12 years agoAdd a test to make sure that a die in a Controller's end sub is propogated as a 500...
Dave Rolsky [Tue, 3 May 2011 16:28:42 +0000]
Add a test to make sure that a die in a Controller's end sub is propogated as a 500 error

13 years agoUnbreak the pod-coverage tests
Florian Ragwitz [Thu, 14 Apr 2011 12:09:13 +0000]
Unbreak the pod-coverage tests

Without removing those imports, it expects us to document them.

13 years agoYes, it is
Florian Ragwitz [Thu, 14 Apr 2011 12:05:49 +0000]
Yes, it is

13 years agoRemove special-casing of '*' for get_action_methods
Florian Ragwitz [Thu, 14 Apr 2011 11:56:17 +0000]
Remove special-casing of '*' for get_action_methods

Instead, pull '*' out of _controller_actions early on. The part where they're
applied to actions is still gross.

13 years agoPull a bunch of gross hacks to get actions => { '*' => ... } working again
Florian Ragwitz [Thu, 14 Apr 2011 11:51:09 +0000]
Pull a bunch of gross hacks to get actions => { '*' => ... } working again

13 years agoAdd a test for configuring attributes for all actions
Florian Ragwitz [Thu, 14 Apr 2011 11:50:50 +0000]
Add a test for configuring attributes for all actions

13 years agoAdd a test view for dumping actions
Florian Ragwitz [Thu, 14 Apr 2011 11:50:06 +0000]
Add a test view for dumping actions

13 years agoMake this error message more readable
Florian Ragwitz [Thu, 14 Apr 2011 11:35:39 +0000]
Make this error message more readable

13 years agoMake this error message more readable
Florian Ragwitz [Thu, 14 Apr 2011 11:06:55 +0000]
Make this error message more readable

13 years agoAgreed
Florian Ragwitz [Thu, 14 Apr 2011 11:01:40 +0000]
Agreed

13 years agoMerge branch 'master' into psgi
Florian Ragwitz [Tue, 12 Apr 2011 00:08:23 +0000]
Merge branch 'master' into psgi

* master:
  A failing reverse lookup doesn't imply it's a local address
  Tiny tidyups to copyright and license

Conflicts:
lib/Catalyst/Engine/HTTP.pm

13 years agoA failing reverse lookup doesn't imply it's a local address
Florian Ragwitz [Tue, 12 Apr 2011 00:05:59 +0000]
A failing reverse lookup doesn't imply it's a local address

13 years agoFix a syntax error introduced in d367082
Florian Ragwitz [Thu, 31 Mar 2011 12:48:36 +0000]
Fix a syntax error introduced in d367082

13 years agoUse the Plack LighttpdScriptNameFix middleware.
Tomas Doran [Tue, 29 Mar 2011 15:05:02 +0000]
Use the Plack LighttpdScriptNameFix middleware.

This replaces our home-baked solution with something which works better
and in more cases

13 years agoUpdate TODO
Tomas Doran [Sun, 27 Mar 2011 17:58:13 +0000]
Update TODO

13 years agoClean up FIXMEs
Tomas Doran [Sun, 27 Mar 2011 17:50:46 +0000]
Clean up FIXMEs

13 years agoFix a warning under aggregation
Florian Ragwitz [Sun, 27 Mar 2011 17:36:28 +0000]
Fix a warning under aggregation

13 years agoPass script options into auto loader
Tomas Doran (t0m) [Sun, 27 Mar 2011 16:21:37 +0000]
Pass script options into auto loader

Plack::Handler::XXX therefore gets the options even when auto-detecting.

13 years agoRetrieve args for old scripts so FCGI --listen works
Tomas Doran [Sun, 27 Mar 2011 16:19:38 +0000]
Retrieve args for old scripts so FCGI --listen works

13 years agoForce FCGI detection for FastCGI scripts.
Tomas Doran [Sun, 27 Mar 2011 15:54:11 +0000]
Force FCGI detection for FastCGI scripts.

13 years agoDo not commit conflicts, idiot
Tomas Doran [Sun, 27 Mar 2011 15:28:25 +0000]
Do not commit conflicts, idiot

13 years agoFix bug stopping fastcgi script working
Tomas Doran [Sun, 27 Mar 2011 15:20:50 +0000]
Fix bug stopping fastcgi script working

13 years agoUpdate todo
Tomas Doran [Sun, 27 Mar 2011 14:41:40 +0000]
Update todo

13 years agoUpdate todo
Tomas Doran [Sun, 27 Mar 2011 14:41:31 +0000]
Update todo

13 years agoFix the breakage the last commit introduced
Florian Ragwitz [Sun, 27 Mar 2011 15:06:51 +0000]
Fix the breakage the last commit introduced

13 years agomake ->setup_engine($engine) supported API again
Florian Ragwitz [Sun, 27 Mar 2011 14:33:30 +0000]
make ->setup_engine($engine) supported API again

It should always take preference over CATALYST_ENGINE. The compat hack is really
just PSGI specific.

13 years agoNo, we don't want this
Florian Ragwitz [Sun, 27 Mar 2011 13:31:13 +0000]
No, we don't want this

If we'd check it, we'd have to check it on every request. That's entirely not
worth it as I'm not even aware of any server that doesn't support psgi.streaming
right now, let alone a server people actually use.

If people complain, documentation suggesting to load the BufferedWrite
middleware needs to be written.

13 years agoDocument middlewares applied automatically
Tomas Doran [Sun, 27 Mar 2011 13:33:36 +0000]
Document middlewares applied automatically

13 years agoExplicitly document the wrong Engine::PSGI thing.
Tomas Doran [Sun, 27 Mar 2011 12:55:20 +0000]
Explicitly document the wrong Engine::PSGI thing.

I did it accidentally, and it took rafl half an hour to chase down in d5024dd882,
so someone as dumb as me is likely to do it (just substituting the method name,
rather than removing the additional closure) when upgrading from ::Engine::PSGI.

13 years agoUpdate TODO
Florian Ragwitz [Sun, 27 Mar 2011 13:23:06 +0000]
Update TODO

13 years agoPartially unify local and remote request codepaths
Florian Ragwitz [Thu, 3 Mar 2011 14:21:53 +0000]
Partially unify local and remote request codepaths

13 years agoUse Plack::Test::ExternalServer in Catalyst::Test remote requests
Florian Ragwitz [Wed, 2 Mar 2011 12:01:46 +0000]
Use Plack::Test::ExternalServer in Catalyst::Test remote requests

13 years agoWe already have a warning for this
Florian Ragwitz [Sun, 27 Mar 2011 12:42:01 +0000]
We already have a warning for this

13 years agoFix the test psgi file
Florian Ragwitz [Sun, 27 Mar 2011 12:39:22 +0000]
Fix the test psgi file

13 years agopsgi_app is just a getter
Florian Ragwitz [Sun, 27 Mar 2011 12:36:53 +0000]
psgi_app is just a getter

13 years agoDon't assume the app gets loaded before the first request
Florian Ragwitz [Sun, 27 Mar 2011 12:36:33 +0000]
Don't assume the app gets loaded before the first request

13 years agoFirst stab documenting writing a .psgi file
Tomas Doran [Sun, 27 Mar 2011 12:22:59 +0000]
First stab documenting writing a .psgi file

13 years agoLegacy .psgi files are never loaded
Florian Ragwitz [Sun, 27 Mar 2011 12:10:29 +0000]
Legacy .psgi files are never loaded

13 years agoIgnore old-style .psgi files
Florian Ragwitz [Sun, 27 Mar 2011 11:58:24 +0000]
Ignore old-style .psgi files

13 years agoDe-uglify a comment
Florian Ragwitz [Sun, 27 Mar 2011 10:55:37 +0000]
De-uglify a comment

13 years agoDocument status of the misc engines.
Tomas Doran [Sun, 27 Mar 2011 11:44:50 +0000]
Document status of the misc engines.

Both SCGI and XMPP appear to be bork

13 years agoAdd a test showing how having an Engine::PSGI .psgi file breaks.
Tomas Doran [Sun, 27 Mar 2011 11:19:31 +0000]
Add a test showing how having an Engine::PSGI .psgi file breaks.

Notably, this .psgi file is in the app root, whereas ::Engine::PSGI's docs and helper suggest that
you create it in script/. So I don't have a problem with not supporting the .psgi working for tests
until you've converted it if you move it into the app root really - however it would be nice to be
able to detect it was broken, rather than just hanging.

13 years agoTest showing how having a .psgi file will ruin your day
Tomas Doran [Sun, 27 Mar 2011 11:17:47 +0000]
Test showing how having a .psgi file will ruin your day

13 years agolocal_request is now tested
Florian Ragwitz [Sun, 27 Mar 2011 10:44:34 +0000]
local_request is now tested

(and deprecated!)

13 years agoAdd tests for the deprecated local_request
Florian Ragwitz [Sun, 27 Mar 2011 10:40:00 +0000]
Add tests for the deprecated local_request

13 years agostrictures and warnings for live_catalyst_test.t
Florian Ragwitz [Sun, 27 Mar 2011 10:35:27 +0000]
strictures and warnings for live_catalyst_test.t

13 years agoThis was fixed already
Tomas Doran [Sun, 27 Mar 2011 10:41:59 +0000]
This was fixed already

13 years agoFix the conditional here.
Tomas Doran [Sun, 27 Mar 2011 10:38:17 +0000]
Fix the conditional here.

We do always need a trailing slash if we have a server path and either
just a /, or no uri at all (e.g. http://localhost). This matters when
we're mounted at a non-root path, as cataltst needs to see
http://localhost/non_root_path/, rather than http://localhost/non_root_path,
otherwise we'll try to resolve /non_root path as an action.

13 years agoTrim hatespace
Tomas Doran [Sun, 27 Mar 2011 09:59:17 +0000]
Trim hatespace

13 years agoTiny tidyups to copyright and license
Tomas Doran [Sun, 27 Mar 2011 07:57:54 +0000]
Tiny tidyups to copyright and license

13 years agoThe IIS 6 script name fix is now part of Plack
Florian Ragwitz [Sat, 26 Mar 2011 13:59:00 +0000]
The IIS 6 script name fix is now part of Plack

Depend on it and use it.

13 years agoMerge branch 'master' into psgi
Florian Ragwitz [Sat, 26 Mar 2011 14:00:18 +0000]
Merge branch 'master' into psgi

* master:
  Fix Path actions debug screen to display number of arguments

13 years agoFix Path actions debug screen to display number of arguments
Dimitar Petrov [Mon, 21 Mar 2011 20:16:42 +0000]
Fix Path actions debug screen to display number of arguments

13 years agoMerge branch 'master' into psgi
Florian Ragwitz [Fri, 18 Mar 2011 14:53:07 +0000]
Merge branch 'master' into psgi

* master:
  Make default test names reflect reality more and fix a buglet, rafl++
  Give action_ok, action_redirect and action_notfound default test names

13 years agoMake default test names reflect reality more and fix a buglet, rafl++
Jonathan "Duke" Leto [Sat, 12 Mar 2011 01:05:27 +0000]
Make default test names reflect reality more and fix a buglet, rafl++

13 years agoGive action_ok, action_redirect and action_notfound default test names
Jonathan "Duke" Leto [Sat, 12 Mar 2011 00:20:51 +0000]
Give action_ok, action_redirect and action_notfound default test names

13 years agoMerge branch 'master' into psgi
Florian Ragwitz [Wed, 9 Mar 2011 21:11:42 +0000]
Merge branch 'master' into psgi

* master:
  Avoid super long command lines when not aggregating
  Improve the documentation for various functions in Catalyst::Test

13 years agoAvoid super long command lines when not aggregating
Florian Ragwitz [Tue, 8 Mar 2011 21:52:15 +0000]
Avoid super long command lines when not aggregating

This probably also fixes prove -lr t/, which people have been complaining about
every now and then.

13 years agoImprove the documentation for various functions in Catalyst::Test
Jonathan "Duke" Leto [Mon, 7 Mar 2011 15:20:13 +0000]
Improve the documentation for various functions in Catalyst::Test

13 years agoDeprecate local_request and remote_request
Florian Ragwitz [Thu, 3 Mar 2011 14:03:17 +0000]
Deprecate local_request and remote_request

13 years agoMove out building of the horrible ctx_request closure
Florian Ragwitz [Thu, 3 Mar 2011 13:43:25 +0000]
Move out building of the horrible ctx_request closure

Also move out the rest, while we're at it. This is intended to aid a subsequent
refactoring.

13 years agoRemove some kane-isms that hurt my eyes
Florian Ragwitz [Thu, 3 Mar 2011 13:22:01 +0000]
Remove some kane-isms that hurt my eyes

13 years agoRun http-server.t tests on something other than the server root
Florian Ragwitz [Wed, 2 Mar 2011 13:11:39 +0000]
Run http-server.t tests on something other than the server root

13 years agoRemove testsuite's assumptions about where it's mounted on a web server
Florian Ragwitz [Wed, 2 Mar 2011 13:11:05 +0000]
Remove testsuite's assumptions about where it's mounted on a web server

13 years agoMake http-server.t use Test::TCP
Florian Ragwitz [Wed, 2 Mar 2011 12:37:18 +0000]
Make http-server.t use Test::TCP

No dep, as it's author-only and pulled in by Plack already anyway.

13 years agoDon't use the server.pl script in the live tests
Florian Ragwitz [Wed, 2 Mar 2011 12:27:37 +0000]
Don't use the server.pl script in the live tests

Instead, just spin up a server with Plack::Loader. The script classes are
already tested separately, so we only need this test to make sure live tests
work.

This is intended to aid a subsequent refactoring.

13 years ago.psgi stuff is now sorted out
Florian Ragwitz [Wed, 2 Mar 2011 11:46:41 +0000]
.psgi stuff is now sorted out

13 years ago::Wx is declared dead
Florian Ragwitz [Wed, 2 Mar 2011 11:46:29 +0000]
::Wx is declared dead

13 years agolighttpd and iis6 now work
Florian Ragwitz [Wed, 2 Mar 2011 11:46:17 +0000]
lighttpd and iis6 now work

13 years agoReduce code repetition
Florian Ragwitz [Wed, 2 Mar 2011 11:38:47 +0000]
Reduce code repetition

13 years agoAdd legacy workaround for nginx PATH_INFO
Florian Ragwitz [Wed, 2 Mar 2011 11:32:22 +0000]
Add legacy workaround for nginx PATH_INFO

13 years agoMake legacy IIS env fixing work again
Florian Ragwitz [Wed, 2 Mar 2011 11:19:12 +0000]
Make legacy IIS env fixing work again

13 years agoMake legacy lighttpd env fixing work again
Florian Ragwitz [Wed, 2 Mar 2011 11:13:41 +0000]
Make legacy lighttpd env fixing work again

13 years agoLoad middlewares where they're actually used
Florian Ragwitz [Wed, 2 Mar 2011 11:13:24 +0000]
Load middlewares where they're actually used

13 years agoMake _wrapped_legacy_psgi_app do wrapping only
Florian Ragwitz [Wed, 2 Mar 2011 10:57:39 +0000]
Make _wrapped_legacy_psgi_app do wrapping only

We now pass in the psgi app to wrap, which should make it quite a bit easier to
test. We'll need that to check if the various quirk fixes for different web
servers actually work.

13 years agoVersion 5.89002 5.89002
Florian Ragwitz [Wed, 2 Mar 2011 10:30:51 +0000]
Version 5.89002

13 years agoThe HTTP engine is no more, and was also never required for this test
Florian Ragwitz [Wed, 2 Mar 2011 10:02:51 +0000]
The HTTP engine is no more, and was also never required for this test

13 years agoAvoid loading actual engines in script tests
Florian Ragwitz [Wed, 2 Mar 2011 09:59:19 +0000]
Avoid loading actual engines in script tests

This way we can run these tests without having FCGI.pm installed.

13 years agoPSGI, not Plack
Florian Ragwitz [Tue, 1 Mar 2011 18:37:44 +0000]
PSGI, not Plack

13 years agoUpdate Upgrading docs
Florian Ragwitz [Tue, 1 Mar 2011 18:37:23 +0000]
Update Upgrading docs

13 years agoMake getting to a PSGI app saner
Florian Ragwitz [Tue, 1 Mar 2011 18:30:31 +0000]
Make getting to a PSGI app saner

All the user should ever need to call is ->psgi_app (previously raw_psgi_app),
to get an unwrapped psgi app as provided by the engine. This is what should be
used when writing .psgi files.

For all other ways of running catalyst apps there's no need to be able to get to
the psgi application that'll actually run, so here we're making everything
that's responsible for reading .psgi files to run them with Catalyst::Script::*
or for wrapping the plain psgi code ref with back-compat middleware private.

Unfortunately this removes some of the docs about the automatic .psgi
loading. These should be added again elsewhere. Also the Upgrading docs need
updating after this.

13 years agoVersion 5.89001 5.89001
Florian Ragwitz [Tue, 1 Mar 2011 14:26:46 +0000]
Version 5.89001

13 years agoStop psgi_file.t from recursing
Florian Ragwitz [Tue, 1 Mar 2011 14:38:40 +0000]
Stop psgi_file.t from recursing

psgi_app is *not* to be used in .psgi files. It's the compat layer that'll give
you either an app loaded from from a .psgi file, or the default PSGI app with
the reverse proxy middleware and stuff.

In .psgi files only raw_psgi_app should ever be used, as that'll return the bare
catalyst psgi app to then be wrapped in middlewares manually.

I see how the naming is shit and confusing though. Suggestions welcome.

13 years agoRename Engine::Loader to EngineLoader
Florian Ragwitz [Tue, 1 Mar 2011 14:22:10 +0000]
Rename Engine::Loader to EngineLoader

It's not an engine, as those exist no more, so move it out of the ::Engine::
namespace.

13 years agoChangelog updates
Florian Ragwitz [Tue, 1 Mar 2011 14:20:08 +0000]
Changelog updates

13 years agoMerge branch 'master' into psgi
Florian Ragwitz [Tue, 1 Mar 2011 13:49:57 +0000]
Merge branch 'master' into psgi

* master: (22 commits)
  Change repos metadata to git
  Version 5.80032
  Pass the extra restart options
  Fix test
  Changelog
  added myself (dd070) in contributors list
  removed duplicate parameters in return value of _restarter_args
  added more parameters in return value of _restarter_args
  Add a few tests
  Patch to make restarter class configurable / settable in prefs
  Apply more correct fix than 8df53b (I hope)
  Rename test
  Make tests more clear
  Merge revert 8df53bed
  Un-TODO abraxxa's tests
  This scares me, but it fixes stuf work work
  Fix 5.80 bug which causes slurp to fail if called multiple times
  Fix body predicate bug/back compat issue
  Version 5.80031
  added test for chained dispatcher fail on multiple CaptureArgs(0) parts introduced in 5.80030
  ...

Conflicts:
Changes
Makefile.PL
lib/Catalyst.pm
lib/Catalyst/Engine/FastCGI.pm
lib/Catalyst/Runtime.pm

13 years agoProperly skip known broken tests
Florian Ragwitz [Tue, 1 Mar 2011 13:43:05 +0000]
Properly skip known broken tests

Faked passing tests are not cool, and nor are __END__ sections, as those will
break aggregation.

If we actually want to prevent the body of the tests from being compiled,
putting the C<< plan skip_all => ... >> in a BEGIN block is the right thing to
do.

13 years agoChange repos metadata to git
Tomas Doran [Mon, 28 Feb 2011 19:41:54 +0000]
Change repos metadata to git

13 years agoVersion 5.80032 5.80032
Tomas Doran [Wed, 23 Feb 2011 01:10:30 +0000]
Version 5.80032

13 years agoPass the extra restart options
Tomas Doran [Wed, 23 Feb 2011 00:47:16 +0000]
Pass the extra restart options

13 years agoFix test
Tomas Doran [Wed, 23 Feb 2011 00:46:13 +0000]
Fix test

13 years agoChangelog
Tomas Doran [Tue, 22 Feb 2011 22:19:16 +0000]
Changelog

13 years agoMerge branch 'restarter_changeclass' into 5.80032_dev
Tomas Doran [Tue, 22 Feb 2011 22:10:31 +0000]
Merge branch 'restarter_changeclass' into 5.80032_dev

* restarter_changeclass:
  added myself (dd070) in contributors list
  removed duplicate parameters in return value of _restarter_args
  added more parameters in return value of _restarter_args
  Add a few tests
  Patch to make restarter class configurable / settable in prefs

13 years agoadded myself (dd070) in contributors list
Dhaval Dhanani [Wed, 16 Feb 2011 16:04:55 +0000]
added myself (dd070) in contributors list

13 years agoremoved duplicate parameters in return value of _restarter_args
Dhaval Dhanani [Wed, 16 Feb 2011 15:40:29 +0000]
removed duplicate parameters in return value of _restarter_args

13 years agoadded more parameters in return value of _restarter_args
Dhaval Dhanani [Wed, 16 Feb 2011 15:35:10 +0000]
added more parameters in return value of _restarter_args

13 years agoAdd a few tests
Tomas Doran [Mon, 14 Feb 2011 21:12:25 +0000]
Add a few tests

13 years agoPatch to make restarter class configurable / settable in prefs
Tomas Doran [Mon, 14 Feb 2011 21:05:30 +0000]
Patch to make restarter class configurable / settable in prefs

13 years agoMerge branch 'chained_fix' into 5.80032_dev
Tomas Doran [Tue, 22 Feb 2011 22:09:40 +0000]
Merge branch 'chained_fix' into 5.80032_dev

* chained_fix:
  Apply more correct fix than 8df53b (I hope)
  Rename test
  Make tests more clear
  Merge revert 8df53bed
  Un-TODO abraxxa's tests

13 years agoApply more correct fix than 8df53b (I hope)
Tomas Doran [Tue, 22 Feb 2011 01:31:11 +0000]
Apply more correct fix than 8df53b (I hope)