This PR builds on #780 by making both version parsing faster, and
perhaps more importantly, making version comparisons much faster.
Overall, these changes result in a considerable improvement for the
`boto3.in` workload. Here's the status quo:
```
$ time puffin pip-compile --no-build --cache-dir ~/astral/tmp/cache/ -o /dev/null ./scripts/requirements/boto3.in
Resolved 31 packages in 34.56s
real 34.579
user 34.004
sys 0.413
maxmem 2867 MB
faults 0
```
And now with this PR:
```
$ time puffin pip-compile --no-build --cache-dir ~/astral/tmp/cache/ -o /dev/null ./scripts/requirements/boto3.in
Resolved 31 packages in 9.20s
real 9.218
user 8.919
sys 0.165
maxmem 463 MB
faults 0
```
This particular workload gets stuck in pubgrub doing resolution, and
thus benefits mightily from a faster `Version::cmp` routine. With that
said, this change does also help a fair bit with "normal" runs:
```
$ hyperfine -w10 \
"puffin-base pip-compile --cache-dir ~/astral/tmp/cache/ -o /dev/null ./scripts/benchmarks/requirements.in" \
"puffin-cmparc pip-compile --cache-dir ~/astral/tmp/cache/ -o /dev/null ./scripts/benchmarks/requirements.in"
Benchmark 1: puffin-base pip-compile --cache-dir ~/astral/tmp/cache/ -o /dev/null ./scripts/benchmarks/requirements.in
Time (mean ± σ): 337.5 ms ± 3.9 ms [User: 310.5 ms, System: 73.2 ms]
Range (min … max): 333.6 ms … 343.4 ms 10 runs
Benchmark 2: puffin-cmparc pip-compile --cache-dir ~/astral/tmp/cache/ -o /dev/null ./scripts/benchmarks/requirements.in
Time (mean ± σ): 189.8 ms ± 3.0 ms [User: 168.1 ms, System: 78.4 ms]
Range (min … max): 185.0 ms … 196.2 ms 15 runs
Summary
puffin-cmparc pip-compile --cache-dir ~/astral/tmp/cache/ -o /dev/null ./scripts/benchmarks/requirements.in ran
1.78 ± 0.03 times faster than puffin-base pip-compile --cache-dir ~/astral/tmp/cache/ -o /dev/null ./scripts/benchmarks/requirements.in
```
There is perhaps some future work here (detailed in the commit
messages), but I suspect it would be more fruitful to explore ways of
making resolution itself and/or deserialization faster.
Fixes#373, Closes#396
Currently, `dbg!` is hard to read because versions are verbose, showing
all optional fields, and we have a lot of versions. Changing debug
formatting to displaying the version number (which can be losslessly
converted to the struct and back) makes this more readable.
See e.g.
https://gist.github.com/konstin/38c0f32b109dffa73b3aa0ab86b9662b
**Before**
```text
version: Version {
epoch: 0,
release: [
1,
2,
3,
],
pre: None,
post: None,
dev: None,
local: None,
},
```
**After**
```text
version: "1.2.3",
```
Extends #517 with a suggestion from @konstin to parse the `SimpleJson`
into an intermediate type `SimpleMetadata(BTreeMap<Version,
VersionFiles>)` before converting to a `VersionMap`. This reduces the
number of times we need to parse the response. Additionally, we cache
the parsed response now instead of `SimpleJson`.
`VersionFiles` stores two vectors with
`WheelFilename`/`SourceDistFilename` and `File` tuples. These can be
iterated over together or separately. A new enum `DistFilename` was
added to capture the `SourceDistFilename` and `WheelFilename` variants
allowing iteration over both vectors.
## Summary
This PR uses the wheel stem (e.g., `foo-1.2.3-py3-none-any`) instead of
the wheel name (e.g., `foo-1.2.3-py3-none-any.whl`) when storing
unzipped wheels in the cache, which removes a class of confusing issues
around overwrites and directory-vs.-file collisions.
For now, we retain _both_ the zipped and unzipped wheels in the cache,
though we can easily change this by storing the zipped wheels in a
temporary directory.
Closes https://github.com/astral-sh/puffin/issues/573.
## Test Plan
Some examples from my local cache:
<img width="835" alt="Screen Shot 2023-12-05 at 4 09 55 PM"
src="784146aa-b080-416e-9767-40c843fe5d6a">
<img width="847" alt="Screen Shot 2023-12-05 at 4 12 14 PM"
src="4bc7f30f-bef3-47f1-b4e8-da9cabf87f28">
<img width="637" alt="Screen Shot 2023-12-05 at 4 09 50 PM"
src="25ca4944-4a06-4a08-ac85-c6f7d8b5c8ea">
## Summary
This PR implements logic to sort wheels by priority, where priority is
defined as preferring more "specific" wheels over less "specific"
wheels. For example, in the case of Black, my machine now selects
`black-23.11.0-cp311-cp311-macosx_11_0_arm64.whl`, whereas sorting by
lowest priority instead gives me `black-23.11.0-py3-none-any.whl`.
As part of this change, I've also modified the resolver to fallback to
using incompatible wheels when determining package metadata, if no
compatible wheels are available.
The `VersionMap` was also moved out of `resolver.rs` and into its own
file with a wrapper type, for clarity.
Closes https://github.com/astral-sh/puffin/issues/380.
Closes https://github.com/astral-sh/puffin/issues/421.
I intend this to become the main form of caching for puffin: You can
make http requests, you tranform the data to what you really need, you
have control over the cache key, and the cache is always json (or
anything else much faster we want to replace it with as long as it's
serde!)
This PR tweaks the representation of `Tags` in order to offer a
faster implementation of `WheelFilename::is_compatible`. We now use a
nested map of tags that lets us avoid looping over every supported
platform tag. As the code comments suggest, that is the essential gain.
We still do not mind looping over the tags in each wheel name since they
tend to be quite small. And pushing our thumb on that side of things can
make things worse overall since it would likely slow down WheelFilename
construction itself.
For micro-benchmarks, we improve considerably for compatibility
checking:
$ critcmp base test3
group base test3
----- ---- -----
build_platform_tags/burntsushi-archlinux 1.00 46.2±0.28µs ? ?/sec 2.48
114.8±0.45µs ? ?/sec
wheelname_parsing/flyte-long-compatible 1.00 624.8±3.31ns 174.0 MB/sec
1.01 629.4±4.30ns 172.7 MB/sec
wheelname_parsing/flyte-long-incompatible 1.00 743.6±4.23ns 165.4 MB/sec
1.00 746.9±4.62ns 164.7 MB/sec
wheelname_parsing/flyte-short-compatible 1.00 526.7±4.76ns 54.3 MB/sec
1.01 530.2±5.81ns 54.0 MB/sec
wheelname_parsing/flyte-short-incompatible 1.00 540.4±4.93ns 60.0 MB/sec
1.01 545.7±5.31ns 59.4 MB/sec
wheelname_parsing_failure/flyte-long-extension 1.00 13.6±0.13ns 3.2
GB/sec 1.01 13.7±0.14ns 3.2 GB/sec
wheelname_parsing_failure/flyte-short-extension 1.00 14.0±0.20ns 1160.4
MB/sec 1.01 14.1±0.14ns 1146.5 MB/sec
wheelname_tag_compatibility/flyte-long-compatible 11.33 159.8±2.79ns
680.5 MB/sec 1.00 14.1±0.23ns 7.5 GB/sec
wheelname_tag_compatibility/flyte-long-incompatible 237.60
1671.8±37.99ns 73.6 MB/sec 1.00 7.0±0.08ns 17.1 GB/sec
wheelname_tag_compatibility/flyte-short-compatible 16.07 223.5±8.60ns
128.0 MB/sec 1.00 13.9±0.30ns 2.0 GB/sec
wheelname_tag_compatibility/flyte-short-incompatible 149.83 628.3±2.13ns
51.6 MB/sec 1.00 4.2±0.10ns 7.6 GB/sec
We do regress slightly on the time it takes for `Tags::new` to run, but
this is somewhat expected. And in absolute terms, 114us is perfectly
acceptable given that it's only executed ~once for each `puffin`
invocation.
Ad hoc benchmarks indicate an overall 25% perf improvement in `puffin
pip-compile` times. This roughly corresponds with how much time
`is_compatible` was taking. Indeed, profiling confirms that it has
virtually disappeared from the profile.
Fixes#157
`PackageName` and `ExtraName` can now only be constructed from valid
names. They share the same rules, so i gave them the same
implementation. Constructors are split between `new` (owned) and
`from_str` (borrowed), with the owned version avoiding allocations.
Closes#279
---------
Co-authored-by: Zanie <contact@zanie.dev>
The normalized name abstractions were not consistently, this PR uses
them where they were previously missing:
* `WheelFilename::distribution`
* `Requirement::name`
* `Requirement::extras`
* `Metadata21::name`
* `Metadata21::provides_dist`
With `puffin-package` depending on `pep508_rs` this would be cyclical
crate dependency, so `puffin-normalize` gets split out from
`puffin-package`.
`DistInfoName` has the same task and semantics as `PackageName`, so it's
merged into the latter.
`PackageName` and `ExtraName` documentation is moved onto the type and
their constructors are called `new` instead of `normalize`. We now use
these constructors rarely enough the implicit allocation by
`to_string()` shouldn't matter anymore, while more actual cloning
becomes visible.
## Summary
This PR adds support for resolving and installing dependencies via
direct URLs, like:
```
werkzeug @ 960bb4017c/Werkzeug-2.0.0-py3-none-any.whl
```
These are fairly common (e.g., with `torch`), but you most often see
them as Git dependencies.
Broadly, structs like `RemoteDistribution` and friends are now enums
that can represent either registry-based dependencies or URL-based
dependencies:
```rust
/// A built distribution (wheel) that exists as a remote file (e.g., on `PyPI`).
#[derive(Debug, Clone)]
#[allow(clippy::large_enum_variant)]
pub enum RemoteDistribution {
/// The distribution exists in a registry, like `PyPI`.
Registry(PackageName, Version, File),
/// The distribution exists at an arbitrary URL.
Url(PackageName, Url),
}
```
In the resolver, we now allow packages to take on an extra, optional
`Url` field:
```rust
#[derive(Debug, Clone, Eq, Derivative)]
#[derivative(PartialEq, Hash)]
pub enum PubGrubPackage {
Root,
Package(
PackageName,
Option<DistInfoName>,
#[derivative(PartialEq = "ignore")]
#[derivative(PartialOrd = "ignore")]
#[derivative(Hash = "ignore")]
Option<Url>,
),
}
```
However, for the purpose of version satisfaction, we ignore the URL.
This allows for the URL dependency to satisfy the transitive request in
cases like:
```
flask==3.0.0
werkzeug @ 254c3e9b5f/werkzeug-3.0.1-py3-none-any.whl
```
There are a couple limitations in the current approach:
- The caching for remote URLs is done separately in the resolver vs. the
installer. I decided not to sweat this too much... We need to figure out
caching holistically.
- We don't support any sort of time-based cache for remote URLs -- they
just exist forever. This will be a problem for URL dependencies, where
we need some way to evict and refresh them. But I've deferred it for
now.
- I think I need to redo how this is modeled in the resolver, because
right now, we don't detect a variety of invalid cases, e.g., providing
two different URLs for a dependency, asking for a URL dependency and a
_different version_ of the same dependency in the list of first-party
dependencies, etc.
- (We don't yet support VCS dependencies.)
This also allows us to get rid of `PinnedPackage` _and_ to remove some
`Result<...>` types due to needless conversions between
otherwise-identical types.
The need for this became clear when working on the source distribution
integration into the resolver.
While at it i also switch the `WheelFilename` version to the parsed
`pep440_rs` version now that we have this crate.
