Rebuild LSP service, socket, and client on reconnect #201
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
DavisVaughan
commented
Jan 10, 2024
Comment on lines
+522
to
+534
| // Forward `client` along to `RMain`. | ||
| // This also updates an outdated `client` after a reconnect. | ||
| // `RMain` should be initialized by now, since the caller of this | ||
| // function waits to receive the init notification sent on | ||
| // `kernel_init_rx`. Even if it isn't, this should be okay because | ||
| // `r_task()` defensively blocks until its sender is initialized. | ||
| r_task({ | ||
| let client = client.clone(); | ||
| move || { | ||
| let main = RMain::get_mut(); | ||
| main.set_lsp_client(client); | ||
| } | ||
| }); |
There was a problem hiding this comment.
The main new bit is here
lionel-
approved these changes
Jan 11, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Follow up to #193
I accidentally introduced a bug related to UI refreshes with
CMD + R. If you run that with an active R session, then you should get a panic related to the LSP (something about unwrapping aNone).It turns out that we can't move the LSP service, socket, and client to
start_kernel(). That code will get run exactly 1 time per ark session lifetime, but thelsp_start()method will get called every time the UI is refreshed (i.e.CMD + R) (which prompts a newcomm_openrequest). This means we really do need "fresh" instances of these objects after each refresh.Most of the code from #193 related to these objects is now reverted, moving their creation back into
start_lsp().However, we now also use an
r_task()to ship theClientover toRMainafter each refresh.R_MAINactually should be started up by the time we get here to call thisr_task(), because theLsp::start()method that callsstart_lsp()will wait for thekernel_init_txnotification that is sent out by the firstr_read_console()iteration, andR_MAINexists at that point. I'm convinced that even if it didn't exist yet, thenr_task()would correctly block until the channel for communicating with the main R thread was set up, due toget_tasks_tx().I did some testing and things like
usethis::edit_r_profile()(which use theClientstored inR_MAIN) work as expected before and after aCMD + Rrefresh, which makes me somewhat confident that the client is getting refreshed properly.