Move server code into its own module

This commit is contained in:
Joscha 2023-08-07 14:23:47 +02:00
parent ad0c1a69cb
commit 45abda2b6d
12 changed files with 15 additions and 4 deletions

View file

@ -0,0 +1,280 @@
//! Add new commits to the database and update the tracked refs.
use std::{collections::HashSet, sync::Arc};
use futures::TryStreamExt;
use gix::{
objs::Kind, prelude::ObjectIdExt, refs::Reference, ObjectId, Repository, ThreadSafeRepository,
};
use sqlx::{Acquire, SqliteConnection, SqlitePool};
use tracing::{debug, info};
use crate::{somehow, util};
async fn get_all_commit_hashes_from_db(
conn: &mut SqliteConnection,
) -> somehow::Result<HashSet<ObjectId>> {
let hashes = sqlx::query!("SELECT hash FROM commits")
.fetch(conn)
.err_into::<somehow::Error>()
.and_then(|r| async move { r.hash.parse::<ObjectId>().map_err(|e| e.into()) })
.try_collect::<HashSet<_>>()
.await?;
Ok(hashes)
}
fn get_all_refs_from_repo(repo: &Repository) -> somehow::Result<Vec<Reference>> {
let mut references = vec![];
for reference in repo.references()?.all()? {
let mut reference = reference.map_err(somehow::Error::from_box)?;
reference.peel_to_id_in_place()?;
// Some repos *cough*linuxkernel*cough* have refs that don't point to
// commits. This makes the rev walk choke and die. We don't want that.
if reference.id().object()?.kind != Kind::Commit {
continue;
}
references.push(reference.detach());
}
Ok(references)
}
fn get_new_commits_from_repo(
repo: &Repository,
refs: &[Reference],
old: &HashSet<ObjectId>,
) -> somehow::Result<Vec<ObjectId>> {
let ref_ids = refs.iter().flat_map(|r| r.peeled.into_iter());
// Walk from those until hitting old references
let mut new = vec![];
for commit in repo.rev_walk(ref_ids).selected(|c| !old.contains(c))? {
new.push(commit?.id);
}
Ok(new)
}
fn get_all_refs_and_new_commits_from_repo(
repo: &Repository,
old: &HashSet<ObjectId>,
) -> somehow::Result<(Vec<Reference>, Vec<ObjectId>)> {
let refs = get_all_refs_from_repo(repo)?;
let new = get_new_commits_from_repo(repo, &refs, old)?;
Ok((refs, new))
}
async fn insert_new_commits(
conn: &mut SqliteConnection,
repo: &Repository,
new: &[ObjectId],
) -> somehow::Result<()> {
for (i, id) in new.iter().enumerate() {
let commit = id.attach(repo).object()?.try_into_commit()?;
let hash = commit.id.to_string();
let author_info = commit.author()?;
let author = util::format_actor(author_info.actor())?;
let author_date = util::time_to_offset_datetime(author_info.time)?;
let committer_info = commit.committer()?;
let committer = util::format_actor(committer_info.actor())?;
let committer_date = util::time_to_offset_datetime(committer_info.time)?;
let message = commit.message_raw()?.to_string();
sqlx::query!(
"\
INSERT OR IGNORE INTO commits ( \
hash, \
author, \
author_date, \
committer, \
committer_date, \
message \
) \
VALUES (?, ?, ?, ?, ?, ?) \
",
hash,
author,
author_date,
committer,
committer_date,
message
)
.execute(&mut *conn)
.await?;
if (i + 1) % 100000 == 0 {
debug!("Inserted {} commits so far", i + 1);
}
}
debug!("Inserted {} commits in total", new.len());
Ok(())
}
async fn insert_new_commit_links(
conn: &mut SqliteConnection,
repo: &Repository,
new: &[ObjectId],
) -> somehow::Result<()> {
for (i, hash) in new.iter().enumerate() {
let commit = hash.attach(repo).object()?.try_into_commit()?;
let child = commit.id.to_string();
for parent in commit.parent_ids() {
let parent = parent.to_string();
// Commits *cough*linuxkernel*cough* may list the same parent
// multiple times, so we just ignore duplicates during insert.
sqlx::query!(
"INSERT OR IGNORE INTO commit_links (parent, child) VALUES (?, ?)",
parent,
child,
)
.execute(&mut *conn)
.await?;
}
if (i + 1) % 100000 == 0 {
debug!("Inserted {} commits' links so far", i + 1);
}
}
debug!("Inserted {} commits' links in total", new.len());
Ok(())
}
async fn mark_all_commits_as_old(conn: &mut SqliteConnection) -> somehow::Result<()> {
sqlx::query!("UPDATE commits SET new = 0")
.execute(conn)
.await?;
Ok(())
}
async fn update_refs(conn: &mut SqliteConnection, refs: Vec<Reference>) -> somehow::Result<()> {
// Remove refs that no longer exist
let existing = refs
.iter()
.map(|r| r.name.to_string())
.collect::<HashSet<_>>();
let current = sqlx::query!("SELECT name FROM refs")
.fetch_all(&mut *conn)
.await?;
for reference in current {
if !existing.contains(&reference.name) {
sqlx::query!("DELETE FROM refs WHERE name = ?", reference.name)
.execute(&mut *conn)
.await?;
}
}
// Add new refs and update existing refs
for reference in refs {
let name = reference.name.to_string();
let Some(hash) = reference.peeled else { continue; };
let hash = hash.to_string();
sqlx::query!(
"\
INSERT INTO refs (name, hash) VALUES (?, ?) \
ON CONFLICT (name) DO UPDATE \
SET hash = excluded.hash \
",
name,
hash
)
.execute(&mut *conn)
.await?;
}
Ok(())
}
async fn track_main_branch(conn: &mut SqliteConnection, repo: &Repository) -> somehow::Result<()> {
let Some(head) = repo.head_ref()? else { return Ok(()); };
let name = head.inner.name.to_string();
sqlx::query!("UPDATE refs SET tracked = true WHERE name = ?", name)
.execute(conn)
.await?;
Ok(())
}
async fn update_commit_tracked_status(conn: &mut SqliteConnection) -> somehow::Result<()> {
sqlx::query!(
"\
WITH RECURSIVE \
tracked (hash) AS ( \
SELECT hash FROM refs WHERE tracked \
UNION \
SELECT parent FROM commit_links \
JOIN tracked ON hash = child \
), \
reachable (hash) AS ( \
SELECT hash FROM refs \
UNION \
SELECT hash FROM tracked \
UNION \
SELECT parent FROM commit_links \
JOIN reachable ON hash = child \
) \
UPDATE commits \
SET reachable = CASE \
WHEN hash IN tracked THEN 2 \
WHEN hash IN reachable THEN 1 \
ELSE 0 \
END \
"
)
.execute(conn)
.await?;
Ok(())
}
pub async fn update(db: &SqlitePool, repo: Arc<ThreadSafeRepository>) -> somehow::Result<()> {
debug!("Updating repo");
let thread_local_repo = repo.to_thread_local();
let mut tx = db.begin().await?;
let conn = tx.acquire().await?;
let old = get_all_commit_hashes_from_db(&mut *conn).await?;
debug!("Loaded {} commits from the db", old.len());
let repo_is_new = old.is_empty();
if repo_is_new {
info!("Initializing new repo");
}
// This can take a while for larger repos. Running it via spawn_blocking
// keeps it from blocking the entire tokio worker.
let (refs, new) = tokio::task::spawn_blocking(move || {
get_all_refs_and_new_commits_from_repo(&repo.to_thread_local(), &old)
})
.await??;
debug!("Found {} new commits in repo", new.len());
// Defer foreign key checks until the end of the transaction to improve
// insert performance.
sqlx::query!("PRAGMA defer_foreign_keys=1")
.execute(&mut *conn)
.await?;
// Inserts are grouped by table so sqlite can process them *a lot* faster
// than if they were grouped by commit (insert commit and parents, then next
// commit and so on).
insert_new_commits(conn, &thread_local_repo, &new).await?;
insert_new_commit_links(conn, &thread_local_repo, &new).await?;
if repo_is_new {
mark_all_commits_as_old(conn).await?;
}
update_refs(conn, refs).await?;
if repo_is_new {
track_main_branch(conn, &thread_local_repo).await?;
}
update_commit_tracked_status(conn).await?;
debug!("Updated tracked refs");
tx.commit().await?;
if repo_is_new {
info!("Initialized new repo");
}
debug!("Updated repo");
Ok(())
}