jemc changed the topic of #ponylang to: Welcome! Please check out our Code of Conduct => https://github.com/ponylang/ponyc/blob/master/CODE_OF_CONDUCT.md | Public IRC logs are available => http://irclog.whitequark.org/ponylang | Please consider participating in our mailing lists => https://pony.groups.io/g/pony
so has joined #ponylang
endformationage has quit [Quit: WeeChat 1.9.1]
_whitelogger has joined #ponylang
aturley has quit [Read error: Connection reset by peer]
aturley has joined #ponylang
<Candle> SeanTAllen: Ha! While I believe it conforms to the websocket spec, it does need some tidy-up. I also need to use it to find the usability & functional bugs.
<vaninwagen> I am dreaming of a future where the https://github.com/ponylang/http repo incorporates websocket support, so one can add websocket support to any webserver :)
<Candle> On my (ever growing) list is starting from scratch with HTTP, focusing on library usability, and starting with the client side. Mixing too much server and client bits seems like a recipe for complexity.
<vaninwagen> It does.
<Candle> My initial attempt at client side websocket was to use the server implementation as a library; but there's too much couplng and private elements to make that a viable option.
<vaninwagen> If you have any design ideas or api suggestions, you are more than welcometon join in shaping: https://github.com/ponylang/http
<vaninwagen> We can also start separating client and server side already
<Candle> Sure. I would expect it to actually be three bits. request/response parsing is almost the same. Both sides need to know about request and response objects, so sharing those seems sensible.
Foaly has joined #ponylang
<Candle> I've also got 3? projects that are on hold because the current HTTP implementation seems to send a TCP FIN before it's transmitted the response body. (I'm sure I created a gh issue somewhere...)
<vaninwagen> K, i am putting that at the front of my queue
<Candle> They're all toy projects, so really not a priority, the core works, I just havn't been able to expose the json-over-http interface, I've just exposed the CLI interface!
* Candle goes digging in his IRC history for GH issues...
<vaninwagen> Isnt life a toy project, Candle ?
<Candle> hehe, yeah, I guess so! https://freenode.irclog.whitequark.org/ponylang/2018-02-13 Reading that, I wasn't quite able to get a minimal testcase.
tiwake has quit [Ping timeout: 252 seconds]
tiwake has joined #ponylang
OtakuSenpai has joined #ponylang
tiwake has quit [Ping timeout: 252 seconds]
tiwake has joined #ponylang
X704 has joined #ponylang
tiwake has quit [Ping timeout: 245 seconds]
X704 is now known as tiwake
Foaly has quit [Quit: Now 'mid shadows deep falls blessed sleep.]
SenasOzys has quit [Ping timeout: 260 seconds]
tiwake has quit [Ping timeout: 252 seconds]
OtakuSenpai has quit [Remote host closed the connection]
OtakuSenpai has joined #ponylang
SenasOzys has joined #ponylang
aturley has quit [Quit: aturley]
<malthe> SeanTAllen: there seems to be some mac errors with travis: https://api.travis-ci.org/v3/job/432893171/log.txt
<malthe> ultimately "/bin/sh: clang-6.0: command not found"
acarrico has joined #ponylang
dx_ob has joined #ponylang
OtakuSenpai has quit [Ping timeout: 252 seconds]
dx_ob has quit [Quit: Leaving]
PrsPrsBK has quit [Read error: Connection reset by peer]
PrsPrsBK has joined #ponylang
<vaninwagen> I merged a PR to fix it
<vaninwagen> Should work now if you rebase your branch on current master
OtakuSenpai has joined #ponylang
<SeanTAllen> vaninwagen: i take it a llvm-6 is now available from homebrew?
<vaninwagen> Yes the package is called llvm@6 , the plain llvm is actually llvm 7
<SeanTAllen> yup I see llvm@6
<SeanTAllen> yeah I had left a note about that in the comments on it
<SeanTAllen> that it would have to be switched
<vaninwagen> That was a very nice note, i was happy about it
<SeanTAllen> i thought it would come in handy
OtakuSenpai has quit [Ping timeout: 252 seconds]
Foaly has joined #ponylang
SenasOzys has quit [Remote host closed the connection]
SenasOzys has joined #ponylang
OtakuSenpai has joined #ponylang
OtakuSenpai has quit [Remote host closed the connection]
aturley has joined #ponylang
OtakuSenpai has joined #ponylang
simuloid has joined #ponylang
<simuloid> Hello everybody. I am very unfamiliar with IRC.
<simuloid> I was writing my first pony code today and discovered that I can't use underscores anywhere in an identifier except the first character (for private members).
OtakuSenpai has quit [Read error: Connection reset by peer]
<simuloid> I was wondering why the restriction on using underscore as a word separator in identifiers is codified in the language. I have found it useful in other languages.
<vaninwagen> There is no such restriction afaik
<vaninwagen> Can you paste an example?
<simuloid> Yes, just one moment...
<simuloid> (IRC question: is there a way to paste code so it's formatted? This is only 4 or 5 lines)
<aturley> simuloid the best way to share might be to post a link to the pony playground.
<aturley> you can generate a sharable link and then we can all see how it runs.
<simuloid> Perfect. That's where I'm coding anyway. Thanks, I'll grab the link.
<simuloid> I think this might only apply to types (or maybe just type aliases). I am able to use underscores in let and var expressions.
Foaly has quit [Quit: Now 'mid shadows deep falls blessed sleep.]
endformationage has joined #ponylang
<Candle> simuloid: There is a restriction, on;y private fields and functions can start with a _. If you want "another X, a bit like X but not quite" then suffix it with any number of ' characters.
<Candle> https://tutorial.ponylang.io/expressions/variables.html See the Fields and the Shadowing sections.
<simuloid> Candle: Thank you. To clarify, I wonder why underscore is restricted from being a word separator in type identifiers. e.g. `My_Type`
<Candle> I'm not sure on the reasoning behind it. In your example, just use LimitedNumber ?
<simuloid> Yes, and since types are always started with an uppercase character, PascalCase is useable.
<simuloid> I was expecting a safety reason, given Pony's focus.
<Candle> Well, it helps with confusing the lexical rile of an identifier.
<Candle> role*
<simuloid> I agree with that.
<simuloid> OK, thank you all for your help. I am going back to the playground. Cheers!
<Candle> Cool, Experiment away!
simuloid has left #ponylang [#ponylang]
SenasOzys has quit [Remote host closed the connection]
SenasOzys has joined #ponylang
SenasOzys has quit [Ping timeout: 245 seconds]
SenasOzys has joined #ponylang
<vaninwagen> Whoa, indeed there is such a restriction! I honestly didn't know of
<vaninwagen> Thought it was only a convention, style guide thingy
aturley has quit [Quit: aturley]
SenasOzys has quit [Ping timeout: 246 seconds]
SenasOzys has joined #ponylang
<SeanTAllen> what restriction is that?
<vaninwagen> The one simuloid talked about: no underscores in type names besides the leading char
<endformationage> vaninwagen: The workaround you pointed out to me the other day re: #1888 seems to work with my example: https://playground.ponylang.io/?gist=17ccb03cc92a9eb4fdb188a513105294
<vaninwagen> BÄM!
<endformationage> Hopefully #1888 can be merged soon, as I'm not sure I want to integrate the extra complexity of the workaround into my code.
<endformationage> I am wondering if this is also what's caused trouble with the assert error I reported in #2867
SenasOzys has quit [Ping timeout: 245 seconds]
SenasOzys has joined #ponylang
OtakuSenpai has joined #ponylang
aturley has joined #ponylang
OtakuSenpai has quit [Ping timeout: 252 seconds]
OtakuSenpai has joined #ponylang
aturley has quit [Quit: aturley]