Lisplog | Blogging in Lisp

Why Elm?

Submitted by Bill St. Clair on Fri, 19 Jul 2019 10:28:57 GMT

An AngularJS user in the Fediverse asked me to sell Elm to him. This is my first attempt.

I have never used AngularJS or any other JS framework besides jQuery, so I'm operating from a quick perusal of the Angular web site, making my comparisons mostly invalid.

Caveats

First off, you're going to miss a lot of Angular's features with Elm. Elm has a functional flavor to it, not the templatey feel of Angular. Your entire site, except for the initial HTML file that loads the elm-generated JS, is Elm. And Elm wants total control. With a couple of exceptions:

You may load Elm into only a single DIV of your larger site, and use regular HTML/JS for the rest of it. This, so I've read, can enable incremental conversion to Elm, but I've never done it.
Elm supports two ways to get to your own JavaScript code:
1. Ports - a send/receive messaging mechanism to get JS data out of and back into Elm. Fully type-safe, since incoming JS data goes through decoders to bring it into Elm, and errors there must be handled by the functional Elm code.
2. Custom HTML Elements - Since Elm's virtual DOM can support ANY HTML element, you can roll your own, to do anything you can dream up.

Elm's package manager is centered on open source, stored at GitHub. I've proposed a simple mechanism to open this up, so that you can easily define your own package plugins, or not use Microsoft's repository system, but Evan Czaplicki, Elm's creator, has not been open to the idea.

A normal way around this for organizations that need their code to be proprietary is a mono-repo for your code, and just include ths whole thing in every project. Elm's compiler is so fast, and so good at rebuilding only what needs to be rebuilt, and including in the output only what is actually used, that this isn't as bad a solution as it sounds.

Or, if you have Haskell chops, you could fork the Elm source, and add the package manager plugin yourself.

Advantages

Now that I've highlighted some caveats, on to Elm's advantages.

No run-time errors. Ever. If it compiles, the JS generated by the Elm compiler will contain logic errors that you need to find and fix, but it will NEVER mistakenly reference a null value or call an undefined function. This is remarkable to experience, for one accustomed to the runtime-error-prone JS environment. And it really works.

The same kind of type-checking that C programmers have had for decades, but without the necessity to say everything twice, and with a much friendlier compiler, and no way to get around the type safety (except in your own custom JS, as mentioned above). JS linters may give you a lot of this, but they don't do a complete job. The Elm compiler does.

As a long-time lisper, accustomed to run-time type checking, but compile-time laxity, I was surprised to find that I really like this. It takes a little longer to get your code to compile, but once you do, you know it won't fail because of a runtime error.

Refactorings, even large refactorings, are a joy, and usually "just work" once you get the refactored code to compile. In Elm, I'm not afraid to make radical changes to pieces of my code. This causes a large list of compiler errors early in the process, but you fix them one at a time until it builds, and then it usually works.

The generated JavaScript is not difficult to read, and is pretty good, performant code. You definitely lost a little in efficiency over hand-coded JS, but you can always move inner loops into ports if necessary, and it is rarely necessary.

Elm webapps behave pretty much like native apps. With all the JS frameworks available, this isn't as much of a sell as it used to be.

Conclusion

I wish I knew how to express my sheer joy when I program in Elm. If I had Shakespeare's skill, I'd write a sonnet. "Shall I compare thee to a summer's day?"

1 comment Edit post Add post

OAuth2 and Mastodon

Submitted by Bill St. Clair on Mon, 01 Jul 2019 22:15:18 GMT

This page explains my thinking of doing OAuth2 authorization for an Elm web frontend talking to a Mastodon API backend.

The OAuth2 Authorization Code Grant Flow Dance

OAuth2 has a number of authentication methods. For end users, using web apps, the Authorization Code Grant Flow is common. It lets the user authenticate with the service she's using, such that the client code never sees her userid or password, then the client code fetches a token it can use to authenticate API requests. If the user notices the client misbehaving, she can go to the service's web site, and remove permission for that client to access her account.

There are three computers involved:

The user's computer, often via a web browser.
The client application's web server, accessed via the redirectUri.
The API service's web server, accessed via the authorizationUri, the tokenUri, and the apiUri.

Examples:

redirectUri:	https://xossbow.com/oath
authorizationUri:	https://mastodon.social/oauth/authorize
tokenUri:	https://mastodon.social/oauth/token
apiUri:	https://mastodon.social/api/v1

Mastodon requires /oauth/authorize and /oauth/token as the OAuth 2 endpoints and /api/v1 as the base of the REST API URLs.

There is some information that the API service uses to identify the client application:

The application name
The application Url (optional)
The clientId
The clientSecret

The clientId can be present in the webapp, on the users's computer, since without the clientSecret, it cannot be used for anything. The clientSecret is kept secret, on the redirect server.

Steps in the dance:

The user clicks a "Login" button in her web browser.
The webapp redirects to the authorizationUri, passing the clientId, the redirectUri, scope descriptors, saying what the client app will be allowed to do, and a state string.
The authorizationUri puts up a form, requesting userid and password.
If the user logs in successfully, the authorizationUri server forwards the user's browser to the redirectUri, passing the clientId, state, and an authentication code that it generates and remembers.
The redirectUri server posts the clientId, clientSecret, and authentication code to the tokenUri, and receives back a token.
The redirectUri server then uses the state to put up a web page for the user to interact with the server via the API.
When the user does something that requires an API call, the token is passed along, for authentication and identification of the user.

The Status Quo

First, I'll explain from where I started, back in December of 2017, before adding Mastodon to the mix.

I published the billstclair/elm-oauth-middleware in the public Elm repository. It works with Google, Facebook, GitHub, Gab Legacy, and likely any other proper implementation of the OAuth2 Authorization Code Grant Flow (but I only tested those four). It is running at https://xossbow.com/oath. It contains both server and client code.

The elm-oauth-middleware server expects application definitions to be mostly static, with the tokenUri, clientId, and clientSecret stored in a JSON file on the server, which is queried periodically, and reloaded if it changes. This allows hot changes to the applications, with a text editor on the server.

I store the clientId, and redirectUri in the Elm client code, compiled to JavaScript in the browser, again loaded from a JSON file that ships with the client application. It redirects to the authorizationUri, passing the clientId, redirectUri, scope, and some state. The authorization server (Google, Facebook, GitHub, Gab, etc.) prompts the user for ID and password, and redirects to the redirectUri with an authorization code and some Base64 encoded JSON state. That server posts the code to the tokenUri to get a token, which it returns to the client code, by using the state to go to a URL on a redirectBackHost that are validated from its configuration file. Validation of that redirectBackHost is my invention.

This is a non-standard use of the Authorization Code Grant Flow. Usually, the token stays on the server, and it uses it to make API calls and then populate HTML for the client browser. Since my clients are all in Elm, and work with no HTML generation by a server, other than an initial static HTML file that loads the Elm JavaScript, that client needs to have the token, and make calls itself to the apiUri. The client typically stores the token in JavaScript localStorage, so that it doesn't have to request it again every time the user goes to its web site.

Enter Mastodon

Normal Mastodon servers have a Your Applications page, linked from </> Development in the left column of the preferences page. This allows you to create a standard, static, OAuth2 application, giving an Application name, Application website, Redirect URI list, and allowed Scopes, and receiving a clientId, clientSecret, and token.

This is fine if your application is targeted to a small set of Mastodon servers, but the nature of Mastodon is hundreds of federated servers, so the API has a POST /api/v1/apps call to create a new client_id and client_secret.

My idea for using this is to have the Elm app send its own URL as the Redirect URI, use POST /api/v1/apps to get a client ID and secret, then redirect to https://<mastodon-host>/oauth/authorize, so the user can log in, get back the authorization code when restarted as the redirectUri, then POST to https://<mastodon-host>/oauth/token to turn that code into a token. The only thing I don't now yet is whether that final POST will pass CORS muster. The API calls have to, but that one doesn't. If it doesn't, then that part of the dance needs to be moved to a server, with the clientId and clientSecret passed in the state, so that the server doesn't need any state itself.

This is likely an unusual use of the POST /api/v1/apps call. I think it's expected that the redirectUri host will save the association of the <mastodon-host> and a clientId/clientSecret pair, so that it doesn't need to request a new one unless a new user specifies a never-before-seen server. I plan to cache the clientId/clientSecret pair (and the most recently issued token) in localStorage on the user's machine, but store no state on the server, even if I need one to get around CORS.

1 comment Edit post Add post

ZAP Meme

Submitted by Bill St. Clair on Thu, 18 Apr 2019 05:25:32 GMT

https://zapmeme.com is a meme maker written in Elm.

It's open source, with a link at the bottom of the page. It uses <svg> to layout the meme, and converts to JPEG or PNG for saving.

It stores meme and images (as data:// URLs) in your browser's localStorage database. They may be exported as JSON and imported into another browser.

kludges

Add comment Edit post Add post

stackoverflow April Fools 2019

Submitted by Bill St. Clair on Mon, 01 Apr 2019 07:57:03 GMT

For April Fools Day, 2019, stackoverflow put fairy dust on every question page, and had a definite retro look.

I saved the fairy dust, and the ASCII art HTML comment at lisplog.org/20190401.

Also see Announcing the Stack Overflow Time Machine.

Add comment Edit post Add post

Image Sizing in elm-ui

Submitted by Bill St. Clair on Thu, 06 Dec 2018 03:53:36 GMT

I've been using Matthew Griffith's wonderful elm-ui package to make the user interface for GabDecker, a TweetDeck-like web app I'm building for Gab.com. I've had inline images since the beginning, scaled to fit the column width. Today I made clicking on one of those image open a dialog with the full-size version, scaled down if necessary to fit the available space.

elm-ui eschews CSS in the source code, usually making it much easier to get what you want. But in this case, I was NOT getting what I wanted, so I wrote my own CSS, and learned about the object-fit property in the process.

Here's the code that's now running:

imageDialog : String -> Model -> Element Msg
imageDialog url model =
    let
        maxw =
            9 * model.windowWidth // 10

        maxws =
            String.fromInt maxw ++ "px"

        maxh =
            9 * model.windowHeight // 10

        maxhs =
            String.fromInt (9 * model.windowHeight // 10) ++ "px"
    in
    column
        -- This is black magic.
        -- It took much play with the CSS to get it right.
        [ centerX
        , centerY
        ]
        [ standardButton "" CloseDialog <|
            (Html.img
                [ Attributes.style "object-fit" "contain"
                , Attributes.style "max-width" maxws
                , Attributes.style "max-height" maxhs
                , Attributes.style "border" "2px solid black"
                , Attributes.style "width" "auto"
                , Attributes.style "height" "auto"
                , Attributes.src url
                ]
                []
                |> Element.html
            )
        ]

Add comment Edit post Add post

Quote

Submitted by Bill St. Clair on Wed, 19 Sep 2018 10:04:00 GMT

"APL is like a diamond. It has a beautiful crystal structure; all of its parts are related in a uniform and elegant way. But if you try to extend this structure in any way - even by adding another diamond - you get an ugly kludge. LISP, on the other hand, is like a ball of mud. You can add any amount of mud to it and it still looks like a ball of mud." -- Joel Moses

Add comment Edit post Add post

How do I run elm-test in Elm 0.19?

Submitted by Bill St. Clair on Mon, 17 Sep 2018 00:24:33 GMT

This is a FAQ, so I'm making a blog post to answer it.

elm-test is still in beta for Elm 0.19. To install it:

$ npm install -g elm-test@beta

The tests directory no longer needs an elm-package.json file (nor does it need elm.json).

Your top-level elm.json file needs to include the following:

    "test-dependencies": {
        "elm-explorations/test": "1.0.0 <= v < 2.0.0"
    }

The easiest way to get it there is with elm-test itself. The following will create a tests directory, containing an example test file, Example.elm, and will add test-dependencies to elm.json.

$ elm-test init

That's it. There is not yet support for running elm-test in the browser, nor is there support for testing DOM output.

billstclair/elm-localstorage is configured with a tests directory to test its JSON encoders and decoders: github.com/billstclair/elm-localstorage. To run the tests:

$ git clone git@github.com:billstclair/elm-localstorage.git
$ cd elm-localstorage
$ elm-test

Add comment Edit post Add post

CLtL2 on Kludges

Submitted by Bill St. Clair on Sat, 15 Sep 2018 11:26:58 GMT

Lest we believe we're building amazing systems as programmers, always remember that Common Lisp the Language, Second Edition, by Guy L. Steele, Jr. (CLtL2), has an index entry for "kludges", saying that they appear on pages 1 to 971. Page 972 is the "References" section.

kludges

Add comment Edit post Add post

iPhone Xs Pricing

Submitted by Bill St. Clair on Wed, 12 Sep 2018 19:22:55 GMT

From the www.apple.com/iphone pages.

Watch the 12 September keynote.

Prices (in parentheses) are per month for Apple financing.

Model	64 gig	128 gig	256 gig	512 gig	Order	Ship
iPhone Xs:	$999 ($49.91)		$1149 ($56.16)	$1349 ($64.50)	9/14	9/21
Xs Max:	$1099 ($54.08)		$1249 ($60.33)	$1449 ($68.66)	9/14	9/21
Xr:	$749 ($37.41)	$799 ($39.50)	$899 ($43.66)		10/19	10/26
iOS 12:	Free					9/17
macOS Mojave	Free					9/24

Add comment Edit post Add post

New Elm Package: billstclair/elm-port-funnel

Submitted by Bill St. Clair on Sat, 08 Sep 2018 20:23:12 GMT

I just published version 1.0.0 of billstclair/elm-port-funnel. It enables communicating with ALL of your custom JavaScript through a single Cmd/Sub port pair. It takes care of loading the JavaScript code, which you simply drop in a directory and add to a list, and dispatching messages from your Elm code. It allows pure-Elm simulators, to make testing your app easy in `elm reactor`.

https://package.elm-lang.org/packages/billstclair/elm-port-funnel/latest

The example is live at https://billstclair.github.io/elm-port-funnel

I will now begin work at converting my existing port modules to funnels. Those will be major version bumps, but will make them much easier to combine.

Thank you to Murphy Randall (@splodingsocks) for the basic idea that I generalized to create this package.

Add comment Edit post Add post