<FromGitter>
<splattael> scratch that! I am sooo yesterday :-( It's already has been fixed in 0.18.5 which I did not notice.
<FromGitter>
<splattael> Sorry for the noise :-(
<FromGitter>
<hkochev> Hello All, I just wonder if Crystal community has bang! method convention inherited from ruby for the methods that could raise or change state
soveran has joined #crystal-lang
soveran has quit [Changing host]
soveran has joined #crystal-lang
<jhass>
@hkochev mostly so. Note that Ruby core convention is pretty strict on this: if there's a bang method there's a non-bang version that returns a copy. If there isn't a bang method you can not expect a non-bang method to not modify state
<jhass>
same holds true in crystal
splattael has joined #crystal-lang
<FromGitter>
<hkochev> Got it, thanks
unshadow has joined #crystal-lang
soveran has quit [Remote host closed the connection]
pawnbox has quit [Remote host closed the connection]
<RX14>
@hkochev: the convention is the same but I see a lot less bang methods, due to immutability being more prevalent in crystal than ruby.
pawnbox has joined #crystal-lang
soveran has joined #crystal-lang
soveran has quit [Remote host closed the connection]
trapped has joined #crystal-lang
<unshadow>
Did anyone tried to do bindings for Krb5 or GSSAPI ?
<BlaXpirit>
no
pawnbox has quit [Remote host closed the connection]
pawnbox has joined #crystal-lang
pawnbox has quit [Remote host closed the connection]
pawnbox has joined #crystal-lang
<unshadow>
using FFI in Ruby a binding looks like this: 'attach_function :krb5_init_context, [:pointer], :uint' is it right to assume this is the way in crystal : 'fun krb5_init_context(buff : UInt8*) : UInt32' ?
<unshadow>
Or should I use Pointer(UInt8) ?
<BlaXpirit>
unshadow, same thing. but 'UInt8*' syntax is not always allowed
<BlaXpirit>
and for futureproofing might be better to use : LibC::UInt
<FromGitter>
<bcardiff> Regarding [ ], thanks. Yet it seems i will move the tagged deploys to a custom script since travis is ignoring it is doing the build over a tagged commit
<FromGitter>
<bcardiff> (and the update of the s3 object to perform a redirect can't be done on the travis deployment s3 provider)
<FromGitter>
<bcardiff> I will end up with a CURRENT_TAG=`git describe --tags --exact-match 2> /dev/null` and see if i can get the tag.
<FromGitter>
<bcardiff> Any issue / input with that?
<jhass>
bcardiff: I think the cleanest approach is indeed adding another job for it, that then can also bail before doing any work on $TRAVIS_PULL_REQUEST set. bin/deploy sounds good to me, you should be able to call bin/ci within it to build the docs
<jhass>
and you can install the awscli tool within it
<jhass>
bcardiff: or actually I wouldn't even mind adding it as command to bin/ci
<FromGitter>
<bcardiff> I understand that another job should be indeed adding one more entry to the matrix with a variable that will scope the script to run in order to just do make docs && deploy.
<FromGitter>
<bcardiff> Is that what you mean? because otherwise by job you mean just a script to be executed after the current ci script
<jhass>
yes, the former
<FromGitter>
<bcardiff> in that case I think is better to use after_success script in travis for all docs deployment. should be 10 bash lines (after I get them right :-$) and skip unless master or tag. will be easier to customize
willl has joined #crystal-lang
<jhass>
I'll hack up my approach for your judgement anyway ;)
paulcsmi_ has quit [Read error: Connection reset by peer]
paulcsmith_ has joined #crystal-lang
<FromGitter>
<bcardiff> or at least my abilities ;-) I will push one more try for tagged commits as I said.
<FromGitter>
<bcardiff> I trust you have strong and good opinions to organize this better, but I don't want to force you do it.
<crystal-gh>
[crystal] bcardiff pushed 1 new commit to master: https://git.io/voFZ2
<crystal-gh>
crystal/master 23a1c5a Brian J. Cardiff: doc deploy of tagged commits moved to after_success script
<jhass>
bcardiff: https://p.jhass.eu/3u.diff would be my proposal roughly, of course I didn't test any of this, might not need/want some of the eval's, I don't remember
<jhass>
afk for 15 minutes or so
trapped has quit [Read error: Connection reset by peer]
<RX14>
jhass, i wish people created nice test suites I could streal for things like multipart
<FromGitter>
<deepj> What’s a bit strange. Any thoughs?
<FromGitter>
<cjgajard> ** remote: Outdated shard.lock (kemal requirements changed). Please run shards update instead. ** ⏎ your project shards (kemal) never got installed...
<jhass>
well, did you run shards update locally and commit that?
pawnbox has joined #crystal-lang
<FromGitter>
<deepj> @jhass I found the problem. I mistakely defined version of Kemal in `shards.yml` as `branch: v0.13.0` instead of `version: 0.13.0`. Interesting, there was no problem with that on local
<jhass>
btw, time to delete gh-pages ?
datanoise has joined #crystal-lang
datanoise has quit [Client Quit]
datanoise has joined #crystal-lang
phoffer has joined #crystal-lang
<maxpowa>
anyone else having issues with crystal-lang.org/api pages?
<FromGitter>
<bcardiff> we moved to /api/{version} /api/master
<maxpowa>
oh now they have the version in the URL
<maxpowa>
it should default to the main version
<BlaXpirit>
it should
<FromGitter>
<bcardiff> /api has a redirect to /api/{latest-tag}
<BlaXpirit>
bcardiff, that should not be a permanent redirect probably
<BlaXpirit>
and /api/* might be nice to redirect, not just the root
<maxpowa>
just add another for /api/([^/]+.html) to /api/{latest-tag}/{1} assuming you can do regex
<maxpowa>
* might be problematic
<BlaXpirit>
it's dangerous to play with these things, SEO will suffer :(
<maxpowa>
with looping redirects and stuff
phoffer has quit [Client Quit]
<maxpowa>
you're already fighting with companies that sell crystal strings lol
<BlaXpirit>
bcardiff, but seriously, crystal api documentation is not permanently moving to /api/0.18.6/. that's exactly how you get google to show old links to everything
<BlaXpirit>
in fact, it is a dangerous idea overall to have versioned links at all
<BlaXpirit>
nah, I'm overstating it
<FromGitter>
<bcardiff> I am not sure if regex will work, the backend is s3. I will need to check some s3 things (and now I am over a really bad 3g...)
<BlaXpirit>
another idea I just had - the link should probably be just 0.18. each point release would just update the 0.18 pages. the changes to docs are supposed to be very minor
Oliphaunte has quit [Remote host closed the connection]
<BlaXpirit>
and I just noticed that /api temporarily-> /api/ permanently-> /api/0.18.6
<crystal-gh>
[crystal] jhass pushed 1 new commit to master: https://git.io/voFST
<crystal-gh>
crystal/master f02b85a Jonne Haß: update Contributing.md for the repo split [ci skip]
paulcsmith_ has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
spiderpug has quit [Quit: Leaving.]
Philpax has joined #crystal-lang
<FromGitter>
<jwaldrip> @bcardiff do you know if @ysbaddaden still maintains the crystal shards site/
Philpax has quit [Ping timeout: 246 seconds]
rolha has quit [Ping timeout: 260 seconds]
<FromGitter>
<bcardiff> I don't have other info than his public profile @jwaldrip
A124 has quit [Quit: '']
Philpax has joined #crystal-lang
<maxpowa>
Also, all the links to the api pages from the docs are broken