Recover from a Nothing with the `alt` method

Andy Van Slaars
InstructorAndy Van Slaars
Share this video with your friends

Social Share Links

Send Tweet
Published 6 years ago
Updated 5 years ago

Once we’re using Maybes throughout our code, it stands to reason that at some point we’ll get a Maybe and want to continue applying transformations. We might get a Nothing and want to perform those same transformations on some default value. In this lesson, we’ll do exactly that! We’ll see how we can use the alt method on the Maybe to recover from a Nothing and continue applying transformations with a Just.

Instructor: [00:01] We're importing crocks as a whole and then using destructuring to grab several of its functions, as well as the maybe type. We also have a couple of utilities being pulled in from the rand library, and we have three functions already defined.

[00:12] It is now an empty string as a predicate function that checks that a value is both a string and that it contains some characters. Create URL slug will take in a string, convert it to lowercase, and replace the spaces with dashes. Then finally, create URL takes in a slug and appends it to the end of our base URL.

[00:30] Our goal is to take this article object and get a URL that contains the name as a slug. We want to do this all within the safe confines of the maybe type.

[00:39] Let's start by collapsing this imports region so we have a little more space. Then, under the article object, I'm going to start by defining a new concept. We'll call this getArticleName, and getArticleName is going to take in an object. Then, we're going to use the prop function from crocks to get the name property from it. This should return us a maybe with that name.

[01:02] Then, we're going to use this to get our URL. Let's start defining that. We'll define a constant called URL. That's going to equal a call to getArticleName. We're going to pass in our article object from above. We'll see that our article name has been wrapped in a just.

[01:22] Let's verify that this works if name isn't available. I'll rename that. We'll see that we get a nothing. That's where it came, but what happens if our name is an empty string? You'll see down here when we call getArticleName, we get a just of an empty string.

[01:37] We'd like to make sure that there's actually a value in this title before we do any transformations to it. We're going to update getArticleName to do an extra check. Prop will give us back the property on the object as long as it exists, but it's not going to do any additional checks.

[01:52] What we want to do is chain another call. This time we're going to call safe. It's going to take that property it got from the object. It will pass it into the chain, which is going to call safe, and we're going to use our is-now-an-empty-string predicate. That's going to return a maybe of its own.

[02:10] That's why we use chain here instead of map, so that it flattens that out and we don't get a nested just. Now, we'll see we're getting our just with our title. If I come up here and I make this an empty string, we get a nothing.

[02:26] We know that any transformation we apply to a just is only going to be applied to a value. Let's add a map on our URL assignment. We're going to map our returned article name over our create URL slug function.

[02:41] Now, we're going to get back a just with our URL slug. We want to create a full URL from that slug. We'll call map again, this time passing it. Create URL. You'll see that we get a full URL with our slug wrapped in a just.

[02:58] Now that we've successfully created our URL, we're going to unwrap it and get the string back. We're going to reach for option here. For now, we'll pass it an empty string as the default.

[03:08] You'll see the result of this is just our URL. If I come up here and I get rid of the name property, we'll get a nothing and our output is going to be an empty string. If I update this, it will be a default value.

[03:25] In order for the consuming code to continue to work as expected, this default really should be a URL. I could hard code a URL here, but the problem now is that base URL is being duplicated. If that changes, I have to remember to update it here. I really don't want to have to do all of that.

[03:40] Luckily, the maybe from crocks gives us an easy way to recover from a nothing. I'm going to come up there between create URL slug and our map to create URL. I'm going to call the alt method on our maybe.

[03:53] Alt is actually going to take a maybe. In the case of a nothing, we're going to get that maybe back, and then it will continue processing.

[04:01] I'm going to use maybe. I'm going to call the of method on that, which is a constructor that's going to take whatever value we give it and put it into a just. I'm going to give this a string, which is going to be a slug for page not found. You'll see that now our result is our URL with the page-not-found slug.

[04:21] We get a nothing. We skip this mapping. We recover from it. Now, we have a just again, so the next map is going to be invoked, which is going to create our URL with this default value.

[04:33] We've recovered and this works, but the problem we have now is that we've basically taken this hard-coded value. We're in a position where we have to make sure that this hard-coded value adheres to whatever logic is used in this create URL slug.

[04:47] A better option would be move this alt up. Then we can pass it a value that we would expect to be passed through the create URL slug, and then we can run our transformations on that default value. We don't have to know about the logic that comes after that. We know that we're recovering with a reasonable default.

[05:10] The other thing we can do is, if this default is going to be applicable any time we use that getArticleName function, we can move it up there. Then, our consuming code doesn't have to worry about recovering at that point in the pipeline.

[05:23] Now that our recovery isn't splitting up our two maps, we can consolidate these using composition. We can compose and we can call create URL after we call create URL slug, and get rid of that second map.

[05:36] We can verify that that's still working and we can make this a little more declarative by cutting this out, moving it into its own utility function. We'll assign that composition to create URL from name. Then, we can use that down here. Everything works and our code is nice and clean.

egghead
egghead
~ 20 minutes ago

Member comments are a way for members to communicate, interact, and ask questions about a lesson.

The instructor or someone from the community might respond to your question Here are a few basic guidelines to commenting on egghead.io

Be on-Topic

Comments are for discussing a lesson. If you're having a general issue with the website functionality, please contact us at support@egghead.io.

Avoid meta-discussion

  • This was great!
  • This was horrible!
  • I didn't like this because it didn't match my skill level.
  • +1 It will likely be deleted as spam.

Code Problems?

Should be accompanied by code! Codesandbox or Stackblitz provide a way to share code and discuss it in context

Details and Context

Vague question? Vague answer. Any details and context you can provide will lure more interesting answers!

Markdown supported.
Become a member to join the discussionEnroll Today