-
Notifications
You must be signed in to change notification settings - Fork 194
Comparing changes
Open a pull request
base repository: jetify-com/devbox
base: 0.10.6
head repository: jetify-com/devbox
compare: 0.10.7
- 13 commits
- 22 files changed
- 5 contributors
Commits on Apr 26, 2024
-
[easy] Cleanup cache action (#2026)
## Summary TSIA ## How was it tested?
Configuration menu - View commit details
-
Copy full SHA for 985d27a - Browse repository at this point
Copy the full SHA 985d27aView commit details -
fix: close archivePath (#2027)
## Summary ## How was it tested? Signed-off-by: guangwu <[email protected]>
Configuration menu - View commit details
-
Copy full SHA for e6e039f - Browse repository at this point
Copy the full SHA e6e039fView commit details
Commits on Apr 27, 2024
-
internal/nix: support older nix versions in DaemonVersion (#2028)
To check if nix daemon is running, Devbox runs: nix store info --json --store daemon This fails on older versions of Nix for a couple reasons: - Before Nix 2.19.0, `nix store info` was `nix store ping`. - Before Nix 2.14.0, the `--json` flag wasn't supported. Check for both of these versions when constructing the `nix store` command so it works for Nix versions 2.12 - 2.21+. Also add a `nix.VersionInfo.AtLeast` method and constants to make checking for supported major Nix versions easier. Tested by running the `nix store` commands on all supported Nix versions.
Configuration menu - View commit details
-
Copy full SHA for 0d3f7f9 - Browse repository at this point
Copy the full SHA 0d3f7f9View commit details
Commits on May 1, 2024
-
[nix] Build all packages in same nix command (#2032)
## Summary This changes how `installNixPackagesToStore` calls `nix.Build` so that it installs all installables in same nix command. It does this by grouping all installables into two groups, one for allow-insecure and other for default and runs `build` on both groups (if not empty) I was able to remove `packageInstallErrorHandler` because it is already called in `packagesToInstallInStore` so we will never actually try to build something that can't build on the system, or if it requires `--allow-insecure` we will notice it then. This does change our segment logging to group packages together for install times, so we will need to do some more work on the metrics side to determine slow packages. ## How was it tested? * Installed multiple packages not present in my nix store at once, including a combination of secure and insecure ones. * Tried to install something that doesn't build on my system, saw correct error message ## Perf test Anecdotal test: Using the following <img width="257" alt="image" src="https://github.com/jetify-com/devbox/assets/544948/8b1ab39c-7a1b-40d3-ab5a-516cde3b1525"> Before <img width="261" alt="image" src="https://github.com/jetify-com/devbox/assets/544948/ce397f59-1eaf-41dc-8256-6306f36e596b"> After <img width="293" alt="image" src="https://github.com/jetify-com/devbox/assets/544948/55ed3214-ff1d-4115-b767-7ad5d29dedd3">
Configuration menu - View commit details
-
Copy full SHA for bc719ca - Browse repository at this point
Copy the full SHA bc719caView commit details
Commits on May 8, 2024
-
[process-compose] Bump version (#2040)
## Summary Bumps to v1.5.0 Note: Because process-compose flake is not in any cache installing a new version can be pretty slow (it has to build which means it has to download dependencies in order to build). A few solutions: * We could use nixpkgs (which lags the repo version by a few versions) * We could use runx to install it (not nix, but it gets the latest version) * We could store in our own cache. ## How was it tested? Tested running services on a few examples.
Configuration menu - View commit details
-
Copy full SHA for c237f94 - Browse repository at this point
Copy the full SHA c237f94View commit details
Commits on May 9, 2024
-
[perf] Skip cache check if store path exists locally (#2042)
## Summary If store path exists locally, we know we can use that store path in `fetchClosure`. This helps speed up almost any operation where state is not up to date and we need to check if some store path is in cache. ## How was it tested? * added print statement `devbox run echo 5`
Configuration menu - View commit details
-
Copy full SHA for ec0a703 - Browse repository at this point
Copy the full SHA ec0a703View commit details
Commits on May 13, 2024
-
Indicate of a package version list in
devbox search
is truncated (#……2044) ## Summary Current `devbox search` UI only shows the 10 newest versions for a package, unless the developer runs the command with `--show-all`. With the current presentation, this may confuse developers by making it seem like only 10 versions are available at all. This pr Adds a `...` at the end to indicate that more versions are available. For example: ``` Found 8+ results for "python": * python (3.13.0a6, 3.13.0a5, 3.13.0a3, 3.13.0a2, 3.13.0a1, 3.12.3, 3.12.2, 3.12.1, 3.12.0, 3.12.0rc3 ...) * python-cosmopolitan (3.6.14) * python-launcher (1.0.0) * python-qt (3.5.2, 3.5.1, 3.4.2, 3.3.0, 3.2) * python-swiftclient (4.2.0, 4.1.0, 4.0.0, 3.13.1, 3.13.0, 3.12.0) * python-language-server (2022-02-18, 2021-09-08, 2021-05-20, 2020-10-08, 2020-06-19, 2020-04-24) * python-matter-server (5.10.0, 5.9.0, 5.8.1, 5.8.0, 5.7.0b2, 5.5.3, 5.1.1, 5.0.3, 4.0.2, 4.0.1 ...) * python-full (3.13.0a6, 3.13.0a5, 3.13.0a3, 3.13.0a2, 3.13.0a1, 3.12.3, 3.12.2, 3.12.1, 3.12.0, 3.11.9 ...) Warning: Showing top 10 results and truncated versions. Use --show-all to show all. ``` ## How was it tested? Debugged locally, tested with `go` and `python` Also verified that the ellipses don't appear when the `--show-all` flag is set
Configuration menu - View commit details
-
Copy full SHA for a592fe5 - Browse repository at this point
Copy the full SHA a592fe5View commit details
Commits on May 14, 2024
-
Configuration menu - View commit details
-
Copy full SHA for 1a31e20 - Browse repository at this point
Copy the full SHA 1a31e20View commit details -
Add mongodb + rabbitmq docs (#2051)
## Summary Add documentation for RabbitMQ + MongoD ## How was it tested? Localhost + `npm run build` to test broken links
Configuration menu - View commit details
-
Copy full SHA for 6186add - Browse repository at this point
Copy the full SHA 6186addView commit details -
Fix errors on rabbitMQ + Index Page (#2052)
## Summary TSIA ## How was it tested? localhost + npm run build
Configuration menu - View commit details
-
Copy full SHA for 35b9265 - Browse repository at this point
Copy the full SHA 35b9265View commit details
Commits on May 15, 2024
-
Configuration menu - View commit details
-
Copy full SHA for d78aedf - Browse repository at this point
Copy the full SHA d78aedfView commit details -
Configuration menu - View commit details
-
Copy full SHA for c198a04 - Browse repository at this point
Copy the full SHA c198a04View commit details -
Revert "[perf] Skip cache check if store path exists locally" (#2054)
Reverts #2042 #2042 introduced a bug where adding non-cached packages would fail to add them to the flake on the first attempt. (it would succeed on subsequent attempts after the package is already in the store). The root cause is that `fetchNarInfoStatus` gets called twice per package (even though we cache it and we're not supposed to call it twice). The first call returns `false` because the package is not cached. The second call returns `true` because the package is already in store. This discrepancy essentially causes the package not to appear on the flake at all. When updating state again, the package is already in the nix store so both `fetchNarInfoStatus` calls return true. I think reverting this is best immediate course of action. In follow up we should fix `fetchNarInfoStatus` so it only gets called once (it will improve performance and is more correct). Later on we can think of better way to do #2042. The current implementation is a bit fragile and not 100% consistent with the initial intention of the function, so I'm concerned it can lead to more bugs in the future.
Configuration menu - View commit details
-
Copy full SHA for 132ad46 - Browse repository at this point
Copy the full SHA 132ad46View commit details
This comparison is taking too long to generate.
Unfortunately it looks like we can’t render this comparison for you right now. It might be too big, or there might be something weird with your repository.
You can try running this command locally to see the comparison on your machine:
git diff 0.10.6...0.10.7