Ecosyste.ms: Timeline
Browse the timeline of events for every public repo on GitHub. Data updated hourly from GH Archive.
cpsievert pushed 1 commit to main rstudio/shinycoreci
- Update os values to `macos-latest`, `ubuntu-latest`, and `windows-latest` (#276) * Update os values to `macos-latest... 82e9fb0
cpsievert closed a pull request on rstudio/shinycoreci
Update os values to `macos-latest`, `ubuntu-latest`, and `windows-latest`
As of this commit... * Ubuntu is migrating from `ubuntu-20` -> `ubuntu-latest`/`ubuntu-22` * macos is migrating from `macos-12` -> `macos-latest`/`macos-14` * The windows change is a no-op. `win...
cpsievert created a comment on an issue on posit-dev/chatlas
Done in https://github.com/posit-dev/chatlas/commit/01557441f37d34c0b89950c71f42a2ece50d1ff9
cpsievert pushed 1 commit to main posit-dev/chatlas
- Fix pre-mature/unnecessary enter and exit of markdown display 7b31f76
cpsievert created a review comment on a pull request on rstudio/shiny
Another direction would be to get rid of the `status` variable altogether and use the length of `problems` to inform the early return / status
cpsievert created a review comment on a pull request on rstudio/shiny
The line below this one should probably be ``` if (duplicateIds.size === 0) return { status: status }; ```
cpsievert created a review comment on a pull request on rstudio/shiny
Seems `status: "error"` is no longer possible/relevant for `checkValidity`? In that case, I'd be favor of removing it completely, or at least a healthy comment about how we imagine it being relevan...
cpsievert created a comment on an issue on rstudio/bslib
> Using a fixedPage() inside a fluidPage() is relatively standard, typically to create a full-width header at the top of the page, and then a fixed-width container with the content of the page. Y...
cpsievert created a comment on a pull request on rstudio/shiny
> @cpsievert @gadenbuie was this too paranoid? https://github.com/rstudio/shiny/commit/d4f144d6e2c5de2053d1ef72c86125395cf5dab4 It does seem like overkill if we're just hashing strings, but if w...
cpsievert created a review comment on a pull request on rstudio/shiny
I guess it does make sense though in terms of not _showing_ the same stack multiple times?
cpsievert created a review comment on a pull request on rstudio/shiny
Are we sure we don't want this to be `rlang::hash(callStack)`? `getCallNames()` drops environments etc from the call, so potentially we'd get false positive matches?
cpsievert created a review comment on a pull request on rstudio/shiny
```suggestion attr(callStack, "shiny.stack.digest") <- getCallStackDigest(callStack, warn = FALSE) ```
cpsievert created a review comment on a pull request on rstudio/shiny
```suggestion dg <- attr(callStack, "shiny.stack.digest", exact = TRUE) ```
cpsievert pushed 2 commits to main posit-dev/chatlas