konsti
47fc90d1b3
This is https://github.com/astral-sh/puffin/pull/947 again but this time merging into main instead of downstack, sorry for the noise. --- Windows has a default stack size of 1MB, which makes puffin often fail with stack overflows. The PR reduces stack size by three changes: * Boxing `File` in `Dist`, reducing the size from 496 to 240. * Boxing the largest futures. * Boxing `CachePolicy` ## Method Debugging happened on linux using https://github.com/astral-sh/puffin/pull/941 to limit the stack size to 1MB. Used ran the command below. ``` RUSTFLAGS=-Zprint-type-sizes cargo +nightly build -p puffin-cli -j 1 > type-sizes.txt && top-type-sizes -w -s -h 10 < type-sizes.txt > sizes.txt ``` The main drawback is top-type-sizes not saying what the `__awaitee` is, so it requires manually looking up with a future with matching size. When the `brotli` features on `reqwest` is active, a lot of brotli types show up. Toggling this feature however seems to have no effect. I assume they are false positives since the `brotli` crate has elaborate control about allocation. The sizes are therefore shown with the feature off. ## Results The largest future goes from 12208B to 6416B, the largest type (`PrioritizedDistribution`, see also #948) from 17448B to 9264B. Full diff: https://gist.github.com/konstin/62635c0d12110a616a1b2bfcde21304f For the second commit, i iteratively boxed the largest file until the tests passed, then with an 800KB stack limit looked through the backtrace of a failing test and added some more boxing. Quick benchmarking showed no difference: ```console $ hyperfine --warmup 2 "target/profiling/main-dev resolve meine_stadt_transparent" "target/profiling/puffin-dev resolve meine_stadt_transparent" Benchmark 1: target/profiling/main-dev resolve meine_stadt_transparent Time (mean ± σ): 49.2 ms ± 3.0 ms [User: 39.8 ms, System: 24.0 ms] Range (min … max): 46.6 ms … 63.0 ms 55 runs Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options. Benchmark 2: target/profiling/puffin-dev resolve meine_stadt_transparent Time (mean ± σ): 47.4 ms ± 3.2 ms [User: 41.3 ms, System: 20.6 ms] Range (min … max): 44.6 ms … 60.5 ms 62 runs Warning: Statistical outliers were detected. Consider re-running this benchmark on a quiet system without any interferences from other programs. It might help to use the '--warmup' or '--prepare' options. Summary target/profiling/puffin-dev resolve meine_stadt_transparent ran 1.04 ± 0.09 times faster than target/profiling/main-dev resolve meine_stadt_transparent ```
1058 lines
39 KiB
Rust
1058 lines
39 KiB
Rust
//! Fetch and build source distributions from remote sources.
|
|
|
|
use std::path::{Path, PathBuf};
|
|
use std::str::FromStr;
|
|
use std::sync::Arc;
|
|
|
|
use anyhow::Result;
|
|
use fs_err::tokio as fs;
|
|
use futures::{FutureExt, TryStreamExt};
|
|
use reqwest::Response;
|
|
use tempfile::TempDir;
|
|
use tokio_util::compat::FuturesAsyncReadCompatExt;
|
|
use tracing::{debug, info_span, instrument, warn, Instrument};
|
|
use url::Url;
|
|
use zip::ZipArchive;
|
|
|
|
use distribution_filename::WheelFilename;
|
|
use distribution_types::{
|
|
DirectArchiveUrl, DirectGitUrl, Dist, FileLocation, GitSourceDist, Identifier, LocalEditable,
|
|
Name, PathSourceDist, RemoteSource, SourceDist,
|
|
};
|
|
use install_wheel_rs::read_dist_info;
|
|
use pep508_rs::VerbatimUrl;
|
|
use platform_tags::Tags;
|
|
use puffin_cache::{CacheBucket, CacheEntry, CacheShard, CachedByTimestamp, WheelCache};
|
|
use puffin_client::{CachedClient, CachedClientError, DataWithCachePolicy};
|
|
use puffin_fs::{write_atomic, LockedFile};
|
|
use puffin_git::{Fetch, GitSource};
|
|
use puffin_traits::{BuildContext, BuildKind, SourceBuildTrait};
|
|
use pypi_types::Metadata21;
|
|
|
|
use crate::reporter::Facade;
|
|
use crate::source::built_wheel_metadata::BuiltWheelMetadata;
|
|
pub use crate::source::error::SourceDistError;
|
|
use crate::source::manifest::{DiskFilenameAndMetadata, Manifest};
|
|
use crate::Reporter;
|
|
|
|
mod built_wheel_metadata;
|
|
mod error;
|
|
mod manifest;
|
|
|
|
/// Fetch and build a source distribution from a remote source, or from a local cache.
|
|
pub struct SourceDistCachedBuilder<'a, T: BuildContext> {
|
|
build_context: &'a T,
|
|
cached_client: &'a CachedClient,
|
|
reporter: Option<Arc<dyn Reporter>>,
|
|
tags: &'a Tags,
|
|
}
|
|
|
|
/// The name of the file that contains the cached metadata, encoded via `MsgPack`.
|
|
const METADATA: &str = "metadata.msgpack";
|
|
|
|
impl<'a, T: BuildContext> SourceDistCachedBuilder<'a, T> {
|
|
/// Initialize a [`SourceDistCachedBuilder`] from a [`BuildContext`].
|
|
pub fn new(build_context: &'a T, cached_client: &'a CachedClient, tags: &'a Tags) -> Self {
|
|
Self {
|
|
build_context,
|
|
reporter: None,
|
|
cached_client,
|
|
tags,
|
|
}
|
|
}
|
|
|
|
/// Set the [`Reporter`] to use for this source distribution fetcher.
|
|
#[must_use]
|
|
pub fn with_reporter(self, reporter: Arc<dyn Reporter>) -> Self {
|
|
Self {
|
|
reporter: Some(reporter),
|
|
..self
|
|
}
|
|
}
|
|
|
|
/// Download and build a [`SourceDist`].
|
|
pub async fn download_and_build(
|
|
&self,
|
|
source_dist: &SourceDist,
|
|
) -> Result<BuiltWheelMetadata, SourceDistError> {
|
|
let built_wheel_metadata = match &source_dist {
|
|
SourceDist::DirectUrl(direct_url_source_dist) => {
|
|
let filename = direct_url_source_dist
|
|
.filename()
|
|
.expect("Distribution must have a filename");
|
|
let DirectArchiveUrl { url, subdirectory } =
|
|
DirectArchiveUrl::from(direct_url_source_dist.url.raw());
|
|
|
|
// For direct URLs, cache directly under the hash of the URL itself.
|
|
let cache_shard = self.build_context.cache().shard(
|
|
CacheBucket::BuiltWheels,
|
|
WheelCache::Url(&url).remote_wheel_dir(direct_url_source_dist.name().as_ref()),
|
|
);
|
|
|
|
self.url(
|
|
source_dist,
|
|
filename,
|
|
&url,
|
|
&cache_shard,
|
|
subdirectory.as_deref(),
|
|
)
|
|
.boxed()
|
|
.await?
|
|
}
|
|
SourceDist::Registry(registry_source_dist) => {
|
|
let url = match ®istry_source_dist.file.url {
|
|
FileLocation::RelativeUrl(base, url) => base
|
|
.join_relative(url)
|
|
.map_err(|err| SourceDistError::UrlParse(url.clone(), err))?,
|
|
FileLocation::AbsoluteUrl(url) => Url::parse(url)
|
|
.map_err(|err| SourceDistError::UrlParse(url.clone(), err))?,
|
|
FileLocation::Path(path) => {
|
|
let path_source_dist = PathSourceDist {
|
|
name: registry_source_dist.filename.name.clone(),
|
|
url: VerbatimUrl::unknown(
|
|
Url::from_file_path(path).expect("path is absolute"),
|
|
),
|
|
path: path.clone(),
|
|
editable: false,
|
|
};
|
|
return self.path(source_dist, &path_source_dist).await;
|
|
}
|
|
};
|
|
|
|
// For registry source distributions, shard by package, then by SHA.
|
|
// Ex) `pypi/requests/a673187abc19fe6c`
|
|
let cache_shard = self.build_context.cache().shard(
|
|
CacheBucket::BuiltWheels,
|
|
WheelCache::Index(®istry_source_dist.index)
|
|
.remote_wheel_dir(registry_source_dist.filename.name.as_ref())
|
|
.join(®istry_source_dist.file.distribution_id().as_str()[..16]),
|
|
);
|
|
|
|
self.url(
|
|
source_dist,
|
|
®istry_source_dist.file.filename,
|
|
&url,
|
|
&cache_shard,
|
|
None,
|
|
)
|
|
.boxed()
|
|
.await?
|
|
}
|
|
SourceDist::Git(git_source_dist) => self.git(source_dist, git_source_dist).await?,
|
|
SourceDist::Path(path_source_dist) => self.path(source_dist, path_source_dist).await?,
|
|
};
|
|
|
|
Ok(built_wheel_metadata)
|
|
}
|
|
|
|
/// Download a [`SourceDist`] and determine its metadata. This typically involves building the
|
|
/// source distribution into a wheel; however, some build backends support determining the
|
|
/// metadata without building the source distribution.
|
|
pub async fn download_and_build_metadata(
|
|
&self,
|
|
source_dist: &SourceDist,
|
|
) -> Result<Metadata21, SourceDistError> {
|
|
let metadata = match &source_dist {
|
|
SourceDist::DirectUrl(direct_url_source_dist) => {
|
|
let filename = direct_url_source_dist
|
|
.filename()
|
|
.expect("Distribution must have a filename");
|
|
let DirectArchiveUrl { url, subdirectory } =
|
|
DirectArchiveUrl::from(direct_url_source_dist.url.raw());
|
|
|
|
// For direct URLs, cache directly under the hash of the URL itself.
|
|
let cache_shard = self.build_context.cache().shard(
|
|
CacheBucket::BuiltWheels,
|
|
WheelCache::Url(&url).remote_wheel_dir(direct_url_source_dist.name().as_ref()),
|
|
);
|
|
|
|
self.url_metadata(
|
|
source_dist,
|
|
filename,
|
|
&url,
|
|
&cache_shard,
|
|
subdirectory.as_deref(),
|
|
)
|
|
.boxed()
|
|
.await?
|
|
}
|
|
SourceDist::Registry(registry_source_dist) => {
|
|
let url = match ®istry_source_dist.file.url {
|
|
FileLocation::RelativeUrl(base, url) => base
|
|
.join_relative(url)
|
|
.map_err(|err| SourceDistError::UrlParse(url.clone(), err))?,
|
|
FileLocation::AbsoluteUrl(url) => Url::parse(url)
|
|
.map_err(|err| SourceDistError::UrlParse(url.clone(), err))?,
|
|
FileLocation::Path(path) => {
|
|
let path_source_dist = PathSourceDist {
|
|
name: registry_source_dist.filename.name.clone(),
|
|
url: VerbatimUrl::unknown(
|
|
Url::from_file_path(path).expect("path is absolute"),
|
|
),
|
|
path: path.clone(),
|
|
editable: false,
|
|
};
|
|
return self
|
|
.path_metadata(source_dist, &path_source_dist)
|
|
.boxed()
|
|
.await;
|
|
}
|
|
};
|
|
|
|
// For registry source distributions, shard by package, then by SHA.
|
|
// Ex) `pypi/requests/a673187abc19fe6c`
|
|
let cache_shard = self.build_context.cache().shard(
|
|
CacheBucket::BuiltWheels,
|
|
WheelCache::Index(®istry_source_dist.index)
|
|
.remote_wheel_dir(registry_source_dist.filename.name.as_ref())
|
|
.join(®istry_source_dist.file.distribution_id().as_str()[..16]),
|
|
);
|
|
|
|
self.url_metadata(
|
|
source_dist,
|
|
®istry_source_dist.file.filename,
|
|
&url,
|
|
&cache_shard,
|
|
None,
|
|
)
|
|
.boxed()
|
|
.await?
|
|
}
|
|
SourceDist::Git(git_source_dist) => {
|
|
self.git_metadata(source_dist, git_source_dist)
|
|
.boxed()
|
|
.await?
|
|
}
|
|
SourceDist::Path(path_source_dist) => {
|
|
self.path_metadata(source_dist, path_source_dist)
|
|
.boxed()
|
|
.await?
|
|
}
|
|
};
|
|
|
|
Ok(metadata)
|
|
}
|
|
|
|
/// Build a source distribution from a remote URL.
|
|
#[allow(clippy::too_many_arguments)]
|
|
async fn url<'data>(
|
|
&self,
|
|
source_dist: &'data SourceDist,
|
|
filename: &'data str,
|
|
url: &'data Url,
|
|
cache_shard: &CacheShard,
|
|
subdirectory: Option<&'data Path>,
|
|
) -> Result<BuiltWheelMetadata, SourceDistError> {
|
|
let cache_entry = cache_shard.entry(METADATA);
|
|
|
|
let download = |response| {
|
|
async {
|
|
// At this point, we're seeing a new or updated source distribution; delete all
|
|
// wheels, and redownload.
|
|
match fs::remove_dir_all(&cache_entry.dir()).await {
|
|
Ok(()) => debug!("Cleared built wheels and metadata for {source_dist}"),
|
|
Err(err) if err.kind() == std::io::ErrorKind::NotFound => (),
|
|
Err(err) => return Err(err.into()),
|
|
}
|
|
|
|
debug!("Downloading source distribution: {source_dist}");
|
|
|
|
// Download the source distribution.
|
|
let source_dist_entry = cache_shard.entry(filename);
|
|
self.persist_source_dist_url(response, source_dist, filename, &source_dist_entry)
|
|
.await?;
|
|
|
|
Ok(Manifest::default())
|
|
}
|
|
.instrument(info_span!("download", source_dist = %source_dist))
|
|
};
|
|
let req = self.cached_client.uncached().get(url.clone()).build()?;
|
|
let manifest = self
|
|
.cached_client
|
|
.get_cached_with_callback(req, &cache_entry, download)
|
|
.await
|
|
.map_err(|err| match err {
|
|
CachedClientError::Callback(err) => err,
|
|
CachedClientError::Client(err) => SourceDistError::Client(err),
|
|
})?;
|
|
|
|
// If the cache contains a compatible wheel, return it.
|
|
if let Some(built_wheel) =
|
|
BuiltWheelMetadata::find_in_cache(self.tags, &manifest, &cache_entry)
|
|
{
|
|
return Ok(built_wheel);
|
|
}
|
|
|
|
// At this point, we're seeing cached metadata (as in, we have an up-to-date source
|
|
// distribution), but the wheel(s) we built previously are incompatible.
|
|
let task = self
|
|
.reporter
|
|
.as_ref()
|
|
.map(|reporter| reporter.on_build_start(source_dist));
|
|
|
|
// Build the source distribution.
|
|
let source_dist_entry = cache_shard.entry(filename);
|
|
let (disk_filename, wheel_filename, metadata) = self
|
|
.build_source_dist(
|
|
source_dist,
|
|
source_dist_entry.path(),
|
|
subdirectory,
|
|
&cache_entry,
|
|
)
|
|
.await?;
|
|
|
|
let cached_data = DiskFilenameAndMetadata {
|
|
disk_filename: disk_filename.clone(),
|
|
metadata: metadata.clone(),
|
|
};
|
|
|
|
// Not elegant that we have to read again here, but also not too relevant given that we
|
|
// have to build a source dist next.
|
|
// Just return if the response wasn't cacheable or there was another errors that
|
|
// `CachedClient` already complained about
|
|
if let Ok(cached) = fs::read(cache_entry.path()).await {
|
|
// If the file exists and it was just read or written by `CachedClient`, we assume it
|
|
// must be correct.
|
|
let mut cached = rmp_serde::from_slice::<DataWithCachePolicy<Manifest>>(&cached)?;
|
|
|
|
cached
|
|
.data
|
|
.insert_wheel(wheel_filename.clone(), cached_data.clone());
|
|
write_atomic(cache_entry.path(), rmp_serde::to_vec(&cached)?).await?;
|
|
};
|
|
|
|
if let Some(task) = task {
|
|
if let Some(reporter) = self.reporter.as_ref() {
|
|
reporter.on_build_complete(source_dist, task);
|
|
}
|
|
}
|
|
|
|
Ok(BuiltWheelMetadata::from_cached(
|
|
wheel_filename,
|
|
cached_data,
|
|
&cache_entry,
|
|
))
|
|
}
|
|
|
|
/// Build the source distribution's metadata from a local path.
|
|
///
|
|
/// If the build backend supports `prepare_metadata_for_build_wheel`, this method will avoid
|
|
/// building the wheel.
|
|
#[allow(clippy::too_many_arguments)]
|
|
async fn url_metadata<'data>(
|
|
&self,
|
|
source_dist: &'data SourceDist,
|
|
filename: &'data str,
|
|
url: &'data Url,
|
|
cache_shard: &CacheShard,
|
|
subdirectory: Option<&'data Path>,
|
|
) -> Result<Metadata21, SourceDistError> {
|
|
let cache_entry = cache_shard.entry(METADATA);
|
|
|
|
let download = |response| {
|
|
async {
|
|
// At this point, we're seeing a new or updated source distribution; delete all
|
|
// wheels, and redownload.
|
|
match fs::remove_dir_all(&cache_entry.dir()).await {
|
|
Ok(()) => debug!("Cleared built wheels and metadata for {source_dist}"),
|
|
Err(err) if err.kind() == std::io::ErrorKind::NotFound => (),
|
|
Err(err) => return Err(err.into()),
|
|
}
|
|
|
|
debug!("Downloading source distribution: {source_dist}");
|
|
|
|
// Download the source distribution.
|
|
let source_dist_entry = cache_shard.entry(filename);
|
|
self.persist_source_dist_url(response, source_dist, filename, &source_dist_entry)
|
|
.await?;
|
|
|
|
Ok(Manifest::default())
|
|
}
|
|
.instrument(info_span!("download", source_dist = %source_dist))
|
|
};
|
|
let req = self.cached_client.uncached().get(url.clone()).build()?;
|
|
let manifest = self
|
|
.cached_client
|
|
.get_cached_with_callback(req, &cache_entry, download)
|
|
.await
|
|
.map_err(|err| match err {
|
|
CachedClientError::Callback(err) => err,
|
|
CachedClientError::Client(err) => SourceDistError::Client(err),
|
|
})?;
|
|
|
|
// If the cache contains compatible metadata, return it.
|
|
if let Some(metadata) = manifest.find_metadata() {
|
|
return Ok(metadata.clone());
|
|
}
|
|
|
|
// Otherwise, we either need to build the metadata or the wheel.
|
|
let source_dist_entry = cache_shard.entry(filename);
|
|
|
|
// If the backend supports `prepare_metadata_for_build_wheel`, use it.
|
|
if let Some(metadata) = self
|
|
.build_source_dist_metadata(source_dist, source_dist_entry.path(), subdirectory)
|
|
.boxed()
|
|
.await?
|
|
{
|
|
if let Ok(cached) = fs::read(cache_entry.path()).await {
|
|
let mut cached = rmp_serde::from_slice::<DataWithCachePolicy<Manifest>>(&cached)?;
|
|
|
|
cached.data.set_metadata(metadata.clone());
|
|
write_atomic(cache_entry.path(), rmp_serde::to_vec(&cached)?).await?;
|
|
};
|
|
|
|
return Ok(metadata);
|
|
}
|
|
|
|
// At this point, we're seeing cached metadata (as in, we have an up-to-date source
|
|
// distribution), but the wheel(s) we built previously are incompatible.
|
|
let task = self
|
|
.reporter
|
|
.as_ref()
|
|
.map(|reporter| reporter.on_build_start(source_dist));
|
|
|
|
// Build the source distribution.
|
|
let (disk_filename, wheel_filename, metadata) = self
|
|
.build_source_dist(
|
|
source_dist,
|
|
source_dist_entry.path(),
|
|
subdirectory,
|
|
&cache_entry,
|
|
)
|
|
.await?;
|
|
|
|
// Not elegant that we have to read again here, but also not too relevant given that we
|
|
// have to build a source dist next.
|
|
// Just return if the response wasn't cacheable or there was another errors that
|
|
// `CachedClient` already complained about
|
|
if let Ok(cached) = fs::read(cache_entry.path()).await {
|
|
// If the file exists and it was just read or written by `CachedClient`, we assume it
|
|
// must be correct.
|
|
let mut cached = rmp_serde::from_slice::<DataWithCachePolicy<Manifest>>(&cached)?;
|
|
|
|
cached.data.insert_wheel(
|
|
wheel_filename.clone(),
|
|
DiskFilenameAndMetadata {
|
|
disk_filename: disk_filename.clone(),
|
|
metadata: metadata.clone(),
|
|
},
|
|
);
|
|
write_atomic(cache_entry.path(), rmp_serde::to_vec(&cached)?).await?;
|
|
};
|
|
|
|
if let Some(task) = task {
|
|
if let Some(reporter) = self.reporter.as_ref() {
|
|
reporter.on_build_complete(source_dist, task);
|
|
}
|
|
}
|
|
|
|
Ok(metadata)
|
|
}
|
|
|
|
/// Build a source distribution from a local path.
|
|
async fn path(
|
|
&self,
|
|
source_dist: &SourceDist,
|
|
path_source_dist: &PathSourceDist,
|
|
) -> Result<BuiltWheelMetadata, SourceDistError> {
|
|
let cache_entry = self.build_context.cache().entry(
|
|
CacheBucket::BuiltWheels,
|
|
WheelCache::Path(&path_source_dist.url)
|
|
.remote_wheel_dir(path_source_dist.name().as_ref()),
|
|
METADATA,
|
|
);
|
|
|
|
// Determine the last-modified time of the source distribution.
|
|
let Some(modified) = puffin_cache::archive_mtime(&path_source_dist.path)? else {
|
|
return Err(SourceDistError::DirWithoutEntrypoint);
|
|
};
|
|
|
|
// Read the existing metadata from the cache.
|
|
let mut manifest = Self::read_cached_metadata(&cache_entry, modified)
|
|
.await?
|
|
.unwrap_or_default();
|
|
|
|
// If the cache contains a compatible wheel, return it.
|
|
if let Some(built_wheel) =
|
|
BuiltWheelMetadata::find_in_cache(self.tags, &manifest, &cache_entry)
|
|
{
|
|
return Ok(built_wheel);
|
|
}
|
|
|
|
// Otherwise, we need to build a wheel.
|
|
let task = self
|
|
.reporter
|
|
.as_ref()
|
|
.map(|reporter| reporter.on_build_start(source_dist));
|
|
|
|
let (disk_filename, filename, metadata) = self
|
|
.build_source_dist(source_dist, &path_source_dist.path, None, &cache_entry)
|
|
.await?;
|
|
|
|
// Store the metadata for this build along with all the other builds.
|
|
manifest.insert_wheel(
|
|
filename.clone(),
|
|
DiskFilenameAndMetadata {
|
|
disk_filename: disk_filename.clone(),
|
|
metadata: metadata.clone(),
|
|
},
|
|
);
|
|
let cached = CachedByTimestamp {
|
|
timestamp: modified,
|
|
data: manifest,
|
|
};
|
|
let data = rmp_serde::to_vec(&cached)?;
|
|
write_atomic(cache_entry.path(), data).await?;
|
|
|
|
if let Some(task) = task {
|
|
if let Some(reporter) = self.reporter.as_ref() {
|
|
reporter.on_build_complete(source_dist, task);
|
|
}
|
|
}
|
|
|
|
let path = cache_entry.dir().join(&disk_filename);
|
|
let target = cache_entry.dir().join(filename.stem());
|
|
|
|
Ok(BuiltWheelMetadata {
|
|
path,
|
|
target,
|
|
filename,
|
|
})
|
|
}
|
|
|
|
/// Build the source distribution's metadata from a local path.
|
|
///
|
|
/// If the build backend supports `prepare_metadata_for_build_wheel`, this method will avoid
|
|
/// building the wheel.
|
|
async fn path_metadata(
|
|
&self,
|
|
source_dist: &SourceDist,
|
|
path_source_dist: &PathSourceDist,
|
|
) -> Result<Metadata21, SourceDistError> {
|
|
let cache_entry = self.build_context.cache().entry(
|
|
CacheBucket::BuiltWheels,
|
|
WheelCache::Path(&path_source_dist.url)
|
|
.remote_wheel_dir(path_source_dist.name().as_ref()),
|
|
METADATA,
|
|
);
|
|
|
|
// Determine the last-modified time of the source distribution.
|
|
let file_metadata = fs_err::metadata(&path_source_dist.path)?;
|
|
let modified = if file_metadata.is_file() {
|
|
// `modified()` is infallible on windows and unix (i.e., all platforms we support).
|
|
file_metadata.modified()?
|
|
} else {
|
|
if let Some(metadata) = path_source_dist
|
|
.path
|
|
.join("pyproject.toml")
|
|
.metadata()
|
|
.ok()
|
|
.filter(std::fs::Metadata::is_file)
|
|
{
|
|
metadata.modified()?
|
|
} else if let Some(metadata) = path_source_dist
|
|
.path
|
|
.join("setup.py")
|
|
.metadata()
|
|
.ok()
|
|
.filter(std::fs::Metadata::is_file)
|
|
{
|
|
metadata.modified()?
|
|
} else {
|
|
return Err(SourceDistError::DirWithoutEntrypoint);
|
|
}
|
|
};
|
|
|
|
// Read the existing metadata from the cache.
|
|
let mut manifest = Self::read_cached_metadata(&cache_entry, modified)
|
|
.await?
|
|
.unwrap_or_default();
|
|
|
|
// If the cache contains compatible metadata, return it.
|
|
if let Some(metadata) = manifest.find_metadata() {
|
|
return Ok(metadata.clone());
|
|
}
|
|
|
|
// If the backend supports `prepare_metadata_for_build_wheel`, use it.
|
|
if let Some(metadata) = self
|
|
.build_source_dist_metadata(source_dist, &path_source_dist.path, None)
|
|
.boxed()
|
|
.await?
|
|
{
|
|
// Store the metadata for this build along with all the other builds.
|
|
manifest.set_metadata(metadata.clone());
|
|
let cached = CachedByTimestamp {
|
|
timestamp: modified,
|
|
data: manifest,
|
|
};
|
|
let data = rmp_serde::to_vec(&cached)?;
|
|
|
|
fs::create_dir_all(&cache_entry.dir()).await?;
|
|
write_atomic(cache_entry.path(), data).await?;
|
|
|
|
return Ok(metadata);
|
|
}
|
|
|
|
// Otherwise, we need to build a wheel.
|
|
let task = self
|
|
.reporter
|
|
.as_ref()
|
|
.map(|reporter| reporter.on_build_start(source_dist));
|
|
|
|
let (disk_filename, filename, metadata) = self
|
|
.build_source_dist(source_dist, &path_source_dist.path, None, &cache_entry)
|
|
.await?;
|
|
|
|
// Store the metadata for this build along with all the other builds.
|
|
manifest.insert_wheel(
|
|
filename.clone(),
|
|
DiskFilenameAndMetadata {
|
|
disk_filename: disk_filename.clone(),
|
|
metadata: metadata.clone(),
|
|
},
|
|
);
|
|
let cached = CachedByTimestamp {
|
|
timestamp: modified,
|
|
data: manifest,
|
|
};
|
|
let data = rmp_serde::to_vec(&cached)?;
|
|
write_atomic(cache_entry.path(), data).await?;
|
|
|
|
if let Some(task) = task {
|
|
if let Some(reporter) = self.reporter.as_ref() {
|
|
reporter.on_build_complete(source_dist, task);
|
|
}
|
|
}
|
|
|
|
Ok(metadata)
|
|
}
|
|
|
|
/// Build a source distribution from a Git repository.
|
|
async fn git(
|
|
&self,
|
|
source_dist: &SourceDist,
|
|
git_source_dist: &GitSourceDist,
|
|
) -> Result<BuiltWheelMetadata, SourceDistError> {
|
|
let (fetch, subdirectory) = self.download_source_dist_git(&git_source_dist.url).await?;
|
|
|
|
// TODO(konstin): Do we want to delete old built wheels when the git sha changed?
|
|
let git_sha = fetch.git().precise().expect("Exact commit after checkout");
|
|
let cache_entry = self.build_context.cache().entry(
|
|
CacheBucket::BuiltWheels,
|
|
WheelCache::Git(&git_source_dist.url, &git_sha.to_short_string())
|
|
.remote_wheel_dir(git_source_dist.name().as_ref()),
|
|
METADATA,
|
|
);
|
|
|
|
// Read the existing metadata from the cache.
|
|
let mut manifest = Self::read_metadata(&cache_entry).await?.unwrap_or_default();
|
|
|
|
// If the cache contains a compatible wheel, return it.
|
|
if let Some(built_wheel) =
|
|
BuiltWheelMetadata::find_in_cache(self.tags, &manifest, &cache_entry)
|
|
{
|
|
return Ok(built_wheel);
|
|
}
|
|
|
|
let task = self
|
|
.reporter
|
|
.as_ref()
|
|
.map(|reporter| reporter.on_build_start(source_dist));
|
|
|
|
let (disk_filename, filename, metadata) = self
|
|
.build_source_dist(
|
|
source_dist,
|
|
fetch.path(),
|
|
subdirectory.as_deref(),
|
|
&cache_entry,
|
|
)
|
|
.await?;
|
|
|
|
// Store the metadata for this build along with all the other builds.
|
|
manifest.insert_wheel(
|
|
filename.clone(),
|
|
DiskFilenameAndMetadata {
|
|
disk_filename: disk_filename.clone(),
|
|
metadata: metadata.clone(),
|
|
},
|
|
);
|
|
let data = rmp_serde::to_vec(&manifest)?;
|
|
write_atomic(cache_entry.path(), data).await?;
|
|
|
|
if let Some(task) = task {
|
|
if let Some(reporter) = self.reporter.as_ref() {
|
|
reporter.on_build_complete(source_dist, task);
|
|
}
|
|
}
|
|
|
|
let path = cache_entry.dir().join(&disk_filename);
|
|
let target = cache_entry.dir().join(filename.stem());
|
|
|
|
Ok(BuiltWheelMetadata {
|
|
path,
|
|
target,
|
|
filename,
|
|
})
|
|
}
|
|
|
|
/// Build the source distribution's metadata from a Git repository.
|
|
///
|
|
/// If the build backend supports `prepare_metadata_for_build_wheel`, this method will avoid
|
|
/// building the wheel.
|
|
async fn git_metadata(
|
|
&self,
|
|
source_dist: &SourceDist,
|
|
git_source_dist: &GitSourceDist,
|
|
) -> Result<Metadata21, SourceDistError> {
|
|
let (fetch, subdirectory) = self.download_source_dist_git(&git_source_dist.url).await?;
|
|
|
|
let git_sha = fetch.git().precise().expect("Exact commit after checkout");
|
|
let cache_entry = self.build_context.cache().entry(
|
|
CacheBucket::BuiltWheels,
|
|
WheelCache::Git(&git_source_dist.url, &git_sha.to_short_string())
|
|
.remote_wheel_dir(git_source_dist.name().as_ref()),
|
|
METADATA,
|
|
);
|
|
|
|
// Read the existing metadata from the cache.
|
|
let mut manifest = Self::read_metadata(&cache_entry).await?.unwrap_or_default();
|
|
|
|
// If the cache contains compatible metadata, return it.
|
|
if let Some(metadata) = manifest.find_metadata() {
|
|
return Ok(metadata.clone());
|
|
}
|
|
|
|
// If the backend supports `prepare_metadata_for_build_wheel`, use it.
|
|
if let Some(metadata) = self
|
|
.build_source_dist_metadata(source_dist, fetch.path(), subdirectory.as_deref())
|
|
.boxed()
|
|
.await?
|
|
{
|
|
// Store the metadata for this build along with all the other builds.
|
|
manifest.set_metadata(metadata.clone());
|
|
let data = rmp_serde::to_vec(&manifest)?;
|
|
|
|
fs::create_dir_all(&cache_entry.dir()).await?;
|
|
write_atomic(cache_entry.path(), data).await?;
|
|
|
|
return Ok(metadata);
|
|
}
|
|
|
|
// Otherwise, we need to build a wheel.
|
|
let task = self
|
|
.reporter
|
|
.as_ref()
|
|
.map(|reporter| reporter.on_build_start(source_dist));
|
|
|
|
let (disk_filename, filename, metadata) = self
|
|
.build_source_dist(
|
|
source_dist,
|
|
fetch.path(),
|
|
subdirectory.as_deref(),
|
|
&cache_entry,
|
|
)
|
|
.await?;
|
|
|
|
// Store the metadata for this build along with all the other builds.
|
|
manifest.insert_wheel(
|
|
filename.clone(),
|
|
DiskFilenameAndMetadata {
|
|
disk_filename: disk_filename.clone(),
|
|
metadata: metadata.clone(),
|
|
},
|
|
);
|
|
let data = rmp_serde::to_vec(&manifest)?;
|
|
write_atomic(cache_entry.path(), data).await?;
|
|
|
|
if let Some(task) = task {
|
|
if let Some(reporter) = self.reporter.as_ref() {
|
|
reporter.on_build_complete(source_dist, task);
|
|
}
|
|
}
|
|
|
|
Ok(metadata)
|
|
}
|
|
|
|
/// Download and unzip a source distribution into the cache from an HTTP response.
|
|
async fn persist_source_dist_url<'data>(
|
|
&self,
|
|
response: Response,
|
|
source_dist: &SourceDist,
|
|
filename: &str,
|
|
cache_entry: &'data CacheEntry,
|
|
) -> Result<&'data Path, SourceDistError> {
|
|
let cache_path = cache_entry.path();
|
|
if cache_path.is_dir() {
|
|
debug!("Distribution is already cached: {source_dist}");
|
|
return Ok(cache_path);
|
|
}
|
|
|
|
// Download the source distribution to a temporary file.
|
|
let span =
|
|
info_span!("download_source_dist", filename = filename, source_dist = %source_dist);
|
|
let (temp_dir, source_dist_archive) =
|
|
self.download_source_dist_url(response, filename).await?;
|
|
drop(span);
|
|
|
|
// Unzip the source distribution to a temporary directory.
|
|
let span =
|
|
info_span!("extract_source_dist", filename = filename, source_dist = %source_dist);
|
|
let source_dist_dir = puffin_extract::extract_source(
|
|
&source_dist_archive,
|
|
temp_dir.path().join("extracted"),
|
|
)?;
|
|
drop(span);
|
|
|
|
// Persist the unzipped distribution to the cache.
|
|
fs::create_dir_all(&cache_entry.dir()).await?;
|
|
if let Err(err) = fs_err::rename(&source_dist_dir, cache_path) {
|
|
// If another thread already cached the distribution, we can ignore the error.
|
|
if cache_path.is_dir() {
|
|
warn!("Downloaded already-cached distribution: {source_dist}");
|
|
} else {
|
|
return Err(err.into());
|
|
};
|
|
}
|
|
|
|
Ok(cache_path)
|
|
}
|
|
|
|
/// Download a source distribution from a URL to a temporary file.
|
|
async fn download_source_dist_url(
|
|
&self,
|
|
response: Response,
|
|
source_dist_filename: &str,
|
|
) -> Result<(TempDir, PathBuf), puffin_client::Error> {
|
|
let reader = response
|
|
.bytes_stream()
|
|
.map_err(|err| std::io::Error::new(std::io::ErrorKind::Other, err))
|
|
.into_async_read();
|
|
let mut reader = tokio::io::BufReader::new(reader.compat());
|
|
|
|
// Create a temporary directory.
|
|
let cache_dir = self.build_context.cache().bucket(CacheBucket::BuiltWheels);
|
|
fs::create_dir_all(&cache_dir)
|
|
.await
|
|
.map_err(puffin_client::Error::CacheWrite)?;
|
|
let temp_dir = tempfile::tempdir_in(cache_dir).map_err(puffin_client::Error::CacheWrite)?;
|
|
|
|
// Download the source distribution to a temporary file.
|
|
let sdist_file = temp_dir.path().join(source_dist_filename);
|
|
let mut writer = tokio::io::BufWriter::new(
|
|
fs_err::tokio::File::create(&sdist_file)
|
|
.await
|
|
.map_err(puffin_client::Error::CacheWrite)?,
|
|
);
|
|
tokio::io::copy(&mut reader, &mut writer)
|
|
.await
|
|
.map_err(puffin_client::Error::CacheWrite)?;
|
|
|
|
Ok((temp_dir, sdist_file))
|
|
}
|
|
|
|
/// Download a source distribution from a Git repository.
|
|
async fn download_source_dist_git(
|
|
&self,
|
|
url: &Url,
|
|
) -> Result<(Fetch, Option<PathBuf>), SourceDistError> {
|
|
debug!("Fetching source distribution from Git: {url}");
|
|
let git_dir = self.build_context.cache().bucket(CacheBucket::Git);
|
|
|
|
// Avoid races between different processes, too.
|
|
let lock_dir = git_dir.join("locks");
|
|
fs::create_dir_all(&lock_dir).await?;
|
|
let canonical_url = cache_key::CanonicalUrl::new(url);
|
|
let _lock = LockedFile::acquire(
|
|
lock_dir.join(cache_key::digest(&canonical_url)),
|
|
&canonical_url,
|
|
)?;
|
|
|
|
let DirectGitUrl { url, subdirectory } =
|
|
DirectGitUrl::try_from(url).map_err(SourceDistError::Git)?;
|
|
|
|
let source = if let Some(reporter) = &self.reporter {
|
|
GitSource::new(url, git_dir).with_reporter(Facade::from(reporter.clone()))
|
|
} else {
|
|
GitSource::new(url, git_dir)
|
|
};
|
|
let fetch = tokio::task::spawn_blocking(move || source.fetch())
|
|
.await?
|
|
.map_err(SourceDistError::Git)?;
|
|
Ok((fetch, subdirectory))
|
|
}
|
|
|
|
/// Build a source distribution, storing the built wheel in the cache.
|
|
///
|
|
/// Returns the un-normalized disk filename, the parsed, normalized filename and the metadata
|
|
#[instrument(skip_all, fields(dist))]
|
|
async fn build_source_dist(
|
|
&self,
|
|
dist: &SourceDist,
|
|
source_dist: &Path,
|
|
subdirectory: Option<&Path>,
|
|
cache_entry: &CacheEntry,
|
|
) -> Result<(String, WheelFilename, Metadata21), SourceDistError> {
|
|
debug!("Building: {dist}");
|
|
|
|
if self.build_context.no_build() {
|
|
return Err(SourceDistError::NoBuild);
|
|
}
|
|
|
|
// Build the wheel.
|
|
fs::create_dir_all(&cache_entry.dir()).await?;
|
|
let disk_filename = self
|
|
.build_context
|
|
.setup_build(
|
|
source_dist,
|
|
subdirectory,
|
|
&dist.to_string(),
|
|
BuildKind::Wheel,
|
|
)
|
|
.await
|
|
.map_err(|err| SourceDistError::Build(dist.to_string(), err))?
|
|
.wheel(cache_entry.dir())
|
|
.await
|
|
.map_err(|err| SourceDistError::Build(dist.to_string(), err))?;
|
|
|
|
// Read the metadata from the wheel.
|
|
let filename = WheelFilename::from_str(&disk_filename)?;
|
|
let metadata = read_metadata(&filename, cache_entry.dir().join(&disk_filename))?;
|
|
|
|
// Validate the metadata.
|
|
if &metadata.name != dist.name() {
|
|
return Err(SourceDistError::NameMismatch {
|
|
metadata: metadata.name,
|
|
given: dist.name().clone(),
|
|
});
|
|
}
|
|
|
|
debug!("Finished building: {dist}");
|
|
Ok((disk_filename, filename, metadata))
|
|
}
|
|
|
|
/// Build the metadata for a source distribution.
|
|
#[instrument(skip_all, fields(dist))]
|
|
async fn build_source_dist_metadata(
|
|
&self,
|
|
dist: &SourceDist,
|
|
source_dist: &Path,
|
|
subdirectory: Option<&Path>,
|
|
) -> Result<Option<Metadata21>, SourceDistError> {
|
|
debug!("Preparing metadata for: {dist}");
|
|
|
|
// Setup the builder.
|
|
let mut builder = self
|
|
.build_context
|
|
.setup_build(
|
|
source_dist,
|
|
subdirectory,
|
|
&dist.to_string(),
|
|
BuildKind::Wheel,
|
|
)
|
|
.await
|
|
.map_err(|err| SourceDistError::Build(dist.to_string(), err))?;
|
|
|
|
// Build the metadata.
|
|
let dist_info = builder
|
|
.metadata()
|
|
.await
|
|
.map_err(|err| SourceDistError::Build(dist.to_string(), err))?;
|
|
let Some(dist_info) = dist_info else {
|
|
return Ok(None);
|
|
};
|
|
|
|
// Read the metadata from disk.
|
|
debug!("Prepared metadata for: {dist}");
|
|
let metadata = Metadata21::parse(&fs::read(dist_info.join("METADATA")).await?)?;
|
|
|
|
// Validate the metadata.
|
|
if &metadata.name != dist.name() {
|
|
return Err(SourceDistError::NameMismatch {
|
|
metadata: metadata.name,
|
|
given: dist.name().clone(),
|
|
});
|
|
}
|
|
|
|
Ok(Some(metadata))
|
|
}
|
|
|
|
/// Build a single directory into an editable wheel
|
|
pub async fn build_editable(
|
|
&self,
|
|
editable: &LocalEditable,
|
|
editable_wheel_dir: &Path,
|
|
) -> Result<(Dist, String, WheelFilename, Metadata21), SourceDistError> {
|
|
debug!("Building (editable) {editable}");
|
|
let disk_filename = self
|
|
.build_context
|
|
.setup_build(
|
|
&editable.path,
|
|
None,
|
|
&editable.to_string(),
|
|
BuildKind::Editable,
|
|
)
|
|
.await
|
|
.map_err(|err| SourceDistError::Build(editable.to_string(), err))?
|
|
.wheel(editable_wheel_dir)
|
|
.await
|
|
.map_err(|err| SourceDistError::Build(editable.to_string(), err))?;
|
|
let filename = WheelFilename::from_str(&disk_filename)?;
|
|
// We finally have the name of the package and can construct the dist
|
|
let dist = Dist::Source(SourceDist::Path(PathSourceDist {
|
|
name: filename.name.clone(),
|
|
url: editable.url().clone(),
|
|
path: editable.path.clone(),
|
|
editable: true,
|
|
}));
|
|
let metadata = read_metadata(&filename, editable_wheel_dir.join(&disk_filename))?;
|
|
|
|
debug!("Finished building (editable): {dist}");
|
|
Ok((dist, disk_filename, filename, metadata))
|
|
}
|
|
|
|
/// Read an existing cache entry, if it exists and is up-to-date.
|
|
async fn read_cached_metadata(
|
|
cache_entry: &CacheEntry,
|
|
modified: std::time::SystemTime,
|
|
) -> Result<Option<Manifest>, SourceDistError> {
|
|
match fs::read(&cache_entry.path()).await {
|
|
Ok(cached) => {
|
|
let cached = rmp_serde::from_slice::<CachedByTimestamp<Manifest>>(&cached)?;
|
|
if cached.timestamp == modified {
|
|
Ok(Some(cached.data))
|
|
} else {
|
|
debug!(
|
|
"Removing stale built wheels for: {}",
|
|
cache_entry.path().display()
|
|
);
|
|
if let Err(err) = fs::remove_dir_all(&cache_entry.dir()).await {
|
|
warn!("Failed to remove stale built wheel cache directory: {err}");
|
|
}
|
|
Ok(None)
|
|
}
|
|
}
|
|
Err(err) if err.kind() == std::io::ErrorKind::NotFound => Ok(None),
|
|
Err(err) => Err(err.into()),
|
|
}
|
|
}
|
|
|
|
/// Read an existing cache entry, if it exists.
|
|
async fn read_metadata(cache_entry: &CacheEntry) -> Result<Option<Manifest>, SourceDistError> {
|
|
match fs::read(&cache_entry.path()).await {
|
|
Ok(cached) => Ok(Some(rmp_serde::from_slice::<Manifest>(&cached)?)),
|
|
Err(err) if err.kind() == std::io::ErrorKind::NotFound => Ok(None),
|
|
Err(err) => Err(err.into()),
|
|
}
|
|
}
|
|
}
|
|
|
|
/// Read the [`Metadata21`] from a built wheel.
|
|
fn read_metadata(
|
|
filename: &WheelFilename,
|
|
wheel: impl Into<PathBuf>,
|
|
) -> Result<Metadata21, SourceDistError> {
|
|
let mut archive = ZipArchive::new(fs_err::File::open(wheel)?)?;
|
|
let dist_info = read_dist_info(filename, &mut archive)?;
|
|
Ok(Metadata21::parse(&dist_info)?)
|
|
}
|