<FromGitter>
<Blacksmoke16> anyone notice sublime crystal plugin doesnt auto format anymoer?
<FromGitter>
<drum445> I was just about to say
<FromGitter>
<drum445> Quite sad
<FromGitter>
<drum445> Switched my finance lib to use decimal %ages as it was a bit ropey before, would appreciate people's opinions on the lib. β https://github.com/drum445/financials
<FromGitter>
<Blacksmoke16> %ages?
<FromGitter>
<drum445> percentages
<FromGitter>
<Blacksmoke16> ah
<FromGitter>
<Blacksmoke16> man i miss that auto format
<FromGitter>
<drum445> haha, same
<FromGitter>
<Blacksmoke16> hopefully get patched soon
<FromGitter>
<drum445> I wonder why it went
<FromGitter>
<drum445> Never seen that issue during version changes before
<FromGitter>
<drum445> Also, have you see the compiler spit out "You found a bug in the compiler...." for just normal syntax errors
<FromGitter>
<Blacksmoke16> havent seen that one
<FromGitter>
<Blacksmoke16> was a pr to refactor formatting, might have changed something that broke the formatter
<FromGitter>
<Blacksmoke16> because under the hood it was just calling crystal tool format
<FromGitter>
<drum445> ah cool
<FromGitter>
<drum445> hope it gets fixed
<FromGitter>
<Blacksmoke16> yea its using a command that got removed now
<FromGitter>
<girng> i'm doing player speech bubbles, need to detect `!message ` β actually, i think this is a remnant from coding ins javascript. i think "!" would have worked
<FromGitter>
<girng> yah prob better, cause empty string will not cause an indexError
<FromGitter>
<girng> thanks
<FromGitter>
<Blacksmoke16> np
<FromGitter>
<dscottboggs_gitlab> Is there a recommended way to set up the appropriate state to run the specs Crystal itself. TBH that's the biggest thing keeping me from contributing to Crystal right now, I can't run the specs on the unmodified repo, let alone to check my changes. I've been trying `make clean && make spec` inside the `jrei/crystal-alpine` docker image at the 0.27.1 branch commit, but there's clearly some specs which
<FromGitter>
... rely on some sort of preconfigured state.
<FromGitter>
<Blacksmoke16> really?
<FromGitter>
<Blacksmoke16> i just do a make clean && make && make std_spec
<FromGitter>
<HarrisonB> And here's the error: `Error in line 5: instance variable '@server' of TestServer was not initialized directly in all of the 'initialize' methods, rendering it nilable. Indirect initialization is not supported.`
<FromGitter>
<dscottboggs_gitlab> looks to me like it's initialized properly..
<FromGitter>
<Blacksmoke16> well
<FromGitter>
<dscottboggs_gitlab> maybe try it without the `self.`?
<FromGitter>
<HarrisonB> If you remove the `self.`, it creates a local
<FromGitter>
<dscottboggs_gitlab> oof, yeah, you're right
<FromGitter>
<Blacksmoke16> ah i see the issue
<FromGitter>
<dscottboggs_gitlab> what about if you do `@last_request`?
<FromGitter>
<Blacksmoke16> `self` is `TestServer`
<FromGitter>
<Blacksmoke16> yea gotta do that
<FromGitter>
<Blacksmoke16> `self.last_request` calls the setter for that ivar
<FromGitter>
<Blacksmoke16> which would make it not set in the initalizer
<FromGitter>
<Blacksmoke16> `@` sets the ivar directly
<FromGitter>
<dscottboggs_gitlab> but that property is nillable?
<FromGitter>
<Blacksmoke16> :thinking:
<FromGitter>
<dscottboggs_gitlab> just got a bunch of failures and errors on `make std_spec`.... I'm not sure why I'm using the alpine image, I'll try `FROM crystal-lang/crystal`
<FromGitter>
<Blacksmoke16> oh you running it in docker?
<FromGitter>
<dscottboggs_gitlab> yes
<FromGitter>
<HarrisonB> ah so the initializer can't call any methods or the ivars might be have been mutated
<FromGitter>
<Blacksmoke16> make sure you have the deps required
<FromGitter>
<Blacksmoke16> like llvm and stuff
<FromGitter>
<dscottboggs_gitlab> Don't really want to run it on my local system, I don't know what kind of things the specs are gonna do
<FromGitter>
<Blacksmoke16> prob something like that, hence the error message `not initialized directly`
<FromGitter>
<Blacksmoke16> yea cant say i tried running it via docker
<FromGitter>
<Blacksmoke16> should be doable tho since it runs on CI
Vexatos has quit [Quit: ZNC Quit]
RX14 has quit [Quit: Fuck this shit, I'm out!]
<FromGitter>
<dscottboggs_gitlab> ugh it's gonna have to wait til tomorrow I'm beat
<FromGitter>
<dscottboggs_gitlab> βοΈ good night
<FromGitter>
<Blacksmoke16> o/
RX14 has joined #crystal-lang
_whitelogger has joined #crystal-lang
Vexatos has joined #crystal-lang
<FromGitter>
<girng> gn
marmotini_ has joined #crystal-lang
gangstacat has quit [Quit: Δis!]
_whitelogger has joined #crystal-lang
marmotini_ has quit [Ping timeout: 246 seconds]
gangstacat has joined #crystal-lang
marmotini_ has joined #crystal-lang
marmotini has joined #crystal-lang
marmotini_ has quit [Ping timeout: 240 seconds]
<FromGitter>
<bew> @dscottboggs_gitlab I also can't build the specs (not enough ram), but you can usually run individual spec files, they should be self contained! And then I let the CI do the full tests
<FromGitter>
<loloof64> Thanks in advance for any help, even for a maybe stupid issue like this
<FromGitter>
<r00ster91> do you get this error when you run `shards`?
<FromGitter>
<loloof64> yes
<FromGitter>
<loloof64> shards build
<FromGitter>
<r00ster91> hmm try removing the `version`
<FromGitter>
<loloof64> Yes, right now
<FromGitter>
<r00ster91> also try only `shards`, not `shards build`
<FromGitter>
<loloof64> shards build works without the version π Thank you π
<FromGitter>
<r00ster91> oh nice. So it looks like the README is a bit outdated
<FromGitter>
<loloof64> Yes, indeed π
<FromGitter>
<loloof64> Also one line during the fetch was strangely : `Installing hedron (0.1.0 at HEAD)`
<FromGitter>
<loloof64> (HEAD !)
<FromGitter>
<loloof64> Maybe I got the error because of recent modifications, even if the version is the same
<FromGitter>
<Blacksmoke16> @loloof64 the issue is they dont have any releases
<FromGitter>
<Blacksmoke16> so not using version works since it fetches from master
<FromGitter>
<loloof64> Ok : I understand better π
<FromGitter>
<loloof64> Not Crystal-specific question, but a generic C/C++/Crystal one : I have the GTK3 developpement files installed, but it seems I still have to link them (got many errors like `undefined reference to « gtk_container_set_border_width »`). As I built libUI as static and as a linux User, does that mean that I have to add an -ld section to my `shards build` command ?
<FromGitter>
<Blacksmoke16> :shrug:
<FromGitter>
<Blacksmoke16> im not too familiar with the C bindings
<FromGitter>
<loloof64> Indeed, I am targeting a bit high from the beginning, as libUI appears to me to be a great UI to start
<FromGitter>
<loloof64> Unfortunately, its repository lacks a bit of documentation
<FromGitter>
<r00ster91> I would also try the QT5 binding and Glass
<FromGitter>
<loloof64> Yes, if I can't manage with LibUI, i'll go for Qt5. I did not want to because of commercial app restriction (it seems that you have to share your sources if you want to use the LGPL version), even if for now, that can't be a problem. Just in case I would have to do commercial app.
<FromGitter>
<loloof64> Oh I forgot that with QT5, one can program commercial application if the qt5 lib is used as dynamic, and not static
<FromGitter>
<loloof64> So maybe, I could use this last approch
<FromGitter>
<loloof64> If I've well undestood from all sources I've been on so far ...
<FromGitter>
<loloof64> Otherwise, as I am convinced that libUI worth the efforts, it seems, from my last `shards build` error, that the compiler has tried to compile the c code dynamically instead of statically
<FromGitter>
<loloof64> ``` code paste, see link```
<FromGitter>
<loloof64> why the -rdynamic flag ?
<FromGitter>
<r00ster91> try `--static`
<FromGitter>
<loloof64> Thanks : I try it right now π
<FromGitter>
<loloof64> Also, failed, but the flags differs :
<FromGitter>
<j8r> Try dynamic linking, then static
<FromGitter>
<loloof64> Yes, so I'll rebuild libUI as a dynamic lib first π
<FromGitter>
<j8r> Usually the OS has everything for dynamic linking, but sometimes `.a` archives required for static linking are missing
<FromGitter>
<loloof64> I understand π
<FromGitter>
<loloof64> Indeed, I haven't any error, though I hadn't manage yet to show the window. But at least, no binding error this time. Thank you π
DTZUZO has joined #crystal-lang
<mps>
j8r: you mentioned in your message "11:57 j8r| @dscottboggs_gitlab i use jrei/crystal-alpine to contribute". Would you mind to share that url
<FromGitter>
<j8r> on my side their new web UI broke the markdown render
<FromGitter>
<j8r> tried on Falkon and Firefox
<FromGitter>
<j8r> that's annoying because there are mandatory steps for static linking in the README :/
<mps>
right, their web ui is annoying, at least. Thanks for url
jemc has quit [Ping timeout: 245 seconds]
jemc has joined #crystal-lang
jemc has quit [Ping timeout: 245 seconds]
<FromGitter>
<HarrisonB> I can't seem to find any info about this syntax: `HTTP::Headers { "Foo" => "bar" }`. It doesn't seem like it's passing in a Hash nor does it appear to be a block...