jgaskins has quit [Quit: This computer has gone to sleep]
jgaskins has joined #rom-rb
lfox has quit [Quit: ZZZzzz…]
jgaskins has quit [Quit: This computer has gone to sleep]
jgaskins has joined #rom-rb
jgaskins has quit [Quit: This computer has gone to sleep]
jgaskins has joined #rom-rb
<dkubb>
xybre: yeah it deep freezes the object
<solnic>
xybre: morpher uses ast gem (you're gonna see this in other gems used by rom)
jgaskins has quit [Quit: This computer has gone to sleep]
<solnic>
dkubb: hah looks like the build is passing again on ruby-head
<dkubb>
wtf
<solnic>
maybe they fixed it?
<dkubb>
lolruby
<solnic>
totally
<dkubb>
I doubt they would fix it that fast
<solnic>
yeah I know
<dkubb>
was it segfaulting?
<solnic>
lolmri
<solnic>
yes it was a segfault
<solnic>
something around rspec::matcher
<dkubb>
yeah, I'm thinking about switching our work ruby to jruby
<solnic>
not adamantium this time
<dkubb>
I just don't have much confidence in mri these days
<solnic>
same here
<dkubb>
even the recent openssl issues
<dkubb>
if you have a security hole you patch it first, then figure out who to contact upstream. once it's fixed completely upstream and everyone has upgraded (which could take years), only then do you even consider removing the patch
<dkubb>
ruby-core argued against patching openssl to not use insecure defaults and said it was the responsibility of all the users to upgrade openssl if they want sane defaults
<dkubb>
responsible maintainers would do both: patch the hole, and recommend everyone upgrade with nasty warnings whenever you're using an old openssl version
<solnic>
I saw some discussions about it yeah
<dkubb>
some of it could be language barrier issues, but I saw lots of english speakers defending it too
<dkubb>
solnic: mbj was talking about how nice it is to deploy a WAR file with all the dependencies self-contained. the only problem is a WAR (seems to) require tomcat or some other java server. I hear there are ways to package it up into a JAR though.
<solnic>
yeah tomcats and wars ain't nice
<dkubb>
it'd be sweet to bundle everything up into a JAR, test it on staging and then pass the same file to production
<solnic>
besides, it's called WAR, cannot be good
<dkubb>
rather than building everything on every server
<dkubb>
heh
<dkubb>
solnic: if we end up doing it I'll let you know
<solnic>
I'm just waiting for a broad adoption of docker
judofyr_ has quit [Remote host closed the connection]
judofyr has joined #rom-rb
judofyr has quit [Ping timeout: 264 seconds]
judofyr has joined #rom-rb
judofyr_ has joined #rom-rb
judofyr has quit [Ping timeout: 248 seconds]
judofyr_ has quit [Ping timeout: 240 seconds]
mbj has joined #rom-rb
judofyr has joined #rom-rb
<mbj>
solnic: hey
<mbj>
solnic: I'm doing a short break and catching up with discussions.
<mbj>
solnic: RE coercions in coercible / morpher.
<mbj>
I think coercibles features can be reproduced in morpher with less loc.
snusnu1 has quit [Quit: Leaving.]
<mbj>
Also we'd gain the ease of inversible coercions if we use morpher as blueprint.
<mbj>
We could still have a coercible gem, (that would probably be backed by morpher) and presenting a more "coercible" like interface.
<mbj>
I totally agree we need support for '10e10' style literals. But not in the default.
<mbj>
BUT the allowed input formats should be choosable fine grained.
<mbj>
avdi: It seems you also attend Joy of Coding in March? I co host the mutant workshop there. If you wanna talk about ROM / mappers / morpher we should take the chance.
<avdi>
Yep, I'll be there!
develop7 has joined #rom-rb
<mbj>
avdi: nice
judofyr has quit [Remote host closed the connection]