Our private tlsync package currently depends on the editor package,
which balloons the size of the cloudflare worker. It also makes it so
that any change to any package triggers a worker refresh, which makes
working on multiplayer stuff kinda miserable.
This is the first PR to fix that problem.
The second PR will need to resolve TLSyncClient's dependency on the
debugFlags somehow. Easiest would be to just remove the offending flag,
but we might want cross-bublic debug flags at some point in the future
so I'll try to find a low-cost way to make that happen while making
`tlsync` not depend on `editor`.
cc @TodePond since you added the flag in question
(`tldrawResetConnectionEveryPing`)
### Release Note
- internal moving stuff around
Right now this examples app looks exactly the same as our old examples
app, but there are a couple of tiny differences:
- We use `vite` instead of our own esbuild setup for development and
bundling
- We use `@tldraw/assets` for smart asset hashing instead of copying the
assets to a public folder
You can use `@tldraw/assets` with vite with a bunch of extra config, but
it (plus a bunch of other bundlers) also support a special syntax for
specifying asset urls: `new URL('./my/asset.svg',
import.meta.url).href`. This approach is more standards-complient, but
doesn't work with every bundler just yet. This diff also adds a
url-based version of `@tldraw/assets`, although I'd like to tweak the
entry point - right now you need to import from
`@tldraw/assets/lib/urls`, but i'd like to find a way to get this to
`@tldraw/assets/urls` or something at some point.
There are a couple other extra fixes in here:
- vscode builds were broken, they're fixed now!
- there's also a little tweak to the `getBundlerAssetUrls` API to allow
passing in a function instead of an object for URL formatting
- there are new internal-only functions for injecting asset urls
globally instead of passing them in via react props. this means we can
get the benefits of cacheable URLs without having to clutter our
examples by passing them in
In this PR I'm adding new versions of the `LiveCollaborators` and
`Collaborators` components for the ephemeral state work. They are behind
a feature flag for now.
Temporarily moving us to a prerelease of this dep. This resolves an
issue where the mobile style menu popover couldn't be closed by tapping
the style button again. See
https://github.com/radix-ui/primitives/issues/2105 for details
This PR adds
- A new `TLInstancePresence` record type, to collect info about the
presence state in a particular instance of the editor. This will
eventually be used to sync presence data instead of sending
instance-only state across the wire.
- **Record Scopes**
`RecordType` now has a `scope` property which can be one of three
things:
- `document`: the record belongs to the document and should be synced
and persisted freely. Currently: `TLDocument`, `TLPage`, `TLShape`, and
`TLAsset`
- `instance`: the record belongs to a single instance of the store and
should not be synced at all. It should not be persisted directly in most
cases, but rather compiled into a kind of 'instance configuration' to
store alongside the local document data so that when reopening the
associated document it can remember some of the previous instance state.
Currently: `TLInstance`, `TLInstancePageState`, `TLCamera`, `TLUser`,
`TLUserDocument`, `TLUserPresence`
- `presence`: the record belongs to a single instance of the store and
should not be persisted, but may be synced using the special presence
sync protocol. Currently just `TLInstancePresence`
This sets us up for the following changes, which are gonna be pretty
high-impact in terms of integrating tldraw into existing systems:
- Removing `instanceId` as a config option. Each instance gets a
randomly generated ID.
- We'd replace it with an `instanceConfig` option that has stuff like
selectedIds, camera positions, and so on. Then it's up to library users
to get and reinstate the instance config at persistence boundaries.
- Removing `userId` as config option, and removing the `TLUser` type
altogether.
- We might need to revisit when doing auth-enabled features like locking
shapes, but I suspect that will be separate.
