Hi Sawyer,
I've read email several times now, and I really can't see anything in it
that I don't like :)
I said which answer I'd really like, and you gave me one better than
that - not only is it fixed in Dancer 2, there's even a good chance that
the fix will be backported to Dancer 1. Not seeing a downside there at all.
None of the work I'm doing is time-critical, so as long as there's a fix
in the pipeline, I'm happy.
If there's anything I can do to help get the Dancer 1 patch out sooner,
then by all means let me know. I'm no expert in Dancer internals, but
I'm happy to help in any way I can.
Cheers,
Dave...
On 01/21/2012 03:52 PM, sawyer x wrote:
Hey Dave. :)
I will probably mention a few things here you might like and a few
things you might not.
In a nutshell: this is *fixed* in Dancer 2, and it might be fixed in
Dancer 1 at some point.
Not in a nutshell: this is actually a very known and painful issue. It's
indeed a Dancer problem, and not a Plack one. This is also one of the
reasons Dancer 2 got started in the first place. One of the design
decisions of Dancer 1 leads to this problem and Dancer 2 was created out
of proper design to set this (and other side-effects) straight. This
means that Dancer 2 will definitely not suffer this malady, from the get-go.
There is a proposed community patch for this (and a fork carrying this
patch) but we found it a bit tricky and somewhat unclean and haven't
merged it in yet. If you have the time (and inclination) contact me on
this issue. I'd love help with reviewing and cleaning it up so we could
actually merge this.
Sorry for this, and hope we could settle it as soon as possible.
S.
On Sat, Jan 21, 2012 at 5:32 PM, Dave Cross <[email protected]
<mailto:[email protected]>> wrote:
I was disappointed to read this in Dancer::Deployment:
NOTE: Only a single Dancer application can be deployed using the
Plack::Handler::Apache2 method. Multiple Dancer applications will not
work properly (The routes will be mixed-up between the applications).
I've just installed my second Dancer app on the same Apache instance
using Plack::Handler::Apache2 and I can confirm that the second app
gets _very_ confused :/
I assume this is a Dancer limitation rather than a Plack one as I
already also have a Catalyst app installed in the same Apache
instance using Plack::Handler::Apache2.
So I was wondering how serious this limitation is? Is it something
that could be fixed if a hacker (and, yes, I'm probably volunteering
myself here) spent some time investigating the problem? Or is it a
hard limit that we're never going to be able to work round?
Of course the answer I'd _really_ like is "oh, yes, that'll be fixed
in Dancer 2.0" :-)
Cheers,
Dave...
--
Dave Cross :: [email protected] <mailto:[email protected]>
http://dave.org.uk/
@davorg
_________________________________________________
Dancer-users mailing list
[email protected] <mailto:[email protected]>
http://www.backup-manager.org/__cgi-bin/listinfo/dancer-users
<http://www.backup-manager.org/cgi-bin/listinfo/dancer-users>
--
Dave Cross :: [email protected]
http://dave.org.uk/
@davorg
_______________________________________________
Dancer-users mailing list
[email protected]
http://www.backup-manager.org/cgi-bin/listinfo/dancer-users