.port_str ? r->parsed_uri.port : r->connection->local_addr->port ? r->connection- >local_addr->port : r->server->port ? r->

Re: ap_get_server_port differences

2005-11-13 Thread Ruediger Pluem
On 11/13/2005 05:16 PM, Jim Jagielski wrote: > > This is something that is documented as a change in 2.0 -> 2.1/2.2 > as well as something that was discussed when the change was > made. I suggested, iirc, some other UCN option which emulated > the 2.0 behavior... This is, I believe, still the r

Re: ap_get_server_port differences

2005-11-13 Thread Jim Jagielski
Sander Temme wrote: > > In the other case, UseCanonicalName On which was default in the 2.0 > branch, we look first in the server_rec and then in the connection. > You could possibly make a case for having the search order in the top > of the if clause match the one in the bottom. This again

Re: ap_get_server_port differences

2005-11-13 Thread Jim Jagielski
Ruediger Pluem wrote: > >if (d->use_canonical_name == USE_CANONICAL_NAME_OFF > || d->use_canonical_name == USE_CANONICAL_NAME_DNS) { > > /* With UseCanonicalName off Apache will form self-referential > * URLs using the hostname and port supplied by the client if >

Re: ap_get_server_port differences

2005-11-13 Thread Ruediger Pluem
On 11/12/2005 06:46 PM, Sander Temme wrote: [..cut..] > > What bit you is probably the change in default value for > UseCanonicalName? Used to be On by default, now it's Off. Try flicking > that switch. I saw that it is now set to off in the default configuration files, but the behaviour of

Re: ap_get_server_port differences

2005-11-12 Thread Sander Temme
On Nov 11, 2005, at 12:33 PM, Brian Akins wrote: in 2.1.9 ap_get_server_port uses the following: port = r->parsed_uri.port_str ? r->parsed_uri.port : r->connection->local_addr->port ? r->connection- >local_addr->port : r->server->port ? r->server->port :

Re: ap_get_server_port differences

2005-11-12 Thread Paul Querna
Sander Temme wrote: .. > Did you force it to 'Off' in 2.0? If you have it 'Off', you believe what > the client sends you, or what you explicitly stashed in the server > record by specifying it in the ServerName directive. In the newer > server, we added the actual port the request came in on, w

Re: ap_get_server_port differences

2005-11-12 Thread Sander Temme
On Nov 11, 2005, at 12:33 PM, Brian Akins wrote: in 2.1.9 ap_get_server_port uses the following: port = r->parsed_uri.port_str ? r->parsed_uri.port : r->connection->local_addr->port ? r->connection- >local_addr->port : r->server->port ? r->server->port :

Re: ap_get_server_port differences

2005-11-11 Thread Brian Akins
Here's the quick workaround module. Just load the modules and it tries to "revert" to 2.0 behavior #include "httpd.h" #include "http_config.h" #include "ap_config.h" #include "apr_strings.h" #include "http_protocol.h" module AP_MODULE_DECLARE_DATA port_module; typedef struct { apr_port_t

ap_get_server_port differences

2005-11-11 Thread Brian Akins
in 2.1.9 ap_get_server_port uses the following: port = r->parsed_uri.port_str ? r->parsed_uri.port : r->connection->local_addr->port ? r->connection->local_addr->port : r->server->port ? r->server->port : ap_default_port(r); in 2.0.55 it uses: p