workspace
stringclasses 4
values | channel
stringclasses 4
values | text
stringlengths 1
3.93k
| ts
stringlengths 26
26
| user
stringlengths 2
11
|
---|---|---|---|---|
elmlang | general | I might benchmark this when I get home. :slightly_smiling_face: | 2019-03-20T13:05:26.470400 | Timika |
elmlang | general | Looking at the `elm/bytes` API, I should be able to do `String -> Bytes` then `Bytes -> Array Int` (decoded as uint8) | 2019-03-20T13:06:02.470600 | Genesis |
elmlang | general | As of now, I push 1426500 ints per second from Elm to JS via arrays of int, a minimal improvement would accumulate to a lot I guess :smile: | 2019-03-20T13:06:38.470900 | Timika |
elmlang | general | If you get some insight by trail and error, I would love to hear your experience :smile: | 2019-03-20T13:07:25.471300 | Timika |
elmlang | general | For you, I guess it would depend on what range of values you use in your ints. If you use the full range then Strings won't help size wise | 2019-03-20T13:08:05.471500 | Genesis |
elmlang | general | The main issue is that not every stream of bytes is valid UT8 | 2019-03-20T13:09:13.471800 | Huong |
elmlang | general | But maybe serialisation/deserialisation is faster or whatnot. I learned _not to trust_ my intuition about performance and just benchmark the hell out of ideas like this. | 2019-03-20T13:09:40.472000 | Timika |
elmlang | general | Right! <@Huong> Does the `elm/bytes` `String -> Bytes` encoder do any checking? Or do you think the issue is on the JS side? | 2019-03-20T13:10:29.472200 | Genesis |
elmlang | general | <@Huong> do you have an example from the top of your head where that might be the case? | 2019-03-20T13:10:59.472400 | Timika |
elmlang | general | I don't have much benchmarking experience in Elm, I'll have to look into setting that up | 2019-03-20T13:11:07.472600 | Genesis |
elmlang | general | Shameless plug: <https://github.com/Malax/elmboy/tree/master/benchmarks> | 2019-03-20T13:11:22.472800 | Timika |
elmlang | general | this is a very bare-bones setup you can just steal | 2019-03-20T13:11:34.473000 | Timika |
elmlang | general | `0xf0 0x00` would do it | 2019-03-20T13:12:00.473200 | Huong |
elmlang | general | With ports, things get tricker to benchmark tho. | 2019-03-20T13:12:05.473400 | Timika |
elmlang | general | Ah! You're the `elmboy` person! Great talk (I watched it on youtube) :slightly_smiling_face: | 2019-03-20T13:12:15.473700 | Genesis |
elmlang | general | (i.e. "broken" surrogate pairs) | 2019-03-20T13:12:58.474000 | Huong |
elmlang | general | And JS will error out with broken UTF8? (which makes sense, I don’t miss the days of `char*` at all :P) | 2019-03-20T13:13:44.474200 | Timika |
elmlang | general | Alright, thanks for your recommendations! The array of int route seems simpler to start with. :smile: | 2019-03-20T13:15:43.474400 | Genesis |
elmlang | general | not sure - JS uses UTF-16 so there's another conversion involved there :sweat_smile: | 2019-03-20T13:16:13.474600 | Huong |
elmlang | general | Right, so nothing will explode, however, the bytes representation won't quote match the bytes it came from. And even that is only because bitshifting `NaN` produces 0 :joy: | 2019-03-20T13:29:24.474800 | Huong |
elmlang | general | Does anyone here use Google's Bazel for building their code (either personally or at work)? | 2019-03-20T15:22:17.476000 | Rosanne |
elmlang | general | I don't though I was just looking at an issue related to it - <https://github.com/elm/compiler/issues/1908> | 2019-03-20T15:54:03.476800 | Huong |
elmlang | general | I think I've read something about it before but can't seem to find it...what are the bug fixes in the Elm npm package related to? The Elm compiler or the package itself? Example: `0.19.0-bugfix6`. <https://www.npmjs.com/package/elm> | 2019-03-20T16:15:44.478900 | Laurena |
elmlang | general | The last related commit I've found is this one for `0.19.0-bugfix2`: <https://github.com/elm/compiler/commit/964c62ffd6ac108d8cc004040c43bdc6acb3e7e3>. | 2019-03-20T16:17:05.479100 | Laurena |
elmlang | general | Only relevant to the npm installer itself, the binaries have not changed | 2019-03-20T16:20:27.479300 | Huong |
elmlang | general | Thanks <@Huong>! | 2019-03-20T16:21:12.479600 | Laurena |
elmlang | general | (the actual changes to the npm installer are related to upgrading the `binwrap` installer, in order to not trigger npm audit warnings) | 2019-03-20T16:21:58.479800 | Huong |
elmlang | general | <https://github.com/elm/compiler/pull/1884/files#diff-b88b915bb43bc1305102fc9079bf9907> | 2019-03-20T16:22:15.480000 | Huong |
elmlang | general | Ah, thank you for the details! Those warnings were unnecessarily ominous. :slightly_smiling_face: | 2019-03-20T16:23:44.480200 | Laurena |
elmlang | general | Ooh, interesting, thanks | 2019-03-20T16:26:48.480400 | Rosanne |
elmlang | general | <https://github.com/elm/compiler/issues/984> anyone know if there's any chance this will get revived? Some of the ideas in there (especially rtfeldman's curried update syntax) seem really nice, but I don't know if things have changed since then. | 2019-03-20T19:12:24.481500 | Vashti |
elmlang | general | I was reading [this](<https://www.infoq.com/presentations/software-languages-panel>), and they talked about elm in such nice terms that I wanted to share it with you: | 2019-03-21T00:45:43.482400 | Yetta |
elmlang | general | > Phelps: I think one thing I wanted to touch on is that Rust and Elm - Elm is a programming language designed for the web - I think they both collectively have kind of played off of each other. I don't want to give one more credit than the other. I'm not sure which came first.
> Williams: No, Rust 100% borrowed from Elm. We copied them shamelessly and we thank them daily.
> Phelps: I thought so, but I didn't want to say it.
> Williams: It's true.
> Phelps: But Elm, in hindsight, it's kind of like, “Duh”. But essentially, Elm, this programming language that was designed for the web, it's kind of like a gateway drug to Haskell, but yes, but it's much much simpler. Sorry about the drug metaphor, I guess.
> Williams: It's cool. It's like hurting cats up here.
> Phelps: I know it. But anyway, it did something novel that is laughable, but that it actually gave us real good air messages. It actually did not only give you a short description that you could google, but also a long description that a human would read, and pointed you exactly to the line of code and in a lot of cases, gave you examples of what you should probably do, what it thinks you meant or what you might want to put here instead. And that was a number of years ago when it first did that. If you're fairly new to programming, you might not have noticed because in the last many years, four or five years, other languages have borrowed from that philosophy. And I just think that that's a huge thing. It really set the bar just totally different. And it's funny, because if you write a lot of Javascript, you still to this day get undefined is not a function errors. | 2019-03-21T00:45:44.482600 | Yetta |
elmlang | general | Hi Suppose I have Msg type like this -> | 2019-03-21T04:26:05.485000 | Sibyl |
elmlang | general | None | 2019-03-21T04:26:49.485100 | Sibyl |
elmlang | general | How do I send (Submit 5 Nothing) with the onclick. | 2019-03-21T04:27:22.486000 | Sibyl |
elmlang | general | I tried using onClick FeedMsg << (Submit 5 Nothing) But could not make it work | 2019-03-21T04:28:06.487200 | Sibyl |
elmlang | general | `onClick (FeedMsg (Submit 5 Nothing))` should work | 2019-03-21T04:28:47.487500 | Jin |
elmlang | general | Thank you. It works. | 2019-03-21T04:30:44.488200 | Sibyl |
elmlang | general | See e.g. here <https://faq.elm-community.org/operators.html>
> Function application (by adjacency) is higher priority than all operators and is left-associative: | 2019-03-21T04:30:45.488300 | Jin |
elmlang | general | Hello, I'm having a hard time wrapping my head around routing in elm. I think I (roughly) understand the principles of parsing a Route, however, I don't understand how to actually show the desired content, instead of saving the current URL. In the documentation (<https://guide.elm-lang.org/webapps/url_parsing.html>), this exact part is missing (it still has a TODO in the example code block). | 2019-03-21T06:23:50.490800 | Shirley |
elmlang | general | The documentation in <https://package.elm-lang.org/packages/elm/browser/latest/Browser#application> should help, it also links to the elm-spa-example. The main idea is that you parse a `Url` into some kind of `Route`, and then use this route to update you model. A route often corresponds to a parametrised `Page`. See the main module in the spa-example, specifically <https://github.com/rtfeldman/elm-spa-example/blob/master/src/Main.elm#L156>. | 2019-03-21T06:42:32.493800 | Jin |
elmlang | general | Thank you for your reply, I will have a look at the material you provided | 2019-03-21T06:43:07.494200 | Shirley |
elmlang | general | Anyone has heard of a lib implementing a datastore in elm ? | 2019-03-21T06:44:58.494900 | Caron |
elmlang | general | Well, there’s this: <https://package.elm-lang.org/packages/billstclair/elm-localstorage/latest/> | 2019-03-21T09:29:01.495700 | Dede |
elmlang | general | But also this page talks about using ports to access localstorage and why one might _not_ want to copy the entire localstorage API: <https://guide.elm-lang.org/interop/ports.html> | 2019-03-21T09:29:40.496200 | Dede |
elmlang | general | (above is for <@Caron>’s question from ~3 hours ago) | 2019-03-21T09:30:01.496600 | Dede |
elmlang | general | Thanks :slightly_smiling_face: | 2019-03-21T09:30:50.496800 | Caron |
elmlang | general | As another example, I’ll point you at <https://github.com/jhbrown94/experimental-elm-spa-skeleton/blob/master/src/Route.elm> and <https://github.com/jhbrown94/experimental-elm-spa-skeleton/blob/master/src/Router.elm>
This approach makes a distinction between Route (the parsed URL) and the Page you end up on (which could vary due to, e.g., authentication.) It’s a work in progress so I’d caution careful consideration before totally cloning it, but maybe looking at it will help clarify how things work generally…? | 2019-03-21T09:31:36.496900 | Dede |
elmlang | general | I'm not sure that's what I needed, but maybe my initial message was not clear.
What I'm looking for is a lib that wraps a type of data and lets you store that type of data from different origins, in order to avoid data duplication and automate gc for the store. | 2019-03-21T09:34:59.498800 | Caron |
elmlang | general | What your use case? I mean implementing that is non trivial and the cost of downloading the library would probably outweigh the benefit in most cases… | 2019-03-21T09:35:53.499400 | Agustin |
elmlang | general | we load the same data for separate reasons, and need a single storage to make sure that modified data is correctly modified in each use case | 2019-03-21T10:01:28.500500 | Caron |
elmlang | general | for reasons unrelated we don't want data to be pushed from the server, so we can't just make a roundtrip with the backend to update data. | 2019-03-21T10:03:54.503000 | Caron |
elmlang | general | normalised data & garbage collection is a bit of a pain with the elm architecture (or redux) | 2019-03-21T10:15:01.507600 | Shenita |
elmlang | general | logic for what data is in use is probably highly specific to your app and very brittle | 2019-03-21T10:15:22.508100 | Shenita |
elmlang | general | probs best just denormalising in a sensible way if memory usage is an issue? | 2019-03-21T10:15:42.508600 | Shenita |
elmlang | general | we do something similar with ports, we send updates out to localstorage, store it, and send the new state back into Elm. Inside elm it is a read-only cache that we can look things up in, to modify it you have to go out a port. You can also avoid localstorage if you want and just keep an in-memory cache in Elm that you pass around | 2019-03-21T10:26:44.510300 | Alicia |
elmlang | general | we put it in our `Session` model that was passed everywhere | 2019-03-21T10:27:13.510600 | Alicia |
elmlang | general | > logic for what data is in use is probably highly specific to your app and very brittle
The general logic is complex, but what it's really is a set of rules and each is simple.
For instance, say we store meetings. Some meetings are not scheduled.
So one rule would be to say that the meetings scheduled this month are in the store, a second rule would be to say that all the unscheduled meetings created recently are in the store, and a third rule would be that the last 3 meetings returned by the search module are in store.
In a read model, you can completely separate these 3 requirements, but as soon as you write, the last search result may be modified from the monthly planning. So we need one store representation. | 2019-03-21T10:41:01.514700 | Caron |
elmlang | general | The idea for such a store would be to say that you have a `Store Meeting` to which you can add rules saying what data is appropriate in the store, and what data is relevant to a given rule (e.g. ask the store all the meetings that are in the month). Also you could modify any data in the store without caring about where it is from... | 2019-03-21T10:43:18.517600 | Caron |
elmlang | general | I had the wild hope that one such lib existed, but I'll code it myself otherwise :slightly_smiling_face: | 2019-03-21T10:44:04.518200 | Caron |
elmlang | general | ah ok, that sounds like a good approach | 2019-03-21T11:02:23.519100 | Shenita |
elmlang | general | if you were to have highly relational data though I think that's where you would run into trouble | 2019-03-21T11:02:58.519700 | Shenita |
elmlang | general | hi all! question for everyone. I watched Evan's "The Life of a File" talk two years ago and certainly agree with his premise of building files around a type, its API, and related helper functions, and not trying to keep files artificially short. however, I work on a large Elm codebase and I'm curious about the practical upper limits of file length. how long is too long, and what do you think is practical? do you think code should only be modularized when reuse is a goal? | 2019-03-21T11:22:15.524800 | Elina |
elmlang | general | <@Elina> I like small single purpose files, for example for a Page I have a `Model.elm, Update.elm, View.elm, Api.elm` etc, unfortunately Elm doesn't allow circular references, so this style of separating your files in almost impossible and you're forced to group things together. | 2019-03-21T11:30:31.528200 | Dayna |
elmlang | general | Hey, all! I've written up a blog post on how I'm handling form fields and abstracting them nicely that I believe could be useful to my fellow newbies. I'd appreciate your expert feedback -- especially on ways to improve what I've got! Thanks very much!
<https://lytedev.io/blog/elm-form-fields-abstractions/> | 2019-03-21T11:31:27.528700 | Amie |
elmlang | general | too long might have more to do with code complexity than with the actual lines of code. | 2019-03-21T12:06:52.529100 | Maida |
elmlang | general | some code just takes a lot of vertical space but it is simple code. | 2019-03-21T12:07:11.529300 | Maida |
elmlang | general | i've run into that too. we approach it by making Api the file that doesn't import from the others | 2019-03-21T12:07:12.529500 | Elina |
elmlang | general | when implementing an webapp, I like to keep all the code belonging to a page in a single file. The largest I have in my largest app is a little bit over 1k lines. | 2019-03-21T12:08:19.529700 | Maida |
elmlang | general | totally. we're considering a similar approach on the team i'm on, but some of our page files would be over 6k lines if we did it. so i'm curious what the tradeoffs are, and whether my hesitation to make files that long is actually justified or not | 2019-03-21T12:10:48.529900 | Elina |
elmlang | general | with this approach in mind I then proceed in extracting code based on domain. I like to keep the Business Objects each in their own module in a `Data` folder. These modules implement a lot of things related to the module’s object like Json Decoding and Encoding, queries and mutations for things that might be too complex. | 2019-03-21T12:11:05.530100 | Maida |
elmlang | general | then there is a module that captures most of the interaction with the backend, this is where queries are defined. Most functions take a message creator the credentials and produce a Cmd. | 2019-03-21T12:12:11.530300 | Maida |
elmlang | general | there is also a folder with all sorts of helpers. Things that should be in some `elm-community/*-extra` kind of a package. | 2019-03-21T12:13:03.530500 | Maida |
elmlang | general | and then there is the `Widgets` folder that holds complex widgets that I feel are complex enough to warrant their own module. For example, an autocomplete widget or the sidebar (which has some interactivity). | 2019-03-21T12:14:39.530700 | Maida |
elmlang | general | If a page has a complex enough subpart, I also create a folder with the name of that page and extract the complex part into its own module. e.g. I have `Pages.Foo` and `Pages.Foo.Subfoo` | 2019-03-21T12:15:49.530900 | Maida |
elmlang | general | <@Maida> it's like you're describing our codebase—so similar it's scary! | 2019-03-21T12:16:24.531100 | Elina |
elmlang | general | This is similar to how `elm-spa-example` used to be structured | 2019-03-21T12:16:42.531300 | Maida |
elmlang | general | I expect that a lot of codebases somehow end up in this kind of a structure. | 2019-03-21T12:17:11.531500 | Maida |
elmlang | general | hey all, is anyone else seeing problems with Elm installing packages at the moment?
We're seeing fairly regular instances of:
```
The following HTTP request failed:
<https://github.com/elm/json/zipball/1.1.3/>
Here is the error message I was able to extract:
HttpExceptionRequest Request { host = "<http://codeload.github.com|codeload.github.com>" port = 443 secure = True requestHeaders =[("Cookie","logged_in=no"),("User-Agent","elm/0.19.0"),("Accept-Encoding","gzip")] path = "/elm/json/legacy.zip/1.1.3" queryString = "" method = "GET" proxy =Nothing rawBody = False redirectCount = 10 responseTimeout = ResponseTimeoutDefault requestVersion = HTTP/1.1 } (StatusCodeException (Response {responseStatus = Status {statusCode = 500, statusMessage =
``` | 2019-03-21T14:28:21.533100 | Brooke |
elmlang | general | I haven’t experienced that personally today. I just tried again installing several packages in the elm/ namespace without a problem | 2019-03-21T14:53:40.533200 | Katharyn |
elmlang | general | That looks like GitHub giving 500 errors - fingers crossed it's temporary service degradation | 2019-03-21T15:01:17.533400 | Huong |
elmlang | general | <@Brooke> if this is a recurring/long-term thing, it's probably a good idea to contact gh support - they might be able to look into it :) | 2019-03-21T15:03:47.533600 | Huong |
elmlang | general | hey, can someone point me to some discourse or anything similar explaining why `elm/http` was changed the way it is. I am kinda puzzled by the fact that previously I had a possibility to simply have a single function making my request and either mapping that to `Cmd` or `Task` if necessary. Now it seems a lot like I have to write two functions not to mention that Task path does not offer json decoding by default from what I can tell. | 2019-03-21T15:42:54.534000 | Floy |
elmlang | general | I am aware there is a package `remotedata-http` that seems to be helping with that | 2019-03-21T15:45:07.534100 | Floy |
elmlang | general | just being curious about the design reasons | 2019-03-21T15:45:14.534300 | Floy |
elmlang | general | previous API seemed flexible and nice | 2019-03-21T15:45:19.534500 | Floy |
elmlang | general | new one...it's not IMHO | 2019-03-21T15:45:26.534700 | Floy |
elmlang | general | <https://elm-lang.org/blog/working-with-files> Evan wrote a little about it here.
You claim working with `Request` was simple, but there is also the factor of familiarity. Agreed that it's a bunch harder to work with requests as tasks right now. Making requests as a command - which in my experience is the most common use of elm/http, by far - is a lot less circumspect now, though. | 2019-03-21T16:01:46.535000 | Huong |
elmlang | general | yeah, although I have bunch of places in my app where the same request is used somewhere as command while being used as cmd (often paired with other requests) in view initialization | 2019-03-21T16:04:25.535200 | Floy |
elmlang | general | i.e. `MyView.init -> Task.Task Err MyView.Model` | 2019-03-21T16:04:41.535400 | Floy |
elmlang | general | it kinda makes an upgrade super hard :confused: | 2019-03-21T16:04:56.535600 | Floy |
elmlang | general | to the point where I am thinking if there's any real benefit in doing that :wink: | 2019-03-21T16:05:17.535800 | Floy |
elmlang | general | is there? | 2019-03-21T16:05:25.536100 | Floy |
elmlang | general | I am asking for an opinion from people who migrated | 2019-03-21T16:07:34.536300 | Floy |
elmlang | general | I'm currently in the process of building a single abstraction layer for all http stuff in our app, because I want to upgrade to elm/[email protected]. The main reasons I want to be able to do so:
- at some point, elm/[email protected] simply won't be supported anymore, and I don't want to hit that wall
- support for sending multipart messages with files means we can get rid of the hacks we currently have for file-uploads | 2019-03-21T16:09:11.536600 | Huong |
elmlang | general | yeah, I don't really do uploading, just downloading :confused: | 2019-03-21T16:35:17.536900 | Floy |
elmlang | general | and just in one place | 2019-03-21T16:35:20.537100 | Floy |
elmlang | general | Is there a fundamental difference between `Task.perform mytask` and `Task.perform (Process.spawn mytask)`? | 2019-03-21T17:26:48.539200 | Isaias |
elmlang | general | <@Isaias> the `Process` module isn't really finished. `spawn` gives you a `Task` that won't produce a result. | 2019-03-21T18:18:55.541300 | Earlean |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.