<FromGitter>
<Blacksmoke16> `SELECT name, address FROM contacts s JOIN settings r ON s.id = r.user_id WHERE (r.address = ? AND s.age >= ? AND s.name = ?);`
<wuehlmaus>
apropos hedron, anybody gets the examples running? i just want to exclude that it's my fault or my os that is the problem
<wuehlmaus>
as i am running crystal on macosx i think that usage of qt is not working. does that leave me with hedron or are there other gui frameworks on osx?
<wuehlmaus>
and libui doesn't do it for me neither :-(
ashirase has quit [Ping timeout: 245 seconds]
ashirase has joined #crystal-lang
Jenz has joined #crystal-lang
<Jenz>
How come crystal doesn't infer the type (Bool) of Enumerable#{all, none, any, one}?
<FromGitter>
<bew> But it doesn't work either because the block is expanded in the struct' and you're not allowed to reference the other ivars in this context
<Jenz>
I'm just wondering *why* the type isn't inferred
<FromGitter>
<bew> Because the return type is not specified on the declaration of these methods
<Jenz>
`true` and `false` are literal Bools, and according to the error messages rule 1, type should be inferred
<FromGitter>
<bew> Hum no, the compiler doesn't check inside the methods
<Jenz>
Huh
<Jenz>
I'd thought it would, but ok, thanks
<Jenz>
So if I have: `def returns_true; true; end`, and in a class, `@foo = returns_true()`, the compiler won't infer the type.
<FromGitter>
<bew> ... I did a few tests and it absolutely not behaved like I thought it would... I guess the compiler has some weird corner case to either make something work or not..
Jenz has quit [Ping timeout: 255 seconds]
<FromGitter>
<bew> Yeah
DTZUZO has joined #crystal-lang
<FromGitter>
<dscottboggs_gitlab> @Blacksmoke16 ooof that's fun :/
<FromGitter>
<dscottboggs_gitlab> ...so why is `Process::Status#exit_code` an `Int32` when it can only ever be a `UInt8`?
laaron has quit [Remote host closed the connection]
laaron has joined #crystal-lang
hightower3 has joined #crystal-lang
_whitelogger has joined #crystal-lang
<oprypin>
dscottboggs_gitlab, im not sure thats true
<oprypin>
what do u think about C's `int main`
<oprypin>
and im certain on Windows there are status codes like minus a million reserved by the system
<FromGitter>
<dscottboggs_gitlab> yeah I thought windows might have something to do with it...but you're right, it is `int main` not `unsigned char main`
<FromGitter>
<dscottboggs_gitlab> returning 300 from a c program returns 44 to the shell, and -5 returns 251 to the shell...so I must be just thinking of a posix shell limitation as opposed to a normal C thing.
<oprypin>
dscottboggs_gitlab, you seem correct there. another confirmation:
<FromGitter>
<Blacksmoke16> @vladfaust Yes but the former is reusable, i.e. could make that query and reuse it, similar to like `HasName.new("name")`. Also the values are all type safe, so if you pass a Float to the `age` `gte` part, it wouldnt compile
<FromGitter>
<Blacksmoke16> which i think is worth the extra bit of typing
<FromGitter>
<Blacksmoke16> are some other differences as welll
<FromGitter>
<Blacksmoke16> @dscottboggs_gitlab yea, had to check if return type was Nil, and if it was map it to a private class that i check for instead of nil
<FromGitter>
<Blacksmoke16> so given you have a complex query, with conditionaly adding additional requirements, could just do `Model.match(IsActive.new("abcd"))`
<FromGitter>
<j8r> mps how goes Crystal 0.28.0 on Alpine? Any help needed?
<mps>
j8r: I'm trying to build static apk package of previous version (0.27.2) which is needed to build 0.28.0
<mps>
i.e. in one build pass to build both, static and shared apk
<mps>
actually I did this, but now wrestling with package manager to make it's conflict and depends options to work correctly
<mps>
for the test I've built 0.28.0 yesterday, and it worked nice
Jenz has joined #crystal-lang
DmitryBochkarev has quit [Ping timeout: 246 seconds]
DmitryBochkarev has joined #crystal-lang
DmitryBochkarev has quit [Remote host closed the connection]
DmitryBochkarev has joined #crystal-lang
Jenz has quit [Ping timeout: 268 seconds]
<FromGitter>
<NinoFurger> Hallo. I just updated crystal and I'm currently trying to build scry, but i get the following error. Any idea why? ⏎ ⏎ `Error target scry failed to compile: ⏎ Error in src/scry.cr:4: while requiring "./scry/context" ⏎ ... [https://gitter.im/crystal-lang/crystal?at=5cbc901ee416b84519b812cd]
<FromGitter>
<Blacksmoke16> scry prob needs updated for 0.28.0
<FromGitter>
<j8r> you may have this issue because you run 0.27.2?
<mps>
j8r: yes, the issue is in 0.27.2, not 0.28.0
<mps>
sorry if I wasn't clear in my previous messages
<FromGitter>
<j8r> so that's totally normal
<mps>
I think 0.27.2 fix is not important because we will have 0.28.0 in repository soon, I hope
<FromGitter>
<j8r> the specs are supposed to run in 0.28.0?
<mps>
in 0.28.0 all works fine
<mps>
maybe you didn't read my previous messages, I'm now building two versions of 0.27.2, one static to have bootstrap version apk for 0.28.0 and old 0.27.2 normal shared apk
<mps>
with this new bootstrap version we will not need anymore to have tar.gz in dev.alpinelinux.org, it will be in 'normal' repo. this is my plan
<mps>
what do you think of it? is it ok or you have some suggestions
japanoise has quit [Ping timeout: 252 seconds]
Jenz has quit [Ping timeout: 245 seconds]
<FromGitter>
<j8r> I thought you were building 0.27.2 on 3.9
<FromGitter>
<j8r> then 0.28.0 on edge
<FromGitter>
<j8r> what's done with rust or go?
<mps>
in 3.9 is 0.27.0 version
<mps>
0.27.2 is in edge, currently
<mps>
because edge will become next stable (3.10) I want to upgrade crystal to 0.28.0 to be version in next stable