imashnake0 / animite Goto Github PK

No stack trace other than this, so it probably crashes in a LazyColumn.

You know what Compose is like 🤷

E/AndroidRuntime: FATAL EXCEPTION: main
    Process: com.imashnake.animite, PID: 20062

I spent 15 minutes looking for the cause, but I'd have to put a log statement after every line, and I don't have the time for that atm.

Padding for the background is messed up in landscape mode:

Hi I cannot install the app it says package is invalid.

Adding a static code analysis tool like Detekt can help us find code quality issues fast, which is great for both CI and local dev.
https://detekt.dev

Some preliminary issues we should resolve first

• Wildcard imports
• Forbidden comment (TODOs)
• Magic numbers
• Swallowed exceptions
• Long parameter list

It shows error on installing the app
It don't have any previous version of this apk

Looks like there are some problems with the current design: Anime titles that are really long don't fit well and the picture on the home page is bigger than it should be.

• Decrease the picture size to what it is on MediaPage.
• #50
  Note: Out of scope.
• Accommodate long media titles.
• Improve the layouts (for instance, use a Scaffold in Home and MediaPage).
  Note: Scaffolds are awful; source: Zhuin.
• Break layouts into multiple Composables (see reviews in #46).
  Note: This is not needed right now since I don't use Previews :dev_cool_emoji:.
• Create a ProgressIndicator composable.
• Add/Improve ripples.
• If needed, fix typography.

Bonus:

• Support landscape mode.
• Rewrite material-color-utilities in Kotlin and publish it. 😩

Noticed that the keyboard launch is not very smooth:

Felt like we could start working on user auth and #39 in parallel since they don't depend on each other a lot (the only potential difference is that there will be a bottom app bar when the user is logged in; along with some metadata). I think it's a good idea to separate this work into a new module since we'll likely need to reuse this code.

• Create a new auth module.
• Implement user authentication.
• Display the user's Media List in the Profile path.
• Allow users to log out.

These should probably all match. They are all "top-level" destinations, so it makes sense for them to all have the same animation

Pls bro this is an issue:

• Rewrite material-color-utils in Kotlin.

What behavior will the Home tab have? Will it have an "explore" and/or "trending" tab, or "seasonal" even?

Here are my thoughts for the Home tab (to get started):

• A searchbar on the top to search stuff with a filter, like in AL-chan:

  AL-chan search bar behavior

  
  

• Display trending media in horizontal views.

• If the user is logged in, display the name and/or the cover image above the search bar.

These are only some considerations, I think these tasks are essential to start with it:

• Create the GraphQL queries to get trending media.
• #23.
• Some UX design to have as reference (maybe I'm going to make one).

There's no way of seeing the entire list without collapsing the keyboard. The layout is also a bit of a mess. Action items:

• Add Modifier.imeNestedScroll().
• Improve the layout: Make the list occupy the entire screen.
• Add item enter and exit animations.
• Fix typography if needed.
• Related to #42?

As mentioned in Discord, light mode support would be nice :kappa:

Sane people start with this first, there will be some refactoring to sort this out now.

It looks like there's only been significant work completed on "Home". Once work commences on other features such as "profile", it might make sense to break things down into feature modules. This has many benefits, from code separation to build speed

Hii is it going to be an only Ani client or you going to add source in it to watch that particular animes?

Because of rate limiting, we now use FetchPolicy.CacheFirst (with #126), this means that data is loaded from the cache if it's available and network if not. The only way to refresh data is to clear storage. Compose Material3 1.2.0 has Pull to refresh, use this to refresh content on demand alongside CacheFirst.

AniList's markdown differs slightly from the standard commonmark markdown.
This requires us to maintain a markdown parser for it separately.

I've been considering forking an existing MD parsing library, the options are:
commonmark-java
JetBrains' MD parser

This forum post describes how AL flavoured Markdown differs from CommonMark: https://anilist.co/forum/thread/6125

Changing the banner is probably a nice... nice to have, see if we can make use of the new photo picker.

Being able to click on media is, I feel like, the next most natural addition. This helps build a foundation for navigation
(consider: Continued use of navigation-compose or simple-stack 👀; decided to use compose-destinations) and heavier requests, which means we'd likely have to use a shimmer or something else to indicate a load. We should also ideally have mockups for this page.

• Modify MediaQuery.
• Create mockups.
• Improve navigation and make media clickable.
• Integrate with #23.

This issue is for tracking the implementation of the ProfileScreen. Currently, there is a login (only) feature which means we can query the current viewer (and other users).

• Cleanup HomeScreen and MediaPage: These haven't been touched in a while and since ProfileScreen will follow a very similar layout, it's worth refactoring them.
• BannerLayout: This will be used in all three layouts, potentially close #50.
• Mockups for ProfileScreen: Create some initial mockups (using Relay?).
• Implement ProfileScreen.

This issue will entail details for creating the search bar component. Potential sub-tasks:

• The data layer needs a refactor after #20 is merged.
• Create a SearchBar Composable.
• Use coroutine APIs to get search results based on what is... searched (what's a Flow?) Create a SearchQuery.
• Animate SearchBar.