mcr

Activity Graph

Page 1 of 1

2019-08-11

[16:59:02] mcr: Quit: Leaving.

2019-08-03

[06:05:23] mcr: Ping timeout: 248 seconds
[06:07:53] mcr: has joined #RubyOnRails

2019-07-23

[14:33:06] mcr: Ping timeout: 248 seconds
[14:39:56] mcr: has joined #RubyOnRails

2019-06-02

[00:06:50] mcr: has joined #RubyOnRails
[00:34:50] mcr: Read error: No route to host
[01:31:52] mcr: has joined #RubyOnRails

2019-06-01

[23:50:32] mcr: Read error: Connection reset by peer

2019-05-21

[14:06:09] mcr: *.net *.split
[14:14:03] mcr: has joined #RubyOnRails

2019-05-16

[21:46:44] mcr: Ping timeout: 252 seconds
[21:55:15] mcr: has joined #RubyOnRails

2019-05-07

[17:35:25] mcr: Ping timeout: 246 seconds
[20:00:36] mcr: has joined #RubyOnRails

2019-05-02

[21:38:27] mcr: has joined #RubyOnRails

2019-04-05

[22:01:19] mcr: Ping timeout: 246 seconds

2019-03-29

[20:03:33] mcr: Ping timeout: 252 seconds
[22:50:30] mcr: has joined #RubyOnRails

2019-03-22

[12:14:37] mcr: *.net *.split
[12:22:22] mcr: has joined #RubyOnRails

2019-03-10

[21:32:07] mcr: Ping timeout: 245 seconds
[21:35:53] mcr: has joined #RubyOnRails

2019-03-09

[23:37:17] mcr: Ping timeout: 250 seconds
[23:58:21] mcr: has joined #RubyOnRails

2019-03-06

[18:16:35] mcr: has joined #RubyOnRails
[18:17:39] mcr: "You have already activated bundler 1.17.3, but your Gemfile requires bundler 1.17.2. Prepending `bundle exec` to your command may solve this." <- yes, I know that I can bundle exec, and that's okay, but I'm just annoyed that I can't get my Gemfile to update to 1.17.3... I don't see 1.17.2 anywhere....
[18:19:25] mcr: Wondering if there are other people trying to deploy to kubernetes with the "distroless" ruby... I sure like not having compilers installed in production, but I would really like a way to further reduce crap. We install a few things from git, and the gemspec files like to call "git ls" to make a list of files, and there is no git in the target...
[20:00:02] mcr: can I get gem to clean an installed gem of the .o files, but leave the .so file alone? I'd even like to just remove all the .c files and documentation from the installed gem. Or can I get the .so file installed somewhere else?

2018-03-08

[00:34:59] mcr: has joined #ruby
[00:35:31] mcr: Client Quit

2018-03-02

[07:08:57] mcr: Ping timeout: 245 seconds

2018-02-27

[17:32:55] mcr: Ping timeout: 252 seconds
[18:46:27] mcr: has joined #ruby
[18:52:42] mcr: Ping timeout: 245 seconds
[19:18:59] mcr: has joined #ruby

2018-02-26

[00:24:53] mcr: Ping timeout: 252 seconds
[00:38:59] mcr: has joined #ruby
[00:43:07] mcr: Ping timeout: 245 seconds
[02:51:01] mcr: has joined #ruby
[02:55:12] mcr: Ping timeout: 245 seconds
[03:09:42] mcr: has joined #ruby
[04:22:29] mcr: Ping timeout: 252 seconds
[05:10:27] mcr: has joined #ruby
[12:30:12] mcr: Ping timeout: 245 seconds
[12:43:38] mcr: has joined #ruby
[14:01:53] mcr: Ping timeout: 240 seconds
[14:54:43] mcr: has joined #ruby
[14:59:11] mcr: Ping timeout: 248 seconds
[15:16:53] mcr: has joined #ruby
[15:21:05] mcr: Ping timeout: 240 seconds
[15:48:10] mcr: has joined #ruby
[16:06:36] mcr: Read error: Connection reset by peer
[16:06:46] mcr: has joined #ruby
[17:07:21] mcr: Ping timeout: 252 seconds
[17:45:01] mcr: has joined #ruby

2018-02-25

[17:46:27] mcr: Ping timeout: 240 seconds
[18:36:58] mcr: has joined #ruby
[19:48:25] mcr: Ping timeout: 252 seconds
[20:44:44] mcr: has joined #ruby
[20:58:07] mcr: Ping timeout: 245 seconds
[22:59:16] mcr: has joined #ruby

2018-02-23

[01:45:18] mcr: Ping timeout: 240 seconds
[01:49:59] mcr: has joined #ruby
[07:53:25] mcr: has joined #ruby
[08:11:50] mcr: *.net *.split
[08:23:01] mcr: has joined #ruby
[17:05:12] mcr: Ping timeout: 260 seconds
[18:20:00] mcr: has joined #ruby

2018-02-22

[23:16:55] mcr: has joined #ruby

2018-01-19

[17:52:56] mcr: Quit: Leaving.

2018-01-11

[14:59:48] mcr: Ping timeout: 246 seconds
[18:36:51] mcr: has joined #RubyOnRails

2018-01-08

[00:16:32] mcr: Ping timeout: 276 seconds
[00:32:36] mcr: has joined #RubyOnRails

2017-12-04

[16:54:12] mcr: I wish that we could get google to find things on ruby-doc.org newer than 1.9.3 by default... or if there were links in the header to easily navigate to the version of the doc I want to see. (I just edit the URL usually)

2017-12-01

[21:11:19] mcr: Ping timeout: 258 seconds
[21:14:40] mcr: has joined #ruby
[21:14:41] mcr: has joined #RubyOnRails
[21:20:30] mcr: Ping timeout: 246 seconds
[21:25:15] mcr: has joined #RubyOnRails

2017-11-27

[13:00:37] mcr: aperiros, crudely, you could strace if you need to know when.

2017-11-20

[00:57:03] mcr: https://github.com/postmodern/ruby-install <- this what you are speaking of?
[01:33:01] mcr: pbmarcano, yes, I've done this before.
[01:34:43] mcr: so, yes, you want to put a token in the URL that you generate. It should index into a table. The token should be some randomly generated number, Base64::urlsafe_encoded().
[01:36:12] mcr: You don't want "fake authentication", just no authentication, with the form brought up because of the URL. You may want to expire the token after use, and perhaps expire it after some period of time as well.
[02:31:59] mcr: what's the text value for their reference?
[02:33:09] mcr: Oh, you might as well just Base64::urlsafe_encode64(SecureRandom.random_bytes(12))
[02:33:34] mcr: and use that as your url token and value in the text column. No reason to store both binary and base64 version.

2017-11-19

[23:29:01] mcr: My only problem with RVM is the |sh method of installing. You can install from tar.gz, and I'd like to have a .deb file. I don't see a problem with rvm in production.

2017-11-15

[21:01:48] mcr: has joined #RubyOnRails
[21:07:14] mcr: I'm working on David (CoAP server), trying to improve the ResourceDiscovery mechanism. I would like to annotate which routes should be discoverable. I'm looking for a gem (or FAQ) that extends ActionDispatch::Journey::Route (and/or Path?)

2017-11-12

[01:10:07] mcr: has left #RubyOnRails: ()

2017-11-02

[13:27:44] mcr: Quit: Leaving.
[13:37:32] mcr: has joined #RubyOnRails

2017-09-17

[18:05:59] mcr: Ping timeout: 255 seconds
[20:41:13] mcr: has joined #RubyOnRails
[22:35:08] mcr: Ping timeout: 240 seconds

2017-09-16

[18:36:03] mcr: Quit: Leaving.
[18:36:10] mcr: has joined #RubyOnRails
[21:42:36] mcr: are you sure you are really getting a double backslash? In a "" string, a single \ will be shown as \\, because otherwise, it's an escape.

2017-09-15

[12:25:21] mcr: Ping timeout: 246 seconds
[13:18:44] mcr: has joined #RubyOnRails
[13:27:23] mcr: Ping timeout: 246 seconds
[14:07:17] mcr: has joined #RubyOnRails

2017-09-13

[13:32:35] mcr: has joined #RubyOnRails
[13:34:06] mcr: I have rake tasks that run from cron. When all is well, I don't need to see any output. When things go wrong, I want to see the debug output. Has anyone seen any Logger replacements that collects the debug logging, and then emits it *only* if a logger.error or some other signal is created? I can create such a thing myself of course.
[14:39:15] mcr: Ping timeout: 246 seconds
[15:03:36] mcr: has joined #RubyOnRails

2017-06-30

[13:08:55] mcr: Quit: Leaving.

2013-08-08

[19:35:23] mcr: I am using Rails.cache to avoid repeative loading of some relatively static data during a rake task.
[19:35:55] mcr: p = Phone.includes([:site => :client, :btn => :client]).first. Client is reachable by two relationships. But: irb(main):010:0> p.site.client.object_id => 67190640 vs irb(main):011:0> p.btn.client.object_id => 67170780.
[19:36:35] mcr: and I see two queries. Before I do my own caching (maybe on top of Rails.cache, maybe not)... is there some other setting I'm missing?

2013-08-07

[20:24:12] mcr: Why does rails keep asking postgresql about the attr.attname in production. Shouldn't this happen once, and then be cached?

2013-02-20

[15:18:35] mcr: there seems to be a difference in rspec / I18n between rails 3.0 and rails 3.1... the lookup by view name does not work for the top view (but works for partials)
[15:18:41] mcr: is this known?
[16:01:33] mcr: it looks like rails 3.1 has simply dropped the :scope that rails 3.0 supplied to t() for views... it continues to provide the :scope for partials. Does anyone know if this is an intentional change?
[20:40:31] mcr: gem has installed rails 3.2.x for me... I need to reproduce a bug in a 3.1.11 application... how to I convince "rails new" to make me a tree based upon 3.1.x?
[21:41:11] mcr: uhm. depends upon your background.
[21:41:48] mcr: if you speak perl, have written some CGIs in shell, hacked some stuff without a framework, then yes.
[21:42:30] mcr: if you haven't a clue what a framework does for you, because you don't understand the transactional nature of HTTP, then the magic that ruby provides get confused with the magic that the rails framework provides.
[21:43:52] mcr: Megamind, I started with Agile Web Development with Rails in 2007. I didn't know ruby. I did the tutorial. But, I started HTTP stuff in 1993.
[22:35:46] mcr: mehwork so you've only used crappy IDEs then? And I guess, not even Emacs.
[22:37:47] mcr: mehwork, vim is an IDE.
[22:48:32] mcr: mehwork, anything that knows that kind of code you are writing and colourizes it, and then lets you compile it with a keystroke is an IDE.
[22:48:58] mcr: mehwork: vim is huge bloated code. It's bigger than emacs.
[22:50:55] mcr: ACTION uses BSD vi.

2012-10-01

[20:53:44] mcr: I have a problem with a model where I allocate a resource. A race condition means that two users can get that resource. Okay, I need a pessimistic lock. Good. How do I write the test case with rspec?
[21:12:48] mcr: so, active_model_serializers goes to/from json, without making me turn my model into a hash first?

2012-09-25

[14:05:21] mcr: given: obiwan-[~/C/hydra/t5219] mcr 3038 %rake routes | grep cancel
[14:05:22] mcr: why: ActionController::RoutingError (No route matches "/fr/account/orders/4203/cancel")... ? my eyes must be missing something.
[14:05:22] mcr: cancel_account_order POST (/:locale)/account/orders/:id/cancel(.:format) {:locale=>/en|fr/, :action=>"cancel", :controller=>"orders"}
[14:07:14] mcr: certainty: Started POST "/fr/account/orders/4203/cancel" for ::ffff:127.0.0.1
[14:08:37] mcr: worse... it *did* work at some point. and... rspec controller and view and routing tests skip routes.rb. (well. routing tests seem to use it, but not the same way Test::Unit routing tests did...)
[14:09:04] mcr: certainty: yup, restarted many times since.. this broke a week ago or more, finally trying to fix this.
[14:11:20] mcr: yeah, same with en. back to puzzling then.
[14:14:55] mcr: certainty, I might have to try that. I know about where it worked, cause I wrote a spec/routing/order_routing_spec.rb test for it, and *it still passes*
[14:15:24] mcr: gonna try to write an rspec integration test. I used to write lots of those on Test::Unit days.
[14:20:14] mcr: oh, my routing tests were get: !
[14:20:42] mcr: nope that didn't fail.
[14:21:17] mcr: this is why I'm skeptical about the routing specs... that should have failed given that I was GET'ing a POST resource. (Of course, you can simulate POST with GET + params, so... maybe it's not all wrong)
[15:47:10] mcr: certainty, found it! my form_for was defaulting to PUT (via _method=>'put'), I probably hacked the HTML and then refactored it. alas the rails server console doesn't log the implied action, but the literal HTTP method.
[15:47:36] mcr: too bad that form_for :action=> doesn't set :method based upon routes.rb... and that spec/routing tests seem to bypass the method check completely.
[15:48:17] mcr: I changed the method to :delete anyway, since really that's what cancel does, but it doesn't actually remove the order, so I might still want to use the D-CRUD method elsewhere.

2012-09-23

[20:57:15] mcr: I'm struggling to get cancan to authorize access for any owner of a has-many-and-belongs-to object (implemented with :through).
[20:58:17] mcr: hmt, I didn't know that acronym. thanks.
[21:00:40] mcr: seems like: can [:read, :update], Student do |student| student.guardians.inject(false) { |result, guard| result ||= (guard.id == guardian.id) } end
[21:00:45] mcr: ought to be what I want.
[21:01:18] mcr: (guardian is passed into the initialize)
[21:02:14] mcr: "The accessible_by call cannot be used with a block 'can' definition."