Activity Graph

Page 1 of 34 | Next »


[00:05:18] FailBit: I could not be passionate about JavaScript until ES6
[00:05:24] FailBit: and then I loved writing it, surprisingly
[07:04:46] FailBit: *.net *.split
[21:07:18] FailBit: has left #RubyonRails-offtopic: ("WeeChat 1.5")


[01:43:24] FailBit: has joined #RubyonRails-offtopic
[01:49:03] FailBit: Radar: I'm taking your advice and trying skylight for a month
[01:49:15] FailBit: can't afford to keep it up tho @ 3000RPM
[01:53:06] FailBit: Radar: /dev/null
[01:53:24] FailBit: we currently use new relic free tier and it's god awful
[02:05:04] FailBit: >Typical response: 46ms
[02:08:08] FailBit: it has some weird behavior about our elasticsearch stuff
[02:51:03] FailBit: I found many of them
[02:51:08] FailBit: it didn't take long
[02:51:12] FailBit: wow newrelic sucks
[02:56:25] FailBit: now I just have to hope I can fix them all before my month is up
[02:56:49] FailBit: and then it's back to shit
[02:57:02] FailBit: unless I self host sentry on a separate box or something
[03:01:05] FailBit: maybe he is helpful
[03:25:04] FailBit: rails on memedows
[12:24:08] FailBit:
[12:24:15] FailBit: god damn I love skylight
[21:46:27] FailBit: lod'samoney
[21:46:46] FailBit: only if you have a big enough server
[21:47:02] FailBit: it's at its limits right now
[21:47:14] FailBit: it'll occasionally peak at 5k and can't really go over
[21:47:26] FailBit: without getting horrid queue waiting times for users
[23:25:48] FailBit: oh shit lads
[23:26:00] FailBit: IT'S HAPPENING


[01:48:49] FailBit: has left #RubyOnRails: ("WeeChat 0.4.2")


[22:25:02] FailBit: has joined #RubyOnRails
[22:25:30] FailBit: how do I post form data using js fetch() to a rails post endpoint


[19:25:47] FailBit: has joined #RubyOnRails
[19:26:13] FailBit: how can I convert this to upsert?
[19:26:56] FailBit: some profiling revealed that this was one of the slowest constants of my site
[19:27:39] FailBit: 9.5 is installed on production
[19:27:42] FailBit: so it has the new upsert stuff
[19:28:31] FailBit: the query itself is not slow per se, and yes the indexes are there
[19:28:44] FailBit: it's actually time spent in ruby (!)
[19:29:00] FailBit: fine, fine, I'll post the explain
[19:30:34] FailBit:
[19:33:49] FailBit:
[19:33:55] FailBit: there's your explain analyze
[19:34:12] FailBit: this is why I believe it is ruby
[19:34:27] FailBit: 0.3 + 0.5 ≠ 5.5
[19:35:42] FailBit: I assume it is slow on new records but I don't particularly care about that usecase because it will only happen once per new user ip
[19:35:57] FailBit: whereas this will happen once every pageload
[19:41:47] FailBit: I just want something that will minimize the amount of time spent in ruby and that's certainly not it (exceptions are slow)
[19:42:08] FailBit: I will try the on conflict stuff
[19:44:51] FailBit: that needs a constraint of some sort
[19:44:59] FailBit: but I have to satisfy it using two columns
[20:04:00] FailBit: I recommend not containerizing until you need to deploy to production
[20:04:07] FailBit: it's a bit of a headache
[20:25:11] FailBit: has left #RubyOnRails: ("WeeChat 0.4.2")


[01:57:55] FailBit: what's your prod schema
[01:59:20] FailBit: that message is implying that your prod schema has some weird column in it named transaction
[01:59:46] FailBit: that or there's a conditional association macro somewhere like has_one :transaction
[02:04:58] FailBit: that's how it's always been
[02:22:36] FailBit: has left #RubyOnRails: ("WeeChat 0.4.2")


[13:07:54] FailBit: has joined #RubyOnRails
[13:09:53] FailBit: happening status: it's
[13:12:53] FailBit: party like it's 2016
[13:16:24] FailBit: bah that's schrodinger
[19:27:55] FailBit: :'this is a symbol'
[19:28:27] FailBit: if you want to use symbols, activesupport has Hash#symbolize_keys
[19:51:23] FailBit: that's bad UX from the start
[19:52:15] FailBit: use the default html number input
[19:54:01] FailBit: terrabl: that's the point
[19:54:40] FailBit: some user agents will support it, some will not, but not doing that users expect on scroll is a very bad idea
[19:54:46] FailBit: s/that/what
[19:55:28] FailBit: it is up to the user agent to make it scrollable
[19:56:16] FailBit: something in particular that I _really_ hate is websites that add their own scrollbar
[19:56:31] FailBit: it annoys me to no end
[20:36:19] FailBit: just run it on ganoo linix like a Normal Person™
[20:36:26] FailBit: or BSD if you're a masochist
[20:36:56] FailBit: remember, virtualization exists
[21:28:18] FailBit: o boi check out my mixtape bra
[21:32:03] FailBit: did you add a command to write one?
[21:32:53] FailBit: access_log /var/log/nginx/access.log
[21:38:14] FailBit: >not sudo -s
[21:39:22] FailBit: pretty sure the main one is always /etc/nginx/nginx.conf
[21:39:28] FailBit: which can include other files from elsewhere
[21:40:43] FailBit: I'm pretty sure it has to be nginx.conf
[21:51:35] FailBit:
[22:41:13] FailBit: puppet is da best
[23:36:34] FailBit: what's the simple_form gem do
[23:36:36] FailBit: ACTION googles


[22:11:27] FailBit: has joined #RubyOnRails
[22:13:15] FailBit: has joined #RubyonRails-offtopic
[22:16:47] FailBit: mixtape is fire
[22:19:08] FailBit: well no not always
[22:19:09] FailBit:
[22:20:33] FailBit: it took me several minutes to figure out where that was coming from
[22:23:36] FailBit: that particular instance was a coffeescript syntax error that I introduced when writing !==
[22:23:44] FailBit: by the way coffeescript sucsk
[22:24:49] FailBit: "so much more features" is incorrect because they actually have the same amount of features
[22:24:58] FailBit: coffeescript is just javascript underneath
[22:25:26] FailBit: also if you count bugs as features coffeescript has a lot of those
[22:25:43] FailBit: can't hear sarcasm over the internet sorry
[22:27:47] FailBit: if I say "X sucks" then you know it's an opinion anyways
[22:46:58] FailBit: has left #RubyonRails-offtopic: ("WeeChat 0.4.2")
[22:47:20] FailBit: has left #RubyOnRails: ("WeeChat 0.4.2")


[00:21:13] FailBit: peek, maybe
[00:21:27] FailBit:
[00:22:47] FailBit: you can't.
[00:24:49] FailBit: there's also rack-mini-profiler if you're looking for a profiler
[01:14:09] FailBit: smathy: nothing substantially different (all the options are the same, just the order is a little different)
[01:16:42] FailBit: it isn't older?
[01:22:55] FailBit: smathy: found it
[01:22:58] FailBit: it was the remotipart gem
[01:29:58] FailBit: one of my gems requires ancient sinatra
[01:30:55] FailBit: I'll just switch to goworker
[01:32:27] FailBit: I'm just gonna pin resque at its old non broken version
[01:36:16] FailBit: bundler downgraded my version of sinatra
[01:40:10] FailBit: it was because of an incompat with rack
[01:40:18] FailBit: there's no release with a compatible rack
[01:40:25] FailBit: so I have to use the git version
[01:41:05] FailBit: the current sinatra has its rack locked at ~> 1.5
[01:41:11] FailBit: git sinatra has it at 2.0
[01:41:30] FailBit: rails 5 requires 2.0
[01:46:46] FailBit: has left #RubyOnRails: ("WeeChat 0.4.2")


[21:28:54] FailBit: has joined #RubyOnRails
[21:29:31] FailBit: starting our migration to rails 5. installed r5 rc1 and ruby 2.3.1. getting this message when running tests: 'No such middleware to insert after: ActionDispatch::ParamsParser'
[21:29:45] FailBit: a grep for ParamsParser in our code... nothing
[21:29:52] FailBit: backtrace provided is completely unhelpful
[21:30:59] FailBit:
[21:43:28] FailBit: drale2k: do you have a rescue_from in your controller
[21:44:06] FailBit: find throws an exception, annoyingly
[21:45:07] FailBit: any help with my migration woes
[21:46:18] FailBit: thanks that's exactly what I needed to solve my issue
[21:55:54] FailBit: forgot that one
[22:07:39] FailBit: I'm going to be afk for a while but if anyone has any suggestions for when I'm back please voice them
[23:18:42] FailBit: >not being a pain in the ass


[11:28:18] FailBit: has joined #RubyOnRails
[11:28:57] FailBit: ravi_sun66: but how much experience not using abbreviations when you talk
[11:48:02] FailBit: I'm having some trouble with sequences and my schema
[11:48:47] FailBit: I first tried the pg_sequencer gem and it got the new sequence but it wouldn't install the sequence in the test env
[11:49:29] FailBit: then I tried using structure.sql and that also did not seem to work correctly
[11:50:35] FailBit: as in, I loaded the structure in the test env, then pg_dump'd it out again and the sequence I had defined was missing
[12:14:06] FailBit: scientists are well known for their inability to use anything other than python
[12:19:48] FailBit: also applicable: matlab code
[12:46:05] FailBit: >being this reactionary
[13:10:43] FailBit: Tichodroma: don't use mongo
[13:10:46] FailBit: just don't
[13:10:50] FailBit: don't do it
[13:46:34] FailBit: don't do it
[13:50:45] FailBit: ACTION gives VeryBewitching a user agent
[13:52:20] FailBit: time to start over again
[15:48:45] FailBit: has left #RubyOnRails: ("WeeChat 0.4.2")


[19:56:32] FailBit: how do I make rails reload a model after saving it
[19:56:38] FailBit: actually, just after creating it
[19:56:56] FailBit: if there's a better way than after_create { reload }
[19:57:31] FailBit: specifically, I now have a sequence set to be the default value for one of my columns and when the model is created
[19:57:33] FailBit: rails assigns the id
[19:57:38] FailBit: but doesn't update the other column (?)
[20:00:12] FailBit: did you bundle after gem'ing rspec?
[20:00:19] FailBit: if you did, rake should detect its tasks
[20:04:33] FailBit: t.integer "id_number", default: "nextval('image_id_number_seq'::regclass)"
[20:04:42] FailBit: when saving a model with this
[20:04:54] FailBit: the model does not seem to be updated with the new value that the database assigns
[20:05:02] FailBit: only with its id
[20:06:05] FailBit: so now I have after_create :reload in the model
[20:08:39] FailBit: hm, maybe I was under the assumption that the database returned that column on save
[20:11:06] FailBit: fox_mulder_cp: because of the past idiot dba
[20:11:11] FailBit: there are two id columns
[20:11:14] FailBit: one internal
[20:11:16] FailBit: one user facing
[20:11:36] FailBit: previously we had extremely race-condition prone logic to assign the public facing id
[20:11:50] FailBit: I moved this to a sequence but was confused as to why rails wasn't updating the model
[20:13:38] FailBit: I don't have a lot of wiggle room here
[20:13:46] FailBit: that would involve several hours of downtime to migrate everything
[20:14:46] FailBit: it's abigint
[20:14:54] FailBit: I doubt we will ever run out
[20:15:27] FailBit: we moved from mongo
[20:37:05] FailBit: has left #RubyOnRails: ("WeeChat 0.4.2")


[01:43:22] FailBit: but does it scale
[02:18:59] FailBit: im trying to play franz liszt
[02:20:16] FailBit: has joined #ruby
[03:50:37] FailBit: has left #RubyOnRails: ("WeeChat 0.4.2")
[03:50:43] FailBit: has left #ruby: ("WeeChat 0.4.2")


[01:33:57] FailBit: yay I got my first segfauly
[01:53:51] FailBit: I wasn't either
[02:03:07] FailBit: has left #RubyOnRails: ("WeeChat 0.4.2")
[18:04:28] FailBit: has joined #RubyOnRails
[19:43:32] FailBit: arup_r: just run it
[19:43:46] FailBit: it will affect performance somewhat for a bit
[19:43:57] FailBit: it's better to version your cache strings
[19:50:13] FailBit: it never invalidates
[19:50:32] FailBit: it stays in there until it's evicted
[19:50:35] FailBit: if it never expires
[19:50:59] FailBit: eviction could be caused by you flushing the cache or by the key being unused
[19:55:34] FailBit: you use a cache key based on the object's updated_at time
[19:56:19] FailBit: cache "someobject-v1-#{@someobject.updated_at.to_i}" do ... end
[20:01:40] FailBit: always remember
[20:01:51] FailBit: caching is for things that are expensive to calculate
[20:02:11] FailBit: if you just have a bunch of string interpolation it probably won't make a difference
[20:39:58] FailBit: a search engine
[20:56:41] FailBit: aren't bitmaps fun
[21:30:14] FailBit: jpcody, unicorn doesn't use threads, does it?
[21:30:24] FailBit: it uses individual processes
[21:31:33] FailBit: only if you have one appserver
[21:31:36] FailBit: and it never reforks