-
Notifications
You must be signed in to change notification settings - Fork 11
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Initial port for wasmJs #170
base: main
Are you sure you want to change the base?
Conversation
4e2d6d5
to
d72f896
Compare
- I put this separate because it's a lot to have in the file - This would be a good candidate for context parameters when that is available
This is awesome, thank you! I'll start looking at it ASAP |
@@ -1,17 +1,17 @@ | |||
package com.juul.indexeddb |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It looks like quite a few files are missing their package directive now. Just gonna leave this one comment to cover all of them.
* This API is delicate. If you're passing in Kotlin objects directly, you're probably doing it wrong. | ||
* | ||
* Generally, you'll want to create an explicit `external interface` and pass that in, to guarantee that Kotlin | ||
* doesn't mangle, prefix, or otherwise mess with your field names. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Now that we have JsAny
, I don't think we need the lines about delicate API and external interface
. The same applies to all the other overloads of this function, and similar functions like put
.
public expect interface JsAny | ||
public expect interface UnsafeJsAny : JsAny |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What's the difference between JsAny
and UnsafeJsAny
?
@PublishedApi | ||
internal expect fun Array<IDBKey>.unsafeCast(): IDBKey | ||
|
||
public expect interface JsAny |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Would love to see a doc comment on this explaining it. From my understanding, JsAny
is meant as a marker of DB-safe types. Ideally something that includes theexternal interface MyRowType : JsAny
use case.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
JsAny
is really just there so that there can be common code for Js and Wasm. Wasm requires externals to extend JsAny
, so that had to be represented in the common code.
I believe Kotlin is working on improving the interop story for Js and Wasm, so this can likely get simplified in the future.
database.close() | ||
deleteDatabase("auto-increment-keys") |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm not opposed to losing the test wrapper if there was a real reason it had to go (I'm guessing it didn't play nice with wasm?), but if it's gone there needs to be a try ... finally
for these cleanup lines. Otherwise a failing test leaves some garbage behind that can make future test runs (which should be successful) fail. Same goes for the other tests
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There was an issue with Wasm, although I have a few ideas about how to make it work.
But worst case scenario I will add the try...finally
Absolutely heroic effort! A few comments, but nothing big/architectural, just some small details.
Love the improvements on this front. The footguns are now, at least, more explicit -- it looks a lot harder to accidentally pass something as |
Any progress on this front? |
I've been distracted, but I hope to get back to this in the coming weeks. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is a bit outside my wheelhouse, but I'll try to review the code for readability/understandability more so than correctness of interacting with IndexedDB.
Just a couple comments so far. Overall, as @cedrickcooke mentioned, what an amazing effort on this! Thanks!
[![Slack](https://img.shields.io/badge/Slack-%23juul--libraries-ECB22E.svg?logo=data:image/svg+xml;base64,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&labelColor=611f69)](https://kotlinlang.slack.com/messages/juul-libraries/) | ||
|
||
# Kotlin IndexedDB | ||
|
||
A wrapper around [IndexedDB] which allows for access from Kotlin/JS code using `suspend` blocks and linear, non-callback based control flow. | ||
A wrapper around [IndexedDB] which allows for access from Kotlin/JS or Kotlin/WasmJs code using `suspend` blocks and linear, non-callback based control flow. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
nitpick: typically "Kotlin/Wasm" is used (without the "Js" attached).
A wrapper around [IndexedDB] which allows for access from Kotlin/JS or Kotlin/WasmJs code using `suspend` blocks and linear, non-callback based control flow. | |
A wrapper around [IndexedDB] which allows for access from Kotlin/JS or Kotlin/Wasm code using `suspend` blocks and linear, non-callback based control flow. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I prefer the original wording because it is more precise. The use of IndexedDB is tied to the browser and thus to the wasmJS target and not WASM in general.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I did that to differentiate it from wasmWasi, although I guess the fact that it's a browser API should be enough to differentiate implicitly.
import kotlin.js.unsafeCast | ||
import kotlin.js.unsafeCast as jsUnsafeCast |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Curious the reason for the import aliasing?
Also, why both unsafeCast
import and the alias for it too?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could be a mistake, I'll double check.
pinging this! |
Sorry, haven't had much free time to get back to this. Hopefully will be able to soon. |
Update: real life continues to get in the way of addressing comments 😞 I haven't forgotten about it, and will come back to it as soon as I'm able, but if anyone wants to pick up where I left off, feel free 😄 |
I'd like to supply a common interface, but at first glance that looks like it might be a lot more work. I started work on commonizing the interface. Tests are currently failing; I am working on that now.Commonizing is done and tests are passing. This solution required having expect/actuals for primitives (so that js and wasmJs have a shared JsAny; see discussion here and elsewhere).I'm trying to rework the API a bit so that none of that has to be exposed. Not sure if it is possible, but I'll give it a try.It should be done now. There's still weirdness to support common code between js and wasmJs, but hopefully the API surface is a little better now, e.g. a lot of the type checking happen when you create a key instead of validating a key after it was created. There are still footguns available to the user, but they would be there anyways (e.g. unsafe casting).
It seems like there is partial support for null keys in js, but the spec seems to suggest that it shouldn't be allowed. I mirrored the null key support in wasmJs for now.
Fixes #157