00:00:02 To create the tag details page, let's go ahead and create a server action that'll fetch all questions associated with each specific tag.

00:00:10 Just before we create it, let's go ahead and create a new schema for this action by heading over to validations.ts.

00:00:18 Currently, we have the paginated search params schema, which will extend by saying export const get tag questions schema,

00:00:30 which is equal to paginated search params schema dot extend.

00:00:37 and we want to extend it with a tag ID, because on top of the page, page size, query, and more, we'll also be searching based on the tag ID.

00:00:47 So, let's say that this is a z.string of main characters 1. And let's also head over to action.d.ts And here we can create a new interface called get tag

00:01:04 questions params.

00:01:04 And it'll also extend the paginated search params, but it'll extend it by omitting the filter property from it because we don't need to filter anything.

00:01:18 as we'll only be able to search by the tag ID of a type string.

00:01:23 So here you can see how we can use some of the more complex TypeScript functionalities such as extends, where now the get tag questions params will extend

00:01:33 all of the page search params, but we can also omit a specific param such as filter and then add additional params such as tag ID on top of the ones that

00:01:43 we already extended.

00:01:44 Now, let's head over to our tag actions.

00:01:48 That's going to be the new file we created.

00:01:50 And I think I called it actions, but just to stay consistent with everything we have so far, we can also just rename it to action,

00:01:58 because it'll be an action file for the tag, similar to as how this is the action file for the question.

00:02:05 And now, let's create a new function by saying export const get tag questions and we'll make that into an async arrow function.

00:02:15 Now, there are two different approaches that we can take.

00:02:18 The first one is to make a call to the questions model and find questions that contain this tag.

00:02:24 And the second one is to make a call to the tag questions model and find all related questions together by finding different documents that have the tags

00:02:33 mentioned in them.

00:02:34 Both approaches are totally valid, but in this case, we'll go with the first option.

00:02:39 The reason behind that is that although making a call to tag questions would have been similar, as it just has two fields where we have the direct reference

00:02:47 ID of the questions, things get difficult when you have to search questions of that tag by tag title.

00:02:53 With that approach, we'll have a slightly difficult filter expressions as we don't have to just populate questions, but also have to get their title and

00:03:01 check if they match the provided query.

00:03:03 This is not as easy to understand, so that's why we're going to go with the first approach.

00:03:08 We can just call the questions model, directly apply a filter query on the questions title, in the same way we call the getQuestions in the question action

00:03:18 right here.

00:03:18 So, let's make our life easy and keep our code clean by doing just that.

00:03:23 Once again, this will be almost exactly the same as what we had in the get tags.

00:03:28 So actually, I'll go ahead and copy the entire get tags function and override this one right here.

00:03:35 But of course, I will rename it to get tag questions.

00:03:41 This time our params are going to be of a type get tag question params and in the action response we can return a tag, now a singular tag,

00:03:52 but rather we'll return a list of questions which will be of a type question array.

00:03:58 Remember that this is not a question model but rather a type we have specified in the global.d.ts file.

00:04:05 that has all the info about a question.

00:04:08 Then we try to validate the params we pass in based off of the get tag questions schema.

00:04:14 We check if there are any errors.

00:04:17 We destructure all the params, but this time we don't have access to the filter.

00:04:22 Instead, we have access to the tag ID.

00:04:24 We check if we need to skip something or limit it.

00:04:27 In this case, we can remove the sort criteria.

00:04:29 So I will remove this part right here.

00:04:32 and I will actually move the filter query and the if for the query right within the try block.

00:04:38 I'll explain soon why I did that.

00:04:40 It's because we first have to fetch the tag and make sure that nothing went wrong and then with its tag id we can do further actions.

00:04:47 For example, I can say const tag is equal to await tag dot find by ID, and then I can pass in the tag ID.

00:05:00 Next, we can do some error handling by saying, if there is no tag, then simply throw a new error, tag not found.

00:05:08 And then we can start building out the query filter for questions to see if tag ID exists in the tags array.

00:05:15 We can do that by saying const filter query of a type of question.

00:05:20 And then here we can say, tags, whether it includes, so dollar sign in, the array of tag ID.

00:05:31 And this question has to be imported from add forward slash database.

00:05:34 So we're filtering only the questions that have that specific tag within their tags array.

00:05:39 Then we can check the query.

00:05:41 in case we want to search for some additional questions within that tag.

00:05:45 This will remain the same, but this time we don't have to do the or.

00:05:48 I'll simply check for the filter query dot title and then we don't have to specify the name here.

00:05:53 We can simply say regular expression where we pass the query with the options of i.

00:05:59 Finally, we want to fetch the total number of questions.

00:06:02 So that'll be total questions is equal to tag documents based off of the filter query but we count questions not tags because this function even though

00:06:13 it is about tags it is actually about fetching questions for that tag after that we want to fetch all the questions that match by saying const questions

00:06:22 is a weight questions.find filter query.

00:06:26 We want to remove the sort criteria.

00:06:28 This is actually question.find and we want to select some information out of each one of these questions.

00:06:35 We don't want to get all the information back.

00:06:37 We only want to get some things like an underscore ID, title, views, answers, upvotes, downvotes, author, as well as create a dad.

00:06:49 we only want to fetch things for each question that we can display on the list.

00:06:54 Then, we also want to populate some things, such as a path of author, and then for the author, we want to select the name and the image.

00:07:04 Again, same things as before, we're selecting what we want to get so we can showcase it on the Tag Details page.

00:07:12 You know how we have that little author name and author avatar of the person that created that question?

00:07:18 Well, this is where we get it.

00:07:19 And in the same way, we have to get the name of the tag this question belongs to.

00:07:23 We skip the number of elements to get to the second page or third one, and then we provide a limit as well.

00:07:29 We check for the isNext by referencing the total number of questions and the questions.length.

00:07:36 And finally, we return success true.

00:07:38 Data will actually be a tag.

00:07:43 and also the list of questions that match that tag.

00:07:47 So JSON parse, JSON stringify questions.

00:07:50 And this is the function that should return the questions that match to a specific tag while allowing us to further query them or filter them.

00:08:01 Great.

00:08:02 Let's go ahead and push that by saying something like implement a tags question action, commit it, sync it, and now we are ready to put it to the test

00:08:13 in the next lesson.