[Catalyst] Job posting!
Hey Catalyst / DBIC folks, I am leaving my job and, thus, my position is becoming available! The title is QA Tools Developer - you would be working on custom internal apps at Riverbed Technology, a network acceleration company in downtown San Francisco, CA. Good company, good people, good location (IMHO). A lot of the work is full-stack OO Perl web development using Catalyst and DBIC. Experience with the following will give you a leg up (ordered from most to least desired): - solid OO Perl background - Catalyst / DBIx::Class - Javascript - managing web servers and MySQL databases - Bugzilla maintenance and customization They'll reimburse up to $5k for relocation. If you're interested please send an email to Stan Russell at stan.russ...@riverbed.com, with your resume attached. Good luck! Byron ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] ldap authentication/db authorization
Steve Rippl wrote on 2009-08-09: Hello! I have a question that is hopefully trivial for someone here, but I haven't seen any examples. I have an app I'm building, used by folks in an Active Directory network. It's got a DBIx model and I'd like to use that for role based authorization and getting staff members grade/building info etc (it's a school district), but I'd like to (just) authenticate against ldap. I don't want to store any user/role data there (obviously their username is in there and it matches what I have in the db), simply try to bind against it as the user, then get store data from the database. This way they can use their network passwords and I don't have to fill AD with application specific info. Make sense? Easy to do? So far I'm doing it all via the db... name WsdSis authentication default_realm dbic realms dbic credential class Password password_type clear /credential store user_class DB::Staff role_relation roles role_field role /store /dbic /realms /authentication ... and ... use Catalyst qw/ -Debug ConfigLoader Static::Simple StackTrace Authentication Authorization::Roles Session Session::Store::FastMmap Session::State::Cookie /; ... What would these look like if I'm able to squeeze the ldap plugin in there? Can I have 2 Authentication::Store backends in there? TIA, Steve -- Steve Rippl Technology Director Woodland School District 360 225 9451 x326 Hey Steve, I do this in my app. The way I do is it I have two authentication realms - LDAP and DBIC. I authenticate against the LDAP realm first, and if that is successful I do a find_or_create with the user info on my user DB table, so any first-time users get a user created in the DBIC realm with default roles and whatnot. Then I authenticate against the DBIC realm. I do that last because $c-user will contain the most recently authenticated user, so in this case $c-user would be your DB::Staff object. HTH Byron ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] ldap authentication/db authorization
Steve Rippl wrote on 2009-08-10:
Byron Young wrote:
Hey Steve,
I do this in my app. The way I do is it I have two
authentication realms - LDAP and DBIC. I authenticate against the
LDAP realm first, and if that is successful I do a find_or_create
with the user info on my user DB table, so any first-time users get
a user created in the DBIC realm with default roles and whatnot.
Then I authenticate against the DBIC realm. I do that last because
$c-user will contain the most recently authenticated user, so in
this case $c-user would be your DB::Staff object.
HTH
Byron
Thanks for the response. Makes perfect sense, but I'm having a problem
with the call to $c-authenticate when I add the realm.
$c-authenticate({ username = $username, password = $password },
'ldap') throws an error about not using a string as HASH ref (I've got
my ldap settings under realms - ldap). Using $c-authenticate({
username = $username, password = $password, realm = $realm }) seems
to work, except it's not authenticating correctly and seems to use dbic
by default when I don't have a default realm set (I set $realm = 'ldap'
before the auth call).
What syntax are using?
Many thanks,
Steve
Same as you: $c-authenticate({id = $username,
password = $password},
'ldap')
Not sure why that doesn't work for you. What's in your ldap realm config?
Maybe you have a problem with the way the config is being parsed.
Byron
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] RE: uri_for() corrupts query parameters hash for caller
Tomas Doran wrote on 2009-06-29: On 26 Jun 2009, at 23:19, Byron Young wrote: Alrighty, here you go, patch + test are attached. There are based off the 5.71001 svn head because that's what I have currently. 5.8's uri_for() looks the same, so it should apply there as well, but let me know if you need another one from 5.8. http://dev.catalyst.perl.org/svnweb/Catalyst/revision/?rev=10736 Thanks very much for the patch, applied ok to 5.80 trunk. I rewrote your fix to just not mangle $_, which fixes the same issue with less code, and avoids the unsafe each.. Unfortunately, this just missed the Catalyst 5.80006 release, so you'll have to wait for the next one to see it in released code, sorry! Cheers t0m Oh, nice, that's a much better solution. If you don't mind, though, can you explain what you mean about the 'unsafe each'? Thanks! Byron ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] RE: uri_for() corrupts query parameters hash for caller
Tomas Doran wrote on 2009-06-29: On 30 Jun 2009, at 00:31, Byron Young wrote: If you don't mind, though, can you explain what you mean about the 'unsafe each'? If your application code half iterates through the params hash with each before calling uri_for, then the param copy would only copy the second half of the hash (as each has an internal iterator). Cheers t0m Ah, makes sense. Learn something new every day! Thanks! Byron ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] RE: uri_for() corrupts query parameters hash for caller
Tomas Doran wrote on 2009-06-26: Byron Young wrote: I know people have been busy (I think there were some perl conferences lately?) and I think my issue slipped through the cracks. Just wanted to know what people thought about this and whether I should submit my patch or take a different approach. Sorry for dropping this on the floor. Yes, please submit a patch :) Cheers t0m Alrighty, here you go, patch + test are attached. There are based off the 5.71001 svn head because that's what I have currently. 5.8's uri_for() looks the same, so it should apply there as well, but let me know if you need another one from 5.8. (I'm not sure how tracking contributors works, or if you do, but if so this was worked on by myself and Amir Sadoughi). Byron uri_for_patch.diff Description: uri_for_patch.diff ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
[Catalyst] RE: uri_for() corrupts query parameters hash for caller
Byron Young wrote on 2009-06-12:
Byron Young wrote on 2009-06-12:
Hey everybody,
I ran into an issue at $work where we keep passing the same
$query_params hashref to a number of uri_for() calls successively, but
if there are characters in the query params that need to be escaped
they get escaped each time, leading to sequences like
?filter=Not%25252BRun
after the same $query_params have been run through uri_for a few of
times (because the '%' keeps getting escaped). The query hash was
originally { filter = 'Not Run' }.
So, we patched uri_for() here at work to create a copy of $params
and work with that, and that fixes the issue. However, it seems
like such a simple fix that I feel like it must have been thought
of and discussed and shot down in the past, but I didn't find
anything in the list archives indicating that. Is there some
reason uri_for() does things that way?
If not I'll gladly supply patch + test.
Thanks,
Byron
I also noticed that the docs for uri_for used to warn of the
destructiveness but that warning has been removed in more recent
versions. I'd like to suggest that it be added back and made more
prominent if there really is a good reason for mangling the
caller's data. I can provide a doc patch in that case, too.
Byron
Hey,
I know people have been busy (I think there were some perl conferences lately?)
and I think my issue slipped through the cracks. Just wanted to know what
people thought about this and whether I should submit my patch or take a
different approach.
Thanks
Byron
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
[Catalyst] uri_for() corrupts query parameters hash for caller
Hey everybody,
I ran into an issue at $work where we keep passing the same $query_params
hashref to a number of uri_for() calls successively, but if there are
characters in the query params that need to be escaped they get escaped each
time, leading to sequences like
?filter=Not%25252BRun
after the same $query_params have been run through uri_for a few of times
(because the '%' keeps getting escaped). The query hash was originally {
filter = 'Not Run' }.
So, we patched uri_for() here at work to create a copy of $params and work with
that, and that fixes the issue. However, it seems like such a simple fix that
I feel like it must have been thought of and discussed and shot down in the
past, but I didn't find anything in the list archives indicating that. Is
there some reason uri_for() does things that way?
If not I'll gladly supply patch + test.
Thanks,
Byron
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
[Catalyst] RE: uri_for() corrupts query parameters hash for caller
Byron Young wrote on 2009-06-12:
Hey everybody,
I ran into an issue at $work where we keep passing the same
$query_params hashref to a number of uri_for() calls successively,
but if there are characters in the query params that need to be
escaped they get escaped each time, leading to sequences like
?filter=Not%25252BRun
after the same $query_params have been run through uri_for a few of
times (because the '%' keeps getting escaped). The query hash was
originally { filter = 'Not Run' }.
So, we patched uri_for() here at work to create a copy of $params
and work with that, and that fixes the issue. However, it seems
like such a simple fix that I feel like it must have been thought
of and discussed and shot down in the past, but I didn't find
anything in the list archives indicating that. Is there some
reason uri_for() does things that way?
If not I'll gladly supply patch + test.
Thanks,
Byron
___ List:
Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-
bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-
archive.com/catalyst@lists.scsys.co.uk/ Dev site:
http://dev.catalyst.perl.org/
I also noticed that the docs for uri_for used to warn of the destructiveness
but that warning has been removed in more recent versions. I'd like to suggest
that it be added back and made more prominent if there really is a good reason
for mangling the caller's data. I can provide a doc patch in that case, too.
Byron
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Passing connection info to an custom model/external lib
Eric Wright wrote on 2009-04-10:
I have what I hope is not a dumb question. I'm trying to follow best
practices here and am writing a piece of business logic that I feel
should be Catalyst agnostic and therefore am creating a custom model.
(However, at the moment Catalyst is the primary consumer of this lib.)
I'd like to use my DBIC::Schema goodness for DB access and happily have
abstracted that via Catalyst::Model::DBIC::Schema. The thing is I
realized that if I connect to this schema outside of the scope of
Catalyst via DBIx::Class::Schema-connect I have to pass in the
connection info separately which really bugs me because I've already
defined it via a configuration file for the Catalyst app (using
Config::General). I suppose I could just pass that info into the model
but that seems messy and redundant to me. If I subclass Catalyst::Model
well then I'm tightly coupled with Catalyst again. I'd like to just be
able to glue the model to Catalyst and do a:
$c-model('MyCustomModel')-foo(...)
and it just work.
It seems to me there should be a simple way for this schema to know its
connection info based on context and I figured there's like a slick
Catalyst ninja kind of method for doing this that my brain is just not
grasping at the moment. How do you do this sort of thing? Is the magic
in the Catalyst glue?
Thanks in advance for any advice/thoughts/comments...
-Eric
I'm not sure if there's a special Catalyst way of doing this, but I use my
schema classes from my various places, so I moved my db connection data into a
separate config file that all apps can access. I have a separate class I wrote
that looks for the config file in a few specific places (depending on whether
you're in dev mode or production mode) and reads it. This class provides the
connection info to the schema class. This way I can easily move my database,
update the info in one file, and all apps automatically get the new connection
info.
Hope that helps.
Byron
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Custom changes to different instances of the same app
That's a pretty broad question, but three basic ideas come to mind. First start by factoring as much common code as possible out into separate classes. If they are asking for major differences in their custom versions, using a version control system and branching for their changes might be a good idea, but it brings with it the overhead of maintaining branches and backporting fixes as needed. If the customizations are not so vast, you can use a Strategy pattern to codify the customizable behaviors into interchangeable classes and then either create an Admin page or a config file where they can select the ones they want. But I think the solution you choose will depend on the sort of customization you're facing. In general I think the design practise of creating well-defined interfaces for the various components of your app makes it easy to create interchangeable components and behaviors. Whether you are picking the components at runtime with some configuration options or at compile time by actually selecting a subset of files to deploy (via makefiles or version control) is not as important in my mind. I'm sure there are many other possible solutions as well. If you could provide some more info about the sorts of customizations I think you'll probably get some interesting answers. Byron From: Michael Reddick [mailto:michael.redd...@gmail.com] Sent: Thursday, April 02, 2009 1:32 PM To: catalyst@lists.scsys.co.uk Subject: [Catalyst] Custom changes to different instances of the same app If you have a Catalyst app installed at several different customers, and they all want their own customizations, whether it be to the interface or the business logic, what are some different strategies for doing that using good coding standards? Thanks, Michael ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
[Catalyst] Catalyst::View::Email tests failing
Hi J Shirley I'm trying to install your neat C::V::Email package but I'm running into some test failures. I thought you might like to know about them. The test failures look benign to me so I'm going to force install, but let me know if you think that the errors could be dangerous. Let me know if you want me to try out a test patch. Thanks Byron CPAN.pm: Going to build J/JS/JSHIRLEY/Catalyst-View-Email-0.11.tar.gz Cannot determine perl version info from lib/Catalyst/View/Email.pm *** Module::AutoInstall version 1.03 *** Checking for Perl dependencies... [Core Features] - Catalyst ...loaded. (5.71000 = 5.7) - Class::C3 ...loaded. (0.19) - Email::Send ...loaded. (2.194 = 2.185) - Email::MIME ...loaded. (1.863 = 1.859) - Email::MIME::Creator ...loaded. (1.455 = 1.453) [Template Toolkit Support] - Catalyst::View::TT...loaded. (0.26) [Mason Support] - Catalyst::View::Mason ...loaded. (0.16) [Net::SMTP Support] - MIME::Base64 ...loaded. (3.07) - Authen::SASL ...missing. == Auto-install the 1 optional module(s) from CPAN? [n] *** Module::AutoInstall configuration finished. Checking if your kit is complete... Warning: the following files are missing in your kit: t/lib/TestApp/cache/obj/ Please inform the author. Writing Makefile for Catalyst::View::Email cp lib/Catalyst/View/Email.pm blib/lib/Catalyst/View/Email.pm cp lib/Catalyst/View/Email/Template.pm blib/lib/Catalyst/View/Email/Template.pm cp lib/Catalyst/Helper/View/Email.pm blib/lib/Catalyst/Helper/View/Email.pm cp lib/Catalyst/Helper/View/Email/Template.pm blib/lib/Catalyst/Helper/View/Email/Template.pm Manifying blib/man3/Catalyst::View::Email.3pm Manifying blib/man3/Catalyst::View::Email::Template.3pm Manifying blib/man3/Catalyst::Helper::View::Email::Template.3pm Manifying blib/man3/Catalyst::Helper::View::Email.3pm JSHIRLEY/Catalyst-View-Email-0.11.tar.gz /usr/bin/make -- OK Running make test PERL_DL_NONLAZY=1 /usr/bin/perl -MExtUtils::Command::MM -e test_harness(0, 'inc', 'blib/lib', 'blib/arch') t/01use.t t/02pod.t t/04basic.t t/05template.t t/06config.t t/07mason.t t/01use.ok t/02pod.skipped: set TEST_POD to enable this test t/04basic...[warn] DEPRECATION WARNING: TestApp::View::Mason sets the use_match config variable to a true value. This has been deprecated. Please see the Catalyst::View::Mason documentation for details on use_match. t/04basic...ok t/05template[warn] DEPRECATION WARNING: TestApp::View::Mason sets the use_match config variable to a true value. This has been deprecated. Please see the Catalyst::View::Mason documentation for details on use_match. t/05template2/11 # Failed test 'text/plain part ok' # at t/05template.t line 33. # got: 'text/plain' # expected: 'text/plain; charset=us-ascii' # Failed test 'text/html ok' # at t/05template.t line 36. # got: 'text/html' # expected: 'text/html; charset=us-ascii' # Looks like you failed 2 tests of 11. t/05template Dubious, test returned 2 (wstat 512, 0x200) Failed 2/11 subtests t/06config..[warn] DEPRECATION WARNING: TestApp::View::Mason sets the use_match config variable to a true value. This has been deprecated. Please see the Catalyst::View::Mason documentation for details on use_match. t/06config..1/13 # Failed test 'text/plain ok' # at t/06config.t line 37. # got: 'text/plain' # expected: 'text/plain; charset=us-ascii' # Failed test 'text/html ok' # at t/06config.t line 39. # got: 'text/html' # expected: 'text/html; charset=us-ascii' # Looks like you failed 2 tests of 13. t/06config.. Dubious, test returned 2 (wstat 512, 0x200) Failed 2/13 subtests t/07mason...[warn] DEPRECATION WARNING: TestApp::View::Mason sets the use_match config variable to a true value. This has been deprecated. Please see the Catalyst::View::Mason documentation for details on use_match. t/07mason...1/10 # Failed test 'text/plain ok' # at t/07mason.t line 32. # got: 'text/plain' # expected: 'text/plain; charset=us-ascii' # Failed test 'text/html ok' # at t/07mason.t line 34. # got: 'text/html' # expected: 'text/html; charset=us-ascii' # Looks like you failed 2 tests of 10. t/07mason... Dubious, test returned 2 (wstat 512, 0x200) Failed 2/10 subtests Test Summary Report --- t/05template (Wstat: 512 Tests: 11 Failed: 2) Failed tests: 8, 10 Non-zero exit status: 2 t/06config (Wstat: 512 Tests: 13 Failed: 2) Failed tests: 10, 12 Non-zero exit status: 2 t/07mason (Wstat: 512 Tests: 10 Failed: 2) Failed tests: 7, 9 Non-zero exit status: 2 Files=6, Tests=40, 3 wallclock secs ( 0.01 usr 0.07 sys + 0.65 cusr 2.12 csys = 2.85 CPU) Result: FAIL Failed 3/6 test programs. 6/40 subtests failed. make: *** [test_dynamic] Error 2 JSHIRLEY/Catalyst-View-Email-0.11.tar.gz /usr/bin/make test -- NOT OK //hint// to see the
RE: [Catalyst] Catalyst::View::Email tests failing
J. Shirley wrote on 2009-02-05: On Thu, Feb 5, 2009 at 11:10 AM, Robert L Cochran cochr...@speakeasy.net wrote: Dangerous? Well I don't think this module runs around with a Uzi spraying everyone. I don't think it is a good practice to force an install on errors. You should always get to the bottom of why the errors are happening to start with. Once you do and have applied a fix, it should install cleanly, and you will be a better programmer for having investigated the cause. Bob Hi Bob, As a matter of etiquette, please don't top-post. Put your reply below in the thread to keep things consistent. In this case, the test failures are actually a problem that was resolved in Email::MIME::Creator and while it certainly doesn't carry firearms, it can break your application under certain circumstances. I think, relatively speaking, that counts as dangerous. However, I'm just totally going crazy and senile. I misread the test failure, as I already have a fixed version that never made it to CPAN though I would have sworn under oath it actually did. Catalyst::View::Email 0.12 just shipped to CPAN, that should resolve the error and had the fixed version of Email::MIME::Creator listed as a dependency. Thanks, -J Thanks for the quick fix, J. My Email::MIME::Creator is up to date at v1.455. I grabbed ftp://ftp.funet.fi/pub/languages/perl/CPAN/modules/by-module/Catalyst/Catalyst-View-Email-0.12.tar.gz and the tests worked. Thanks Byron ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Re: Supressing passwords in debug messages
Tomas Doran wrote on 2009-01-29: On 29 Jan 2009, at 18:53, Byron Young wrote: Hi - I'm not sure what the repost policy on patches, but I have the feeling this one slipped through the cracks. Let me know if it's generally annoying to repost stuff. No, reposting if things get dropped on the floor good :) If you have time, then arriving on #catalyst-dev and making noise also gets stuff done. This is a patch that allows you to suppress printing the value of certain query or body parameters when running Catalyst in debug mode - For example, if you want to hide passwords sent from the login page, you can put this in your app config (yaml): Having been discussed in #catalyst-dev, we think that the patch could be made both more generic, and more elegant. The key thing is to split the table drawing, and the data filtering into separate methods (maybe filter_debug_data?). This would then allow you to filter per-type, and support things such as redact_parameters (all), redact_body_parameters, redact_query_parameters, and even potentially to add support for filtering things like the URI (I can see use-cases where that'd be significant - e.g. not wanting to log session IDs which are in URIs).. Have a look at the way the debug screen stuff works (in Catalyst::Engine), this is more elegant and would also benefit from being able to have things redacted I guess - as with the current patch, you're going to display the things you're redacting in the logs to the end user... Cheers t0m Tom, Thanks for the feedback. I think you're referring to $c-dump_these() and it's usage in finalize_error(). I'll refactor log_parameters() to call a separate method that will return the params to log, akin to dump_these(). Not sure when I'll have time for it since my current solution is working for me and I have some big deadlines coming up. Hopefully within the next month. Thanks byron ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
[Catalyst] Re: Supressing passwords in debug messages
Hi - I'm not sure what the repost policy on patches, but I have the feeling
this one slipped through the cracks. Let me know if it's generally annoying to
repost stuff.
This is a patch that allows you to suppress printing the value of certain query
or body parameters when running Catalyst in debug mode - For example, if you
want to hide passwords sent from the login page, you can put this in your app
config (yaml):
Debug:
redact_parameters:
- password
and the resulting log will look like:
[debug] Query Parameters are:
.-+--.
| Parameter | Value|
+-+--+
| password| (redacted by config) |
| username| some_user|
'-+--'
There are two patches attached
- redact-patch.diff - contains patch and test
- cookbook-patch.diff - patch for cookbook entry about this
Thanks to J Shirley for help with this.
Thanks
Byron
Byron Young wrote on 2009-01-16:
-Original Message-
From: Byron Young [mailto:byron.yo...@riverbed.com]
Sent: Friday, January 16, 2009 6:39 PM
To: The elegant MVC web framework
Subject: RE: [Catalyst] Re: Supressing passwords in debug messages
Byron Young wrote on 2009-01-12:
J. Shirley wrote on 2009-01-12:
On Mon, Jan 12, 2009 at 2:35 PM, Byron Young
byron.yo...@riverbed.com wrote:
J. Shirley wrote on 2009-01-12:
On Mon, Jan 12, 2009 at 10:45 AM, Byron Young
byron.yo...@riverbed.com wrote:
[snip]
The patch I'm creating needs to be configured in some way, I am
thinking at this point it can be configured as follows:
package MyApp;
__PACKAGE__-config(
'Debug' = {
skip_dump_parameters = 1, # Simply don't render the
parameters incoming, very shotgunny skip_dump_parameters =
[ qw/password/ ], # Show '(redacted
by
config)' as the value of these fields
}
);
I'll need to bake tests for this, which there are currently no tests
for handling the dumping of parameters so it will be a bit more. If
someone wants to help with that, let me know and I can help guide.
-J
I'd be happy to write some unit tests. I haven't worked with
any
of the Catalyst unit tests before so I'm not sure what the process is
like for getting the code, setting up the test environment, making and
submitting changes and unit tests, etc. Is there a doc you can point
me to? I don't see anything in the manual or wiki.
Byron
Mostly it is just checking out the code from svn and starting.
The
patch that I've started is at http://scsys.co.uk:8001/22410 - you can
apply this to a svn checkout of
http://dev.catalystframework.org/repos/Catalyst/Catalyst- Runtime/5.70
It doesn't have the actual testing part, just a stub. I'll be working
on it more over today and tomorrow when I get free moments, but
they're few and far between.
Ditto on the lack of free time. I'll check it out and let you know
what I come up with.
byron
J Shirley - I finally got a chance to look at this today. You did
most of the work for me. I just updated the unit test, changed the
'skip_dump_parameters' parameter to 'redact_parameters', and
expanded the log_parameters() documentation a bit. I also added a
section to the cookbook explaining how to use the parameter.
Attached are two patches:
redact-patch.diff - patch containing the new unit test and changes to
Catalyst.pm. cookbook-patch.diff - patch containing a new cookbook
section on
this feature, for the Catalyst-Manual repository
Anything else I need to do?
Byron
redact-patch.diff
Description: redact-patch.diff
cookbook-patch.diff
Description: cookbook-patch.diff
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] generic way of session has expired message?
Jens Schwarz wrote on 2009-01-28: Hi, I there some generic way of providing a session has expired message to the user after a session has expired? I don't know how to distinct between a user whose session just expired and a user that has not yet logged in (a session has expired message does not make sense for the latter one). Any hints? Thanks Jens -- NUR NOCH BIS 31.01.! GMX FreeDSL - Telefonanschluss + DSL für nur 16,37 EURO/mtl.!* http://dsl.gmx.de/?ac=OM.AD.PD003K11308T4569a Hi Jens I haven't used it myself, but check out $c-session_delete_reason() in the Catalyst::Plugin::Session docs. Byron ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Re: OT: Better TT pager?
Aristotle Pagaltzis wrote on 2008-12-24: [snip] This is also a perfect demonstration of why I say that TT sucks. This should by all rights be handled in the template since it’s display logic, but writing it in that TT mini-language rather than Perl is fugly. And PERL and RAWPERL blocks have their own problems. Anyway. Someday I’ll quit yapping and write some code. Have you tried Mason? It's delicious. Byron ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Re: Supressing passwords in debug messages
Byron Young wrote on 2009-01-12:
J. Shirley wrote on 2009-01-12:
On Mon, Jan 12, 2009 at 2:35 PM, Byron Young
byron.yo...@riverbed.com wrote:
J. Shirley wrote on 2009-01-12:
On Mon, Jan 12, 2009 at 10:45 AM, Byron Young
byron.yo...@riverbed.com wrote:
[snip]
The patch I'm creating needs to be configured in some way, I am
thinking at this point it can be configured as follows:
package MyApp;
__PACKAGE__-config(
'Debug' = {
skip_dump_parameters = 1, # Simply don't render the
parameters incoming, very shotgunny skip_dump_parameters = [
qw/password/ ], # Show '(redacted
by
config)' as the value of these fields
}
);
I'll need to bake tests for this, which there are currently no tests
for handling the dumping of parameters so it will be a bit more. If
someone wants to help with that, let me know and I can help guide.
-J
I'd be happy to write some unit tests. I haven't worked with
any
of the Catalyst unit tests before so I'm not sure what the process is
like for getting the code, setting up the test environment, making and
submitting changes and unit tests, etc. Is there a doc you can point
me to? I don't see anything in the manual or wiki.
Byron
Mostly it is just checking out the code from svn and starting.
The
patch that I've started is at http://scsys.co.uk:8001/22410 - you can
apply this to a svn checkout of
http://dev.catalystframework.org/repos/Catalyst/Catalyst- Runtime/5.70
It doesn't have the actual testing part, just a stub. I'll be working
on it more over today and tomorrow when I get free moments, but they're
few and far between.
Ditto on the lack of free time. I'll check it out and let you know
what I come up with.
byron
J Shirley - I finally got a chance to look at this today. You did most of the
work for me. I just updated the unit test, changed the 'skip_dump_parameters'
parameter to 'redact_parameters', and expanded the log_parameters()
documentation a bit. I also added a section to the cookbook explaining how to
use the parameter.
Attached are two patches:
redact-patch.diff - patch containing the new unit test and changes to
Catalyst.pm.
cookbook-patch.diff - patch containing a new cookbook section on this
feature, for the Catalyst-Manual repository
Anything else I need to do?
Byron
redact-patch.diff
Description: redact-patch.diff
cookbook-patch.diff
Description: cookbook-patch.diff
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Re: Supressing passwords in debug messages
Ansgar Burchardt wrote on 2009-01-11:
Hi,
J. Shirley jshir...@gmail.com writes:
=== lib/Catalyst.pm
== ---
lib/Catalyst.pm (revision 18145) +++ lib/Catalyst.pm (local) @@
-1830,7 +1830,11 @@
if ( $c-debug keys %{ $c-request-query_parameters } )
{
my $t = Text::SimpleTable-new( [ 35, 'Parameter' ], [
36, 'Value' ] );
+my %skip = map { $_ = $_ } @{
+$c-config-{'Plugin::Debug'}-
{'skip_dump_parameters'} || []
+};
for my $key ( sort keys %{ $c-req-query_parameters } )
{
+next if $skip{$key};
my $param = $c-req-query_parameters-{$key};
my $value = defined($param) ? $param : '';
$t-row( $key,
I think it would be better to show that the parameter was sent, but
Catalyst configured to not display its value. This can be done for
example by displaying a value of `(hidden)'.
If the parameter is simply skipped, it might be confusing if you forget
that you configured Catalyst to not display it.
Regards,
Ansgar
Yeah, I agree that the parameter should be shown as sent, but just not show the
value.
J Shirley - Thanks for looking into it. Let me know if there's anything I can
do to help.
Thanks,
Byron
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Re: Supressing passwords in debug messages
J. Shirley wrote on 2009-01-12:
On Mon, Jan 12, 2009 at 10:45 AM, Byron Young
byron.yo...@riverbed.com wrote:
Ansgar Burchardt wrote on 2009-01-11:
Hi,
J. Shirley jshir...@gmail.com writes:
=== lib/Catalyst.pm
==
---
lib/Catalyst.pm (revision 18145) +++ lib/Catalyst.pm (local) @@
-1830,7 +1830,11 @@
if ( $c-debug keys %{ $c-request-query_parameters }
)
{
my $t = Text::SimpleTable-new( [ 35, 'Parameter' ], [
36, 'Value' ] );
+my %skip = map { $_ = $_ } @{
+$c-config-{'Plugin::Debug'}-
{'skip_dump_parameters'} || []
+};
for my $key ( sort keys %{ $c-req-query_parameters }
)
{
+next if $skip{$key};
my $param = $c-req-query_parameters-{$key};
my $value = defined($param) ? $param : '';
$t-row( $key,
I think it would be better to show that the parameter was sent, but
Catalyst configured to not display its value. This can be done for
example by displaying a value of `(hidden)'.
If the parameter is simply skipped, it might be confusing if you
forget that you configured Catalyst to not display it.
Regards,
Ansgar
Yeah, I agree that the parameter should be shown as sent, but just not
show the value.
J Shirley - Thanks for looking into it. Let me know if there's
anything I can do to help.
Thanks,
Byron
The patch I'm creating needs to be configured in some way, I am
thinking at this point it can be configured as follows:
package MyApp;
__PACKAGE__-config(
'Debug' = {
skip_dump_parameters = 1, # Simply don't render the parameters
incoming, very shotgunny skip_dump_parameters = [ qw/password/
], # Show '(redacted
by
config)' as the value of these fields
}
);
I'll need to bake tests for this, which there are currently no tests for
handling the dumping of parameters so it will be a bit more. If someone
wants to help with that, let me know and I can help guide.
-J
I'd be happy to write some unit tests. I haven't worked with any of the
Catalyst unit tests before so I'm not sure what the process is like for getting
the code, setting up the test environment, making and submitting changes and
unit tests, etc. Is there a doc you can point me to? I don't see anything in
the manual or wiki.
Byron
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Re: Supressing passwords in debug messages
J. Shirley wrote on 2009-01-12:
On Mon, Jan 12, 2009 at 2:35 PM, Byron Young
byron.yo...@riverbed.com wrote:
J. Shirley wrote on 2009-01-12:
On Mon, Jan 12, 2009 at 10:45 AM, Byron Young
byron.yo...@riverbed.com wrote:
[snip]
The patch I'm creating needs to be configured in some way, I am
thinking at this point it can be configured as follows:
package MyApp;
__PACKAGE__-config(
'Debug' = {
skip_dump_parameters = 1, # Simply don't render the
parameters incoming, very shotgunny skip_dump_parameters = [
qw/password/ ], # Show '(redacted
by
config)' as the value of these fields
}
);
I'll need to bake tests for this, which there are currently no tests
for handling the dumping of parameters so it will be a bit more. If
someone wants to help with that, let me know and I can help guide.
-J
I'd be happy to write some unit tests. I haven't worked with any
of the Catalyst unit tests before so I'm not sure what the process
is like for getting the code, setting up the test environment,
making and submitting changes and unit tests, etc. Is there a doc
you can point me to? I don't see anything in the manual or wiki.
Byron
Mostly it is just checking out the code from svn and starting. The
patch that I've started is at http://scsys.co.uk:8001/22410 - you can
apply this to a svn checkout of
http://dev.catalystframework.org/repos/Catalyst/Catalyst- Runtime/5.70
It doesn't have the actual testing part, just a stub. I'll be working
on it more over today and tomorrow when I get free moments, but they're
few and far between.
Ditto on the lack of free time. I'll check it out and let you know what I come
up with.
byron
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Supressing passwords in debug messages
Jesse Sheidlower wrote on 2009-01-07: On Wed, Jan 07, 2009 at 10:39:34AM -0800, Byron Young wrote: I like the CATALYST_DEBUG mode for the test server - it's really nice to be able to see all the GET and POST params and requests as they happen. However, my app uses LDAP authentication and I really don't want people's LDAP passwords getting printed with the rest of the parameters. Is there a way to suppress certain parameters from being printed? I didn't see anything in the docs about it, but thought I'd ask before jumping into the code. This is a FAQ: http://dev.catalystframework.org/wiki/faq How do I hide certain variables (e.g. user/password) from the debug screen? Jesse Sheidlower Jesse, Thanks for the reply, but that doesn't quite do what I'm asking (or I'm using it wrong?). I mean the debug log that's prints request info when -Debug or CATALYST_DEBUG is turned on. For example: [debug] Body Parameters are: .-+--. | Parameter | Value| +-+--+ | password| REDACTED | | submit | Go | | username| youngb | '-+--' [debug] POST request for login from 10.16.5.10 [debug] Path is login [debug] Found sessionid e4202e839e17004bc05baff653ad659f7b165ee7 in cookie [debug] Restored session e4202e839e17004bc05baff653ad659f7b165ee7 [debug] Icebox::Controller::Login - Found username youngb, attempting login [debug] Icebox::Controller::Login - LDAP login successful for youngb [debug] Icebox::Controller::Login - Database login successful for youngb [debug] ***Login::index - redirecting to http://icebox-dev.lab.nbttech.com:3000/ [debug] Redirecting to http://icebox-dev.lab.nbttech.com:3000/; [info] Request took 1.282297s (0.780/s) .+---. | Action | Time | ++---+ | /auto | 0.000270s | | /login/index | 0.078559s | | /end | 0.000765s | '+---' It's in that 'Body Parameters' section that I don't want the password to be displayed. It ends up there in plain text if debugging is turned on. Is there a simple way to remove or it or replace the value with ''? (but thanks for the link to the FAQ - I was only reading the POD. I have been using Catalyst for a while and have never seen a link to the Catalyst Wiki before - Maybe it would be a good idea to add a link to the Manual?) Thanks, Byron ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
[Catalyst] Supressing passwords in debug messages
I like the CATALYST_DEBUG mode for the test server - it's really nice to be able to see all the GET and POST params and requests as they happen. However, my app uses LDAP authentication and I really don't want people's LDAP passwords getting printed with the rest of the parameters. Is there a way to suppress certain parameters from being printed? I didn't see anything in the docs about it, but thought I'd ask before jumping into the code. Thanks, Byron ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] configuring plugin's
Dermot wrote on 2008-06-30:
Hi,
I seem to be struggling with the configuring of plugins, in this case
Authorization::Roles.
I have the following in lib/MyApp.pm
__PACKAGE__-setup(qw/-Debug
ConfigLoader
Static::Simple
StackTrace
Authentication
Authorization::Roles
Session::State
Session::Store
Session::Store::FastMmap
Session
/);
and MyApp.yml has
name: MyApp
Model::MyAppDB:
connect_info:
- DBI:SQLite:__HOME__/MyApp.db
Authentication:
default_realm: user
realms:
user:
credential:
class: Password password_field: password password_type: clear store:
class: DBIx::Class user_class: UsersRoles role_relation: role_id
role_field: role
I think that all the above is correctly configured. The grapping void
in my understanding is making the schema available in my methods. So
sub list : Local {
my ($self, $c) = @_;
print STDERR login=,$c-check_roles('admin'),\n;
$c-detach('denied') unless $c-check_roles('admin');
$c-stash-{stuff} = [$c-model('MyAppDB::Stuff')-all];
$c-stash-{template} = 'submissions/list.tt2
}
Give me the error:
Can't locate object method asser_user_roles via package MyApp
Looks like you have a typo somewhere. Should be 'assert_user_roles', maybe?
byron
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Multiple inheritance of controllers
I ran into the same problem. But I found that adding this to my controllers
that inherit from multiple controller base classes works as a work around:
use Class::C3;
sub create_action {
my $self = shift;
return $self-maybe::next::method(@_);
}
It's an annoying workaround, but I didn't have time to pursue the issue any
further :P Does anybody know why this might be happening?
Anyway, I hope that helps somewhat..
Byron
Zbigniew Lukasiak wrote on 2008-06-13:
I've tried to build a module that would merge two base controllers and
then use it as another base controller. Unfortunately it looks like
only actions from the first base class in the 'use base ...' phrase are
being inherited:
I've created a test case for that - it is attached and I paste it
below for your convenience.
If someone sent me some instructions how to modify it to fit into the
Moose model I'll test it on the Moose branch as well (I am not sure if
they would be base controllers there at all - perhaps Roles instead?).
Cheers,
Zbigniew
package TestApp::Controller::Action::SubclassedChained;
use strict;
use warnings;
use base qw/Intermediate/;
1;
==
package Intermediate; use strict; use warnings;
use base qw/Catalyst::Controller::REST Base/; # this one does not
inherit actions from Base
#use base qw/Base Catalyst::Controller::REST/; # this one does
1; ==
package Base; use strict; use warnings;
use base qw/Catalyst::Controller/;
sub basefoo :CaptureArgs(1) :Chained('/') { }
sub baseendpoint :Args(1) :Chained('basefoo') { }
1;
--
Zbigniew Lukasiak
http://brudnopis.blogspot.com/
http://perlalchemy.blogspot.com/
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Using URIs for my app in another program
Tomas Doran wrote on 2008-06-06: On 7 Jun 2008, at 01:17, Byron Young wrote: Is there any way to fake a request when my app starts? It seems like a bit of a catch-22 -- I need to know a URL in order to fake the request, but in order to generate the URL I need to have a request... The problem which you have here is that the application doesn't know what URLs it's attached to until run time. Remember - you can have multiple virtual hosts pointing to the same application instance, therefore catalyst *can't* know in advance where it is going to be served from. True. All I need to know in this case is just one valid URL, though. The simple perl HTTP engine which script/myapp_server.pl uses prints a nice message with the app URL - but that's because it's a toy compared to what you might do in real deployment. Am I doomed to storing the URL for my 'populate database with urls' action in a config file and updating it by hand when I move the web app? I think so, but I think that you *want* to do this. Otherwise, how do you deal with testing the app? And also testing your thing which requests the app. Surely you're gonna want to be able to test against demo servers you've brought up on localhost? Actually, that's specifically the use case I'm thinking of here. Each developer on this project has his own testing setup on the dev machine, while the production version runs on another host. Each test setup includes a database, daemon, and web app. We do use config files to point everything to the right codebase and database, so it's not the end of the world to add another entry for the url, but I try to be careful of the amount of configuration we need. I want to avoid the situation where a developer forgets to change his dev config file to point to his dev web app and so his daemon ends up talking to the production server instead, potentially resulting in something bad happening. So, if there is some totally automated way to do things instead, I opt for that. Of course, this may just be exposing some bad practices on our end wrt to dev environments and config files and protecting the production server from fallible developers such as myself. I think that you're trying to solve the wrong problem here, sorry. To be honest - how often are you really going to move the webapp once it's deployed? What's wrong with giving it it's own DNS CNAME which you switch between boxen hosting it? Yeah, that's true - we probably won't move the production app, and we'll use a CNAME if we do. Thanks for your help Byron Cheers Tom ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
[Catalyst] Using URIs for my app in another program
Hi all I've run into a situation that I can't find a good solution to and am hoping to solicit some ideas from people on the list here. An app I've been working on consists of two main parts - a Catalyst-based web app which provides the UI, and a daemon that does the heavy lifting. They are totally separate applications that talk to one another via database. The daemon will sometimes send emails to users when it finishes certain tasks that the users are interested in, and I want to include links in the emails that would bring the user to a Results page on the web app where they can get more information about the result of the task. So, what I would like is to be able to call $c-uri_for() from my daemon. Is there any 'sanctioned' way of doing that? If possible, I'd like to avoid loading my whole catalyst app and somehow tricking it into believing its running under the same environment as my real app. Another option is to use the database to pass URIs around. Maybe a URI table that maps finished tasks to URIs. However, I would rather just be able to arbitrarily call $c-uri_for() whenever I decide I need to include a URI in my daemon instead of having to create a table or column whenever I decide I want to send a link to a new type of resource. Any suggestions? Maybe I'm just going about it all wrong? Thanks! Byron ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Using URIs for my app in another program
Ah, right, I need a HTTP request to get base. I wasn't thinking straight about that. I think storing links in the database is what I'll do, because that leaves the responsibility of generating URIs solely with the catalyst app. If I refactor code or move things around I don't want to have to update my daemon at all. Thanks for your feedback, Robert. Byron From: Robert Krimen [mailto:[EMAIL PROTECTED] Sent: Thursday, May 01, 2008 3:34 PM To: The elegant MVC web framework Subject: Re: [Catalyst] Using URIs for my app in another program The most straightforward method I can think of is to factor out the URIs to a common config file. You can have your daemon read in the config file and generate URIs from that. Your Catalyst application can either continue to use base to generate URIs or you can generate some special URIs from the config file as well. The problem with having your daemon load up Catalyst to generate a URI is that I'm not sure how that would work. Catalyst generates URIs given a HTTP request. No HTTP request, no base URI. Another safety measure is that you can have your Catalyst appl do a sanity check against the URI config file to make sure it doesn't become out of date (maybe a startup test request). Rob On Thu, May 1, 2008 at 2:27 PM, Byron Young [EMAIL PROTECTED]mailto:[EMAIL PROTECTED] wrote: Hi all I've run into a situation that I can't find a good solution to and am hoping to solicit some ideas from people on the list here. An app I've been working on consists of two main parts - a Catalyst-based web app which provides the UI, and a daemon that does the heavy lifting. They are totally separate applications that talk to one another via database. The daemon will sometimes send emails to users when it finishes certain tasks that the users are interested in, and I want to include links in the emails that would bring the user to a Results page on the web app where they can get more information about the result of the task. So, what I would like is to be able to call $c-uri_for() from my daemon. Is there any 'sanctioned' way of doing that? If possible, I'd like to avoid loading my whole catalyst app and somehow tricking it into believing its running under the same environment as my real app. Another option is to use the database to pass URIs around. Maybe a URI table that maps finished tasks to URIs. However, I would rather just be able to arbitrarily call $c-uri_for() whenever I decide I need to include a URI in my daemon instead of having to create a table or column whenever I decide I want to send a link to a new type of resource. Any suggestions? Maybe I'm just going about it all wrong? Thanks! Byron ___ List: Catalyst@lists.scsys.co.ukmailto:Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/ ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Using URIs for my app in another program
J. Shirley wrote on 2008-05-01: On Thu, May 1, 2008 at 4:00 PM, Byron Young [EMAIL PROTECTED] wrote: Ah, right, I need a HTTP request to get base. I wasn't thinking straight about that. I think storing links in the database is what I'll do, because that leaves the responsibility of generating URIs solely with the catalyst app. If I refactor code or move things around I don't want to have to update my daemon at all. Thanks for your feedback, Robert. Another option is to use a distributed cache system like memcached. Pretty easy to setup, and then you can have a warm cache action that populates the configured URIs in the cache. So, your daemon queries cache, if it comes up empty then make a call out to the app to generate the URLs and stuff them in cache. -J Hey J. Yes, that could work too, but it requires making a call to the catalyst app, which means at least one hard-coded url will be in the daemon. I considered having a page on my app that would just serve up a requested url, which is similar to your proposal, but I'd rather avoid hard-coding any urls in the daemon. Thanks for the suggestion. byron ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Usage of $c-uri_for and $c-res-redirect
-Original Message-
From: Dustin Suchter [mailto:[EMAIL PROTECTED]
Sent: Wednesday, April 02, 2008 5:44 PM
To: The elegant MVC web framework
Subject: Re: [Catalyst] Usage of $c-uri_for and $c-res-redirect
So I tried to look at the source of the test that failed, and I'm
not sure I even understand what it is doing. Here's the actual
error
message I'm getting:
t/01useok
t/02podskipped: set TEST_POD to enable this test
t/03podcoverageskipped: Test::Pod::Coverage 1.04 required
t/04ssl1/15
# Failed test 'redirect uri ok'
# at t/04ssl.t line 43.
# got: 'http://localhost:443/ssl/unsecured'
# expected: 'http://localhost/ssl/unsecured'
# Failed test 'redirect with params ok'
# at t/04ssl.t line 47.
# got:
'http://localhost:443/ssl/unsecured?a=1a=2b=3c=4'
# expected: 'http://localhost/ssl/unsecured?a=1a=2b=3c=4'
# Looks like you failed 2 tests of 15.
t/04ssl Dubious, test returned 2 (wstat 512, 0x200)
Failed 2/15 subtests
t/05ssl_host...ok
t/06remain_in_ssl..ok
The top error came from:
40 # test redirect back to http mode
41 ok( $res = request('https://localhost/ssl/unsecured'),
'request ok' );
42 is( $res-code, 302, 'redirect back to http ok' );
43 is( $res-header('location'),
'http://localhost/ssl/unsecured', 'redirect uri ok' );
So I'm guessing that line #41 is trying to request a secure action
that would normally redirect you back to an insecure place. This
jives, because it is requesting an HTTPS URL. However, If I am
right, then why is it considered a failure to end up on a non-
secure
URL? I guess I don't understand what these tests are doing.
-d
Those don't make sense to anyone else, either, right? I mean, if
you
are going to request a URL from port 443,
Heh, you know, I think I force installed that plugin because those test results
looked like false negatives. It has been working fine for me, though. I
probably should have investigated it more, but I just went with it and never
looked back. I'm going to be installing a bunch of the catalyst modules on
another server this week, so I'll let you know if I run into the same thing
again, and I'll try to look into it if I have time.
Byron
Dustin Suchter wrote:
So I like the idea of the plugin, but I'm having a hard time
installing
it. I found one hit on the Internet that with a report of someone
else
having the exact same problem as me, but I don't see the solution
online:
http://www.nntp.perl.org/group/perl.cpan.testers/2007/09/msg631733.
html
Any ideas?
Byron Young wrote:
-Original Message-
From: Dustin Suchter [mailto:[EMAIL PROTECTED]
Sent: Wednesday, March 26, 2008 1:13 PM
To: The elegant MVC web framework
Subject: [Catalyst] Usage of $c-uri_for and $c-res-redirect
Let's say I want to send people back and forth between an HTTP
connection and an HTTPS connection on a server based on some
action. For
example, clicking on a logout button from within my
application
while
connected via HTTPS does something like:
$c-res-redirect(http://foo.com/;);
The above seems like a fine solution except it totally
disregards
the
beauty of uri_for, which I would love to be using for stuff
like
this.
Without uri_for, problems arise when you do things like test
via
the
built in Perl webserver (the one running on port 3000 by
default)
while
on the same webserver as my production application.
So the real question is, how do I properly refer to my
webserver
and/or
application root and include port or SSL flags? I guess I'm
looking
for
something like $0 within uri_for.
-d
Hey Dustin,
There's actually a neat plugin for doing this called
Catalyst::Plugin::RequireSSL. You just include it in your
plugins and
then call $c-require_ssl() at the top of any actions you want
to use
SSL for. It will disable itself automatically on the test
server, too.
Cheers,
Byron
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-
bin/mailman/listinfo/catalyst
Searchable archive:
http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-
bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-
archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Usage of $c-uri_for and $c-res-redirect
-Original Message- From: Dustin Suchter [mailto:[EMAIL PROTECTED] Sent: Wednesday, March 26, 2008 1:13 PM To: The elegant MVC web framework Subject: [Catalyst] Usage of $c-uri_for and $c-res-redirect Let's say I want to send people back and forth between an HTTP connection and an HTTPS connection on a server based on some action. For example, clicking on a logout button from within my application while connected via HTTPS does something like: $c-res-redirect(http://foo.com/;); The above seems like a fine solution except it totally disregards the beauty of uri_for, which I would love to be using for stuff like this. Without uri_for, problems arise when you do things like test via the built in Perl webserver (the one running on port 3000 by default) while on the same webserver as my production application. So the real question is, how do I properly refer to my webserver and/or application root and include port or SSL flags? I guess I'm looking for something like $0 within uri_for. -d Hey Dustin, There's actually a neat plugin for doing this called Catalyst::Plugin::RequireSSL. You just include it in your plugins and then call $c-require_ssl() at the top of any actions you want to use SSL for. It will disable itself automatically on the test server, too. Cheers, Byron ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] On authentication again
Alex Povolotsky wrote:
I even don't understand what __PACKAGE__-config- I should use -
{'Plugin::Authentication'}, {authentication}
or{authentication}{dbic}.
Docs are unclear.
Alex,
It should be {authentication}. Here's mine:
authentication:
default_realm: database
realms:
database:
credential:
class: Password
password_field: password
password_type: hashed
password_hash_type: SHA-1
store:
class: DBIx::Class
user_class: DB::Users
id_field: username
role_field: name
role_relation: roles
Hope that helps
Byron
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] Base controllers and multiple inheritance
Peter Karman wrote:
Byron Young wrote on 3/5/08 4:29 PM:
Everything looks good, right? My List actions work! However, the
CRUD actions don't work because the FormConfig actions seem to have
been ignored -- $c-stash-{form} is undef :(
I wouldn't expect anything in stash() to persist from request to
request. Don't do anything with stash during new(). new() is called
once per app life, at startup. stash() is for each request.
Peter - thanks for taking the time to respond.
Sorry, I was unclear about that $c-stash-{form} comment.
$c-stash-{form} is created by the FormConfig action provided by
Catalyst::Controller::HTML::FormFu, which is a base class of my CRUD
base controller. For some reason, even though it appears new() is being
called down both inheritance chains, the HTML::FormFu actions aren't set
up properly.
I want to be using new() in my base controllers mainly to set up and
error check config(). I understand it's only called once per app life -
I don't want to do config() error checking every request.
$class-NEXT::new() doesn't do the right thing with this crazy
diamond-ish inheritance I have, and I think using Class::C3 in my base
classes while Catalyst::Controller and C::C::HTML::FormFu use NEXT is
causing some problems. I'll try to reproduce the behavior with a small
test later. I guess it's not really Catalyst-specific, but I figured
Catalyst people would know since I saw some notes in the docs about
wanting to switch Catalyst over to c3.
My workaround for now is to have new() in each of my actual controller
classes which passes control up to Catalyst::Controller::new() and then
calls each superclass's __setup() method specifically. I don't like
doing it this way, but for now I need to just keep moving ahead with
this project so I will leave it until a better solution appears.
Does anyone have any suggestions? Should I just forego using new()
and do initial setup the first time auto() is called in each of my
base controllers?
depends on what kind of thing you are setting up. I have used multiple
inheritance in my controllers to good effect (see
CatalystX::CRUD::Controller
and its descendents for example). In general, only set up things in
new() (or its cousins) for stuff you want to set up once per app:
config, persistent objects, etc. IME, there isn't much you want like
that.
I think the difference between your CX::CRUD::Controller's multiple
inheritance and mine is that you don't have the diamond pattern.
As an aside -- I wish I had seen your CatalystX::CRUD::Controller
before! I actually based mine off of your C::C::Rose::CRUD but wanted
to use DBIC and C::C::HTML::FormFu, so I made my own. Anyway, kudos!
Thanks,
Byron
___
List: Catalyst@lists.scsys.co.uk
Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst
Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/
Dev site: http://dev.catalyst.perl.org/
RE: [Catalyst] sending a variable to more than one tt2 template
You can call another template from within the template you specified in the stash, and pass the stash parameters along. I don't use TT, but I think PROCESS is how you call other templates. HTH Byron From: Emily Heureux [mailto:[EMAIL PROTECTED] Sent: Thursday, March 06, 2008 2:36 PM To: 'The elegant MVC web framework' Subject: [Catalyst] sending a variable to more than one tt2 template Hi, I have a variable in a Controller method sent to the stash for a particular tt file. What if I want that same variable to be available to another tt file? I assume I cannot set two templates in one method. Thanks, Emily ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
[Catalyst] Base controllers and multiple inheritance
Hi, I've been using Catalyst for a project and decided to make some base controllers for common functionality. So far I've made two: CRUD, and List, for standard database operations and for sortable, paged listings of data. They work fine on their own but I'm running into trouble with multiple inheritance. If I have a controller for some database table that I want to inherit the actions from both CRUD and List, only the first one inherited will initialize properly. The inheritance looks like this: Catalyst::Controller /\ / \ /\ /Catalyst::Controller::HTML::FormFu / | MyApp::Base::Controller::List | \ | \MyApp::Base::Controller::CRUD \/ \ / \/ MyApp::Controller::SomeDatabaseTable Each of the controller base classes has a new() method which sets up config data and does error checking and whatnot. If I inherit both of my base classes, only new() on the first inherited is called. I've tried calling $class-NEXT::new(@_) in each, and I've also tried Class::C3's $class-maybe::next::method(@_), but new() is still not called down the second inheritance chain. When I tried the Class::C3 approach I called Class::C3::initialize() in MyApp.pm - I wasn't sure where to call it so I tried before and after MyApp-setup() - still no luck. Can somebody tell me why new() isn't called down both paths? Thanks, Byron ___ List: Catalyst@lists.scsys.co.uk Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/catalyst@lists.scsys.co.uk/ Dev site: http://dev.catalyst.perl.org/
