Reduce the overhead of `uv run` in large workspaces. Instead of
re-discovering the entire workspace each time we resolve the metadata of
a member, we can the discovered set of workspace members. Care needs to
be taken to not cache the discovery for `uv init`, `uv add` and `uv
remove`, which change the definitions of workspace members.
Below is apache airflow e3fe06382df4b19f2c0de40ce7c0bdc726754c74 `uv run
python` with a minimal payload. With this change, we avoid a ~350ms
overhead of each `uv run` invocation.
```
$ hyperfine --warmup 2 \
"uv run --no-dev python -c \"print('hi')\"" \
"uv-profiling run --no-dev python -c \"print('hi')\""
Benchmark 1: uv run --no-dev python -c "print('hi')"
Time (mean ± σ): 492.6 ms ± 7.0 ms [User: 393.2 ms, System: 97.1 ms]
Range (min … max): 482.3 ms … 501.5 ms 10 runs
Benchmark 2: uv-profiling run --no-dev python -c "print('hi')"
Time (mean ± σ): 129.7 ms ± 2.5 ms [User: 105.4 ms, System: 23.2 ms]
Range (min … max): 126.0 ms … 136.1 ms 22 runs
Summary
uv-profiling run --no-dev python -c "print('hi')" ran
3.80 ± 0.09 times faster than uv run --no-dev python -c "print('hi')"
```
The profile after those change below. We still spend a large chunk in
toml parsing (both `uv.lock` and `pyproject.toml`), but it's not
excessive anymore.
Build failures are one of the most common user facing failures that
aren't "obivous" errors (such as typos) or resolver errors. Currently,
they show more technical details than being focussed on this being an
error in a subprocess that is either on the side of the package or -
more likely - in the build environment, e.g. the user needs to install a
dev package or their python version is incompatible.
The new error message clearly delineates the part that's important (this
is a build backend problem) from the internals (we called this hook) and
is consistent about which part of the dist building stage failed. We
have to calibrate the exact wording of the error message some more. Most
of the implementation is working around the orphan rule, (this)error
rules and trait rules, so it came out more of a refactoring than
intended.
Example:
This is like #9556, but at the level of all other builds, including the
resolver and installer. Going through PEP 517 to build a package is
slow, so when building a package with the uv build backend, we can call
into the uv build backend directly instead: No temporary virtual env, no
temp venv sync, no python subprocess calls, no uv subprocess calls.
This fast path is gated through preview. Since the uv wheel is not
available at test time, I've manually confirmed the feature by comparing
`uv venv && cargo run pip install . -v --preview --reinstall .` and `uv
venv && cargo run pip install . -v --reinstall .`. When hacking the
preview so that the python uv build backend works without the setting
the direct build also (wheel built with `maturin build --profile
profiling`), we can see the perfomance difference:
```
$ hyperfine --prepare "uv venv" --warmup 3 \
"UV_PREVIEW=1 target/profiling/uv pip install --no-deps --reinstall scripts/packages/built-by-uv --preview" \
"target/profiling/uv pip install --no-deps --reinstall scripts/packages/built-by-uv --find-links target/wheels/"
Benchmark 1: UV_PREVIEW=1 target/profiling/uv pip install --no-deps --reinstall scripts/packages/built-by-uv --preview
Time (mean ± σ): 33.1 ms ± 2.5 ms [User: 25.7 ms, System: 13.0 ms]
Range (min … max): 29.8 ms … 47.3 ms 73 runs
Benchmark 2: target/profiling/uv pip install --no-deps --reinstall scripts/packages/built-by-uv --find-links target/wheels/
Time (mean ± σ): 115.1 ms ± 4.3 ms [User: 54.0 ms, System: 27.0 ms]
Range (min … max): 109.2 ms … 123.8 ms 25 runs
Summary
UV_PREVIEW=1 target/profiling/uv pip install --no-deps --reinstall scripts/packages/built-by-uv --preview ran
3.48 ± 0.29 times faster than target/profiling/uv pip install --no-deps --reinstall scripts/packages/built-by-uv --find-links target/wheels/
```
Do we need a global option to disable the fast path? There is one for
`uv build` because `--force-pep517` moves `uv build` much closer to a
`pip install` from source that a user of a library would experience (See
discussion at #9610), but uv overall doesn't really make guarantees
around the build env of dependencies, so I consider the direct build a
valid option.
Best reviewed commit-by-commit, only the last commit is the actual
implementation, while the preview mode introduction is just a
refactoring touching too many files.
As per
https://matklad.github.io/2021/02/27/delete-cargo-integration-tests.html
Before that, there were 91 separate integration tests binary.
(As discussed on Discord — I've done the `uv` crate, there's still a few
more commits coming before this is mergeable, and I want to see how it
performs in CI and locally).
uv will soon support both a build frontend (`uv build`) and a build
backend (`build-system = "uv"`). To avoid the name clash, I'm renaming
the `uv-build` crate to `uv-build-frontend`. In a follow-up PR, I will
add a `uv-build-backend` crate with the build backend implementation.
Whew this is a lot.
The user-facing changes are:
- `uv toolchain` to `uv python` e.g. `uv python find`, `uv python
install`, ...
- `UV_TOOLCHAIN_DIR` to` UV_PYTHON_INSTALL_DIR`
- `<UV_STATE_DIR>/toolchains` to `<UV_STATE_DIR>/python` (with
[automatic
migration](https://github.com/astral-sh/uv/pull/4735/files#r1663029330))
- User-facing messages no longer refer to toolchains, instead using
"Python", "Python versions" or "Python installations"
The internal changes are:
- `uv-toolchain` crate to `uv-python`
- `Toolchain` no longer referenced in type names
- Dropped unused `SystemPython` type (previously replaced)
- Clarified the type names for "managed Python installations"
- (more little things)
## Summary
This PR removes the static resolver map:
```rust
static RESOLVED_GIT_REFS: Lazy<Mutex<FxHashMap<RepositoryReference, GitSha>>> =
Lazy::new(Mutex::default);
```
With a `GitResolver` struct that we now pass around on the
`BuildContext`. There should be no behavior changes here; it's purely an
internal refactor with an eye towards making it cleaner for us to
"pre-populate" the list of resolved SHAs.
With the change, we remove the special casing of workspace dependencies
and resolve `tool.uv` for all git and directory distributions. This
gives us support for non-editable workspace dependencies and path
dependencies in other workspaces. It removes a lot of special casing
around workspaces. These changes are the groundwork for supporting
`tool.uv` with dynamic metadata.
The basis for this change is moving `Requirement` from
`distribution-types` to `pypi-types` and the lowering logic from
`uv-requirements` to `uv-distribution`. This changes should be split out
in separate PRs.
I've included an example workspace `albatross-root-workspace2` where
`bird-feeder` depends on `a` from another workspace `ab`. There's a
bunch of failing tests and regressed error messages that still need
fixing. It does fix the audited package count for the workspace tests.
## Summary
This PR consolidates the concurrency limits used throughout `uv` and
exposes two limits, `UV_CONCURRENT_DOWNLOADS` and
`UV_CONCURRENT_BUILDS`, as environment variables.
Currently, `uv` has a number of concurrent streams that it buffers using
relatively arbitrary limits for backpressure. However, many of these
limits are conflated. We run a relatively small number of tasks overall
and should start most things as soon as possible. What we really want to
limit are three separate operations:
- File I/O. This is managed by tokio's blocking pool and we should not
really have to worry about it.
- Network I/O.
- Python build processes.
Because the current limits span a broad range of tasks, it's possible
that a limit meant for network I/O is occupied by tasks performing
builds, reading from the file system, or even waiting on a `OnceMap`. We
also don't limit build processes that end up being required to perform a
download. While this may not pose a performance problem because our
limits are relatively high, it does mean that the limits do not do what
we want, making it tricky to expose them to users
(https://github.com/astral-sh/uv/issues/1205,
https://github.com/astral-sh/uv/issues/3311).
After this change, the limits on network I/O and build processes are
centralized and managed by semaphores. All other tasks are unbuffered
(note that these tasks are still bounded, so backpressure should not be
a problem).
Needed to prevent circular dependencies in my toolchain work (#2931). I
think this is probably a reasonable change as we move towards persistent
configuration too?
Unfortunately `BuildIsolation` needs to be in `uv-types` to avoid
circular dependencies still. We might be able to resolve that in the
future.
Previously, we did not consider installed distributions as candidates
while performing resolution. Here, we update the resolver to use
installed distributions that satisfy requirements instead of pulling new
distributions from the registry.
The implementation details are as follows:
- We now provide `SitePackages` to the `CandidateSelector`
- If an installed distribution satisfies the requirement, we prefer it
over remote distributions
- We do not want to allow installed distributions in some cases, i.e.,
upgrade and reinstall
- We address this by introducing an `Exclusions` type which tracks
installed packages to ignore during selection
- There's a new `ResolvedDist` wrapper with `Installed(InstalledDist)`
and `Installable(Dist)` variants
- This lets us pass already installed distributions throughout the
resolver
The user-facing behavior is thoroughly covered in the tests, but
briefly:
- Installing a package that depends on an already-installed package
prefers the local version over the index
- Installing a package with a name that matches an already-installed URL
package does not reinstall from the index
- Reinstalling (--reinstall) a package by name _will_ pull from the
index even if an already-installed URL package is present
- To reinstall the URL package, you must specify the URL in the request
Closes https://github.com/astral-sh/uv/issues/1661
Addresses:
- https://github.com/astral-sh/uv/issues/1476
- https://github.com/astral-sh/uv/issues/1856
- https://github.com/astral-sh/uv/issues/2093
- https://github.com/astral-sh/uv/issues/2282
- https://github.com/astral-sh/uv/issues/2383
- https://github.com/astral-sh/uv/issues/2560
## Test plan
- [x] Reproduction at `charlesnicholson/uv-pep420-bug` passes
- [x] Unit test for editable package
([#1476](https://github.com/astral-sh/uv/issues/1476))
- [x] Unit test for previously installed package with empty registry
- [x] Unit test for local non-editable package
- [x] Unit test for new version available locally but not in registry
([#2093](https://github.com/astral-sh/uv/issues/2093))
- ~[ ] Unit test for wheel not available in registry but already
installed locally
([#2282](https://github.com/astral-sh/uv/issues/2282))~ (seems
complicated and not worthwhile)
- [x] Unit test for install from URL dependency then with matching
version ([#2383](https://github.com/astral-sh/uv/issues/2383))
- [x] Unit test for install of new package that depends on installed
package does not change version
([#2560](https://github.com/astral-sh/uv/issues/2560))
- [x] Unit test that `pip compile` does _not_ consider installed
packages
This is driving me a little crazy and is becoming a larger problem in
#2596 where I need to move more types (like `Upgrade` and `Reinstall`)
into this crate. Anything that's shared across our core resolver,
install, and build crates needs to be defined in this crate to avoid
cyclic dependencies. We've outgrown it being a single file with some
shared traits.
There are no behavioral changes here.
Scott schafer got me the idea: We can avoid repeating the path for
workspaces dependencies everywhere if we declare them in the virtual
package once and treat them as workspace dependencies from there on.
## Summary
I tried out `cargo shear` to see if there are any unused dependencies
that `cargo udeps` isn't reporting. It turned out, there are a few. This
PR removes those dependencies.
## Test Plan
`cargo build`
The architecture of uv does not necessarily match that of the python
interpreter (#2326). In cross compiling/testing scenarios the operating
system can also mismatch. To solve this, we move arch and os detection
to python, vendoring the relevant pypa/packaging code, preventing
mismatches between what the python interpreter was compiled for and what
uv was compiled for.
To make the scripts more manageable, they are now a directory in a
tempdir and we run them with `python -m` . I've simplified the
pypa/packaging code since we're still building the tags in rust. A
`Platform` is now instantiated by querying the python interpreter for
its platform. The pypa/packaging files are copied verbatim for easier
updates except a `lru_cache()` python 3.7 backport.
Error handling is done by a `"result": "success|error"` field that allow
passing error details to rust:
```console
$ uv venv --no-cache
× Can't use Python at `/home/konsti/projects/uv/.venv/bin/python3`
╰─▶ Unknown operation system `linux`
```
I've used the [maturin sysconfig
collection](855f6d2cb1/sysconfig)
as reference. I'm unsure how to test these changes across the wide
variety of platforms.
Fixes#2326
<!--
Thank you for contributing to uv! To help us out with reviewing, please
consider the following:
- Does this pull request include a summary of the change? (See below.)
- Does this pull request include a descriptive title?
- Does this pull request include references to any relevant issues?
-->
## Summary
With this PR I've added the option environment variables to the wheel
building process, through the `BuildDispatch`. When integrating uv with
our project pixi (https://github.com/prefix-dev/pixi/pull/863). We ran
into this missing requirement, I've made a rough version here, could
maybe use some refinement.
### Why do we need this?
Because pixi allow the user to use a conda activated prefix for wheel
building, this comes with a number of environment variables, like `PATH`
but also `CONDA_PREFIX` amongst others. This allows the user to use
system dependencies from conda-forge to use during an sdist build.
Because we use `uv` as a library we need to pass in the options
programatically. Additionally, in general there is nothing holding a
python sdist back from actually depending on an environment variable,
see
e.g the test package: https://pypi.org/project/env-test-package/
### What about `ConfigSettings`
I think `ConfigSettings` does not suffice because e.g. CMake could
function differently when the `CONDA_PREFIX` is set. Also, we do not
know if the user supplied backend actually support these settings.
### Path handling
Because the user can now also supply a PATH in the environment map, the
logic I had was the following, I format the path so that it has the
following precedence
1. venv scripts dir.
2. user supplied path.
3. system path.
### Improvements
There is some path modification and copying happening everytime we use
the `run_python_script` function, I think we could improve this but
would like some pointers where to best put the maybe split and cached
version, we might also want to use some types to split these things up.
### Finally
I did not add any of these options to the uv executables, I first would
like to know if this is a direction we would want to go in. I'm happy to
do this or make any changes that you feel would benefit this project.
Also tagging @wolfv to keep track of this as well.
## Test Plan
<!-- How was it tested? -->
---------
Co-authored-by: konsti <konstin@mailbox.org>
First, replace all usages in files in-place. I used my editor for this.
If someone wants to add a one-liner that'd be fun.
Then, update directory and file names:
```
# Run twice for nested directories
find . -type d -print0 | xargs -0 rename s/puffin/uv/g
find . -type d -print0 | xargs -0 rename s/puffin/uv/g
# Update files
find . -type f -print0 | xargs -0 rename s/puffin/uv/g
```
Then add all the files again
```
# Add all the files again
git add crates
git add python/uv
# This one needs a force-add
git add -f crates/uv-trampoline
```
2024-02-15 11:19:46 -06:00
Renamed from crates/puffin-dispatch/Cargo.toml (Browse further)
