Eric Meier

**Checklist** * [x] Are you reporting a bug? Use github issues for bug reports and feature requests. For general questions, please use https://discuss.yjs.dev/ * [x] Try to report your issue...

Fix toolbar and mentions examples on android

8

**Description** Updates the mention and hovering toolbar example so that they properly work on android. This PR also contains a potentially breaking change: It memos the Editable component to make...

React package rewrite

2

Rewrite to make `@slate-yjs/react` ready for async react, add cursors decorations and other utility hooks for commonly requested features

While going further down the history rabbit hole, I discovered another issue, this time not related to history, and I'm still trying to figure out the best approach to fix...

Apollo client like generic typing

1

Currently, data and options have the "any" type. It would be nice to have apollo client like type safety using generics. e.g. ``` useQuery(LEARNING_VIEW_QUERY, options) ```

[SQLite Dialect] Properly translate .ignore()

4

Currently `.ignore()` using the SQLite dialect gets translated into `INSERT IGNORE INTO ....` Which doesn't match the SQLite expected `INSERT OR IGNORE INTO` (see [here](https://www.sqlite.org/lang_insert.html))

How-swaping model weights of torch.complied or traced model

10

I've been trying to get module weight hot-swapping working for almost two days now. All ways I found were blocked by some bug in torch XLA or pytorch itself: Here...

Minimal reproduction example: ``` import torch import torch_xla import torch_xla.core.xla_model as xm class SimpleModel(torch.nn.Module): def __init__(self): super().__init__() self.conv_in = torch.nn.Conv2d(3, 64, kernel_size=3, padding=1) def forward(self, sample): return self.conv_in(sample) def trace_model(device):...

feat: add storage_options to _BaseLanceDatasink, LanceDatasink, LanceCommitter

1

Adds the ability to pass `storage_options` to `_BaseLanceDatasink` to allow specifying them for `LanceDatasink` and `LanceCommitter` (`LanceFragmentWriter` already supports passing these). We need to customize them because some of our...

# What does this PR do? Adds support for xla devices via `torch_xla` (https://pytorch.org/xla/release/2.3/index.html) Test code: ```py import torch import torch_xla.core.xla_model as xm from safetensors.torch import save_file from safetensors import...