split meilisearch-http and meilisearch-lib

This commit is contained in:
mpostma 2021-09-21 13:23:22 +02:00
parent 09d4e37044
commit 60518449fc
63 changed files with 608 additions and 324 deletions

View file

@ -0,0 +1,157 @@
use std::collections::HashMap;
use std::path::{Path, PathBuf};
use std::sync::Arc;
use async_stream::stream;
use chrono::Utc;
use futures::{lock::Mutex, stream::StreamExt};
use log::{error, trace};
use tokio::sync::{mpsc, oneshot, RwLock};
use update_actor::UpdateActorHandle;
use uuid_resolver::UuidResolverHandle;
use super::error::{DumpActorError, Result};
use super::{DumpInfo, DumpMsg, DumpStatus, DumpTask};
use crate::index_controller::{update_actor, uuid_resolver};
pub const CONCURRENT_DUMP_MSG: usize = 10;
pub struct DumpActor<UuidResolver, Update> {
inbox: Option<mpsc::Receiver<DumpMsg>>,
uuid_resolver: UuidResolver,
update: Update,
dump_path: PathBuf,
lock: Arc<Mutex<()>>,
dump_infos: Arc<RwLock<HashMap<String, DumpInfo>>>,
update_db_size: usize,
index_db_size: usize,
}
/// Generate uid from creation date
fn generate_uid() -> String {
Utc::now().format("%Y%m%d-%H%M%S%3f").to_string()
}
impl<UuidResolver, Update> DumpActor<UuidResolver, Update>
where
UuidResolver: UuidResolverHandle + Send + Sync + Clone + 'static,
Update: UpdateActorHandle + Send + Sync + Clone + 'static,
{
pub fn new(
inbox: mpsc::Receiver<DumpMsg>,
uuid_resolver: UuidResolver,
update: Update,
dump_path: impl AsRef<Path>,
index_db_size: usize,
update_db_size: usize,
) -> Self {
let dump_infos = Arc::new(RwLock::new(HashMap::new()));
let lock = Arc::new(Mutex::new(()));
Self {
inbox: Some(inbox),
uuid_resolver,
update,
dump_path: dump_path.as_ref().into(),
dump_infos,
lock,
index_db_size,
update_db_size,
}
}
pub async fn run(mut self) {
trace!("Started dump actor.");
let mut inbox = self
.inbox
.take()
.expect("Dump Actor must have a inbox at this point.");
let stream = stream! {
loop {
match inbox.recv().await {
Some(msg) => yield msg,
None => break,
}
}
};
stream
.for_each_concurrent(Some(CONCURRENT_DUMP_MSG), |msg| self.handle_message(msg))
.await;
error!("Dump actor stopped.");
}
async fn handle_message(&self, msg: DumpMsg) {
use DumpMsg::*;
match msg {
CreateDump { ret } => {
let _ = self.handle_create_dump(ret).await;
}
DumpInfo { ret, uid } => {
let _ = ret.send(self.handle_dump_info(uid).await);
}
}
}
async fn handle_create_dump(&self, ret: oneshot::Sender<Result<DumpInfo>>) {
let uid = generate_uid();
let info = DumpInfo::new(uid.clone(), DumpStatus::InProgress);
let _lock = match self.lock.try_lock() {
Some(lock) => lock,
None => {
ret.send(Err(DumpActorError::DumpAlreadyRunning))
.expect("Dump actor is dead");
return;
}
};
self.dump_infos
.write()
.await
.insert(uid.clone(), info.clone());
ret.send(Ok(info)).expect("Dump actor is dead");
let task = DumpTask {
path: self.dump_path.clone(),
uuid_resolver: self.uuid_resolver.clone(),
update_handle: self.update.clone(),
uid: uid.clone(),
update_db_size: self.update_db_size,
index_db_size: self.index_db_size,
};
let task_result = tokio::task::spawn(task.run()).await;
let mut dump_infos = self.dump_infos.write().await;
let dump_infos = dump_infos
.get_mut(&uid)
.expect("dump entry deleted while lock was acquired");
match task_result {
Ok(Ok(())) => {
dump_infos.done();
trace!("Dump succeed");
}
Ok(Err(e)) => {
dump_infos.with_error(e.to_string());
error!("Dump failed: {}", e);
}
Err(_) => {
dump_infos.with_error("Unexpected error while performing dump.".to_string());
error!("Dump panicked. Dump status set to failed");
}
};
}
async fn handle_dump_info(&self, uid: String) -> Result<DumpInfo> {
match self.dump_infos.read().await.get(&uid) {
Some(info) => Ok(info.clone()),
_ => Err(DumpActorError::DumpDoesNotExist(uid)),
}
}
}

View file

@ -0,0 +1,52 @@
use meilisearch_error::{Code, ErrorCode};
use crate::index_controller::update_actor::error::UpdateActorError;
use crate::index_controller::uuid_resolver::error::UuidResolverError;
pub type Result<T> = std::result::Result<T, DumpActorError>;
#[derive(thiserror::Error, Debug)]
pub enum DumpActorError {
#[error("Another dump is already in progress")]
DumpAlreadyRunning,
#[error("Dump `{0}` not found")]
DumpDoesNotExist(String),
#[error("Internal error: {0}")]
Internal(Box<dyn std::error::Error + Send + Sync + 'static>),
#[error("{0}")]
UuidResolver(#[from] UuidResolverError),
#[error("{0}")]
UpdateActor(#[from] UpdateActorError),
}
macro_rules! internal_error {
($($other:path), *) => {
$(
impl From<$other> for DumpActorError {
fn from(other: $other) -> Self {
Self::Internal(Box::new(other))
}
}
)*
}
}
internal_error!(
heed::Error,
std::io::Error,
tokio::task::JoinError,
serde_json::error::Error,
tempfile::PersistError
);
impl ErrorCode for DumpActorError {
fn error_code(&self) -> Code {
match self {
DumpActorError::DumpAlreadyRunning => Code::DumpAlreadyInProgress,
DumpActorError::DumpDoesNotExist(_) => Code::NotFound,
DumpActorError::Internal(_) => Code::Internal,
DumpActorError::UuidResolver(e) => e.error_code(),
DumpActorError::UpdateActor(e) => e.error_code(),
}
}
}

View file

@ -0,0 +1,52 @@
use std::path::Path;
use tokio::sync::{mpsc, oneshot};
use super::error::Result;
use super::{DumpActor, DumpActorHandle, DumpInfo, DumpMsg};
#[derive(Clone)]
pub struct DumpActorHandleImpl {
sender: mpsc::Sender<DumpMsg>,
}
#[async_trait::async_trait]
impl DumpActorHandle for DumpActorHandleImpl {
async fn create_dump(&self) -> Result<DumpInfo> {
let (ret, receiver) = oneshot::channel();
let msg = DumpMsg::CreateDump { ret };
let _ = self.sender.send(msg).await;
receiver.await.expect("IndexActor has been killed")
}
async fn dump_info(&self, uid: String) -> Result<DumpInfo> {
let (ret, receiver) = oneshot::channel();
let msg = DumpMsg::DumpInfo { ret, uid };
let _ = self.sender.send(msg).await;
receiver.await.expect("IndexActor has been killed")
}
}
impl DumpActorHandleImpl {
pub fn new(
path: impl AsRef<Path>,
uuid_resolver: crate::index_controller::uuid_resolver::UuidResolverHandleImpl,
update: crate::index_controller::update_actor::UpdateActorHandleImpl,
index_db_size: usize,
update_db_size: usize,
) -> anyhow::Result<Self> {
let (sender, receiver) = mpsc::channel(10);
let actor = DumpActor::new(
receiver,
uuid_resolver,
update,
path,
index_db_size,
update_db_size,
);
tokio::task::spawn(actor.run());
Ok(Self { sender })
}
}

View file

@ -0,0 +1,2 @@
pub mod v1;
pub mod v2;

View file

@ -0,0 +1,224 @@
use std::collections::{BTreeMap, BTreeSet};
use std::marker::PhantomData;
use std::path::Path;
use log::{error, info, warn};
use milli::update::Setting;
use serde::{Deserialize, Deserializer, Serialize};
use uuid::Uuid;
use crate::index_controller::{self, uuid_resolver::HeedUuidStore, IndexMetadata};
use crate::index_controller::{asc_ranking_rule, desc_ranking_rule};
use crate::{
index::Unchecked,
options::IndexerOpts,
};
#[derive(Serialize, Deserialize, Debug)]
#[serde(rename_all = "camelCase")]
pub struct MetadataV1 {
db_version: String,
indexes: Vec<IndexMetadata>,
}
impl MetadataV1 {
pub fn load_dump(
self,
src: impl AsRef<Path>,
dst: impl AsRef<Path>,
size: usize,
indexer_options: &IndexerOpts,
) -> anyhow::Result<()> {
info!(
"Loading dump, dump database version: {}, dump version: V1",
self.db_version
);
let uuid_store = HeedUuidStore::new(&dst)?;
for index in self.indexes {
let uuid = Uuid::new_v4();
uuid_store.insert(index.uid.clone(), uuid)?;
let src = src.as_ref().join(index.uid);
load_index(
&src,
&dst,
uuid,
index.meta.primary_key.as_deref(),
size,
indexer_options,
)?;
}
Ok(())
}
}
pub fn deserialize_some<'de, T, D>(deserializer: D) -> std::result::Result<Option<T>, D::Error>
where
T: Deserialize<'de>,
D: Deserializer<'de>,
{
Deserialize::deserialize(deserializer).map(Some)
}
// These are the settings used in legacy meilisearch (<v0.21.0).
#[derive(Default, Clone, Serialize, Deserialize, Debug)]
#[serde(rename_all = "camelCase", deny_unknown_fields)]
struct Settings {
#[serde(default, deserialize_with = "deserialize_some")]
pub ranking_rules: Option<Option<Vec<String>>>,
#[serde(default, deserialize_with = "deserialize_some")]
pub distinct_attribute: Option<Option<String>>,
#[serde(default, deserialize_with = "deserialize_some")]
pub searchable_attributes: Option<Option<Vec<String>>>,
#[serde(default, deserialize_with = "deserialize_some")]
pub displayed_attributes: Option<Option<BTreeSet<String>>>,
#[serde(default, deserialize_with = "deserialize_some")]
pub stop_words: Option<Option<BTreeSet<String>>>,
#[serde(default, deserialize_with = "deserialize_some")]
pub synonyms: Option<Option<BTreeMap<String, Vec<String>>>>,
#[serde(default, deserialize_with = "deserialize_some")]
pub attributes_for_faceting: Option<Option<Vec<String>>>,
}
fn load_index(
_src: impl AsRef<Path>,
_dst: impl AsRef<Path>,
_uuid: Uuid,
_primary_key: Option<&str>,
_size: usize,
_indexer_options: &IndexerOpts,
) -> anyhow::Result<()> {
todo!("fix dump obkv documents")
//let index_path = dst.as_ref().join(&format!("indexes/index-{}", uuid));
//create_dir_all(&index_path)?;
//let mut options = EnvOpenOptions::new();
//options.map_size(size);
//let index = milli::Index::new(options, index_path)?;
//let index = Index(Arc::new(index));
//// extract `settings.json` file and import content
//let settings = import_settings(&src)?;
//let settings: index_controller::Settings<Unchecked> = settings.into();
//let mut txn = index.write_txn()?;
//let handler = UpdateHandler::new(indexer_options)?;
//index.update_settings_txn(&mut txn, &settings.check(), handler.update_builder(0))?;
//let file = File::open(&src.as_ref().join("documents.jsonl"))?;
//let mut reader = std::io::BufReader::new(file);
//reader.fill_buf()?;
//if !reader.buffer().is_empty() {
//index.update_documents_txn(
//&mut txn,
//IndexDocumentsMethod::ReplaceDocuments,
//Some(reader),
//handler.update_builder(0),
//primary_key,
//)?;
//}
//txn.commit()?;
//// Finaly, we extract the original milli::Index and close it
//Arc::try_unwrap(index.0)
//.map_err(|_e| "Couldn't close the index properly")
//.unwrap()
//.prepare_for_closing()
//.wait();
//// Updates are ignored in dumps V1.
//Ok(())
}
/// we need to **always** be able to convert the old settings to the settings currently being used
impl From<Settings> for index_controller::Settings<Unchecked> {
fn from(settings: Settings) -> Self {
Self {
distinct_attribute: match settings.distinct_attribute {
Some(Some(attr)) => Setting::Set(attr),
Some(None) => Setting::Reset,
None => Setting::NotSet
},
// we need to convert the old `Vec<String>` into a `BTreeSet<String>`
displayed_attributes: match settings.displayed_attributes {
Some(Some(attrs)) => Setting::Set(attrs.into_iter().collect()),
Some(None) => Setting::Reset,
None => Setting::NotSet
},
searchable_attributes: match settings.searchable_attributes {
Some(Some(attrs)) => Setting::Set(attrs),
Some(None) => Setting::Reset,
None => Setting::NotSet
},
filterable_attributes: match settings.attributes_for_faceting {
Some(Some(attrs)) => Setting::Set(attrs.into_iter().collect()),
Some(None) => Setting::Reset,
None => Setting::NotSet
},
sortable_attributes: Setting::NotSet,
ranking_rules: match settings.ranking_rules {
Some(Some(ranking_rules)) => Setting::Set(ranking_rules.into_iter().filter_map(|criterion| {
match criterion.as_str() {
"words" | "typo" | "proximity" | "attribute" | "exactness" => Some(criterion),
s if s.starts_with("asc") => asc_ranking_rule(s).map(|f| format!("{}:asc", f)),
s if s.starts_with("desc") => desc_ranking_rule(s).map(|f| format!("{}:desc", f)),
"wordsPosition" => {
warn!("The criteria `attribute` and `wordsPosition` have been merged \
into a single criterion `attribute` so `wordsPositon` will be \
ignored");
None
}
s => {
error!("Unknown criterion found in the dump: `{}`, it will be ignored", s);
None
}
}
}).collect()),
Some(None) => Setting::Reset,
None => Setting::NotSet
},
// we need to convert the old `Vec<String>` into a `BTreeSet<String>`
stop_words: match settings.stop_words {
Some(Some(stop_words)) => Setting::Set(stop_words.into_iter().collect()),
Some(None) => Setting::Reset,
None => Setting::NotSet
},
// we need to convert the old `Vec<String>` into a `BTreeMap<String>`
synonyms: match settings.synonyms {
Some(Some(synonyms)) => Setting::Set(synonyms.into_iter().collect()),
Some(None) => Setting::Reset,
None => Setting::NotSet
},
_kind: PhantomData,
}
}
}
// /// Extract Settings from `settings.json` file present at provided `dir_path`
//fn import_settings(dir_path: impl AsRef<Path>) -> anyhow::Result<Settings> {
//let path = dir_path.as_ref().join("settings.json");
//let file = File::open(path)?;
//let reader = std::io::BufReader::new(file);
//let metadata = serde_json::from_reader(reader)?;
//Ok(metadata)
//}
#[cfg(test)]
mod test {
use super::*;
#[test]
fn settings_format_regression() {
let settings = Settings::default();
assert_eq!(
r##"{"rankingRules":null,"distinctAttribute":null,"searchableAttributes":null,"displayedAttributes":null,"stopWords":null,"synonyms":null,"attributesForFaceting":null}"##,
serde_json::to_string(&settings).unwrap()
);
}
}

View file

@ -0,0 +1,59 @@
use std::path::Path;
use chrono::{DateTime, Utc};
use log::info;
use serde::{Deserialize, Serialize};
use crate::index::Index;
use crate::index_controller::{update_actor::UpdateStore, uuid_resolver::HeedUuidStore};
use crate::options::IndexerOpts;
#[derive(Serialize, Deserialize, Debug)]
#[serde(rename_all = "camelCase")]
pub struct MetadataV2 {
db_version: String,
index_db_size: usize,
update_db_size: usize,
dump_date: DateTime<Utc>,
}
impl MetadataV2 {
pub fn new(index_db_size: usize, update_db_size: usize) -> Self {
Self {
db_version: env!("CARGO_PKG_VERSION").to_string(),
index_db_size,
update_db_size,
dump_date: Utc::now(),
}
}
pub fn load_dump(
self,
src: impl AsRef<Path>,
dst: impl AsRef<Path>,
index_db_size: usize,
update_db_size: usize,
indexing_options: &IndexerOpts,
) -> anyhow::Result<()> {
info!(
"Loading dump from {}, dump database version: {}, dump version: V2",
self.dump_date, self.db_version
);
info!("Loading index database.");
HeedUuidStore::load_dump(src.as_ref(), &dst)?;
info!("Loading updates.");
UpdateStore::load_dump(&src, &dst, update_db_size)?;
info!("Loading indexes.");
let indexes_path = src.as_ref().join("indexes");
let indexes = indexes_path.read_dir()?;
for index in indexes {
let index = index?;
Index::load_dump(&index.path(), &dst, index_db_size, indexing_options)?;
}
Ok(())
}
}

View file

@ -0,0 +1,14 @@
use tokio::sync::oneshot;
use super::error::Result;
use super::DumpInfo;
pub enum DumpMsg {
CreateDump {
ret: oneshot::Sender<Result<DumpInfo>>,
},
DumpInfo {
uid: String,
ret: oneshot::Sender<Result<DumpInfo>>,
},
}

View file

@ -0,0 +1,204 @@
use std::fs::File;
use std::path::{Path, PathBuf};
use anyhow::Context;
use chrono::{DateTime, Utc};
use log::{info, trace, warn};
#[cfg(test)]
use mockall::automock;
use serde::{Deserialize, Serialize};
use tokio::fs::create_dir_all;
use loaders::v1::MetadataV1;
use loaders::v2::MetadataV2;
pub use actor::DumpActor;
pub use handle_impl::*;
pub use message::DumpMsg;
use super::{update_actor::UpdateActorHandle, uuid_resolver::UuidResolverHandle};
use crate::index_controller::dump_actor::error::DumpActorError;
use crate::options::IndexerOpts;
use error::Result;
mod actor;
pub mod error;
mod handle_impl;
mod loaders;
mod message;
const META_FILE_NAME: &str = "metadata.json";
#[async_trait::async_trait]
#[cfg_attr(test, automock)]
pub trait DumpActorHandle {
/// Start the creation of a dump
/// Implementation: [handle_impl::DumpActorHandleImpl::create_dump]
async fn create_dump(&self) -> Result<DumpInfo>;
/// Return the status of an already created dump
/// Implementation: [handle_impl::DumpActorHandleImpl::dump_info]
async fn dump_info(&self, uid: String) -> Result<DumpInfo>;
}
#[derive(Debug, Serialize, Deserialize)]
#[serde(tag = "dumpVersion")]
pub enum Metadata {
V1(MetadataV1),
V2(MetadataV2),
}
impl Metadata {
pub fn new_v2(index_db_size: usize, update_db_size: usize) -> Self {
let meta = MetadataV2::new(index_db_size, update_db_size);
Self::V2(meta)
}
}
#[derive(Debug, Serialize, Deserialize, PartialEq, Clone)]
#[serde(rename_all = "snake_case")]
pub enum DumpStatus {
Done,
InProgress,
Failed,
}
#[derive(Debug, Serialize, Clone)]
#[serde(rename_all = "camelCase")]
pub struct DumpInfo {
pub uid: String,
pub status: DumpStatus,
#[serde(skip_serializing_if = "Option::is_none")]
pub error: Option<String>,
started_at: DateTime<Utc>,
#[serde(skip_serializing_if = "Option::is_none")]
finished_at: Option<DateTime<Utc>>,
}
impl DumpInfo {
pub fn new(uid: String, status: DumpStatus) -> Self {
Self {
uid,
status,
error: None,
started_at: Utc::now(),
finished_at: None,
}
}
pub fn with_error(&mut self, error: String) {
self.status = DumpStatus::Failed;
self.finished_at = Some(Utc::now());
self.error = Some(error);
}
pub fn done(&mut self) {
self.finished_at = Some(Utc::now());
self.status = DumpStatus::Done;
}
pub fn dump_already_in_progress(&self) -> bool {
self.status == DumpStatus::InProgress
}
}
pub fn load_dump(
dst_path: impl AsRef<Path>,
src_path: impl AsRef<Path>,
index_db_size: usize,
update_db_size: usize,
indexer_opts: &IndexerOpts,
) -> anyhow::Result<()> {
let tmp_src = tempfile::tempdir_in(".")?;
let tmp_src_path = tmp_src.path();
crate::from_tar_gz(&src_path, tmp_src_path)?;
let meta_path = tmp_src_path.join(META_FILE_NAME);
let mut meta_file = File::open(&meta_path)?;
let meta: Metadata = serde_json::from_reader(&mut meta_file)?;
let dst_dir = dst_path
.as_ref()
.parent()
.with_context(|| format!("Invalid db path: {}", dst_path.as_ref().display()))?;
let tmp_dst = tempfile::tempdir_in(dst_dir)?;
match meta {
Metadata::V1(meta) => {
meta.load_dump(&tmp_src_path, tmp_dst.path(), index_db_size, indexer_opts)?
}
Metadata::V2(meta) => meta.load_dump(
&tmp_src_path,
tmp_dst.path(),
index_db_size,
update_db_size,
indexer_opts,
)?,
}
// Persist and atomically rename the db
let persisted_dump = tmp_dst.into_path();
if dst_path.as_ref().exists() {
warn!("Overwriting database at {}", dst_path.as_ref().display());
std::fs::remove_dir_all(&dst_path)?;
}
std::fs::rename(&persisted_dump, &dst_path)?;
Ok(())
}
struct DumpTask<U, P> {
path: PathBuf,
uuid_resolver: U,
update_handle: P,
uid: String,
update_db_size: usize,
index_db_size: usize,
}
impl<U, P> DumpTask<U, P>
where
U: UuidResolverHandle + Send + Sync + Clone + 'static,
P: UpdateActorHandle + Send + Sync + Clone + 'static,
{
async fn run(self) -> Result<()> {
trace!("Performing dump.");
create_dir_all(&self.path).await?;
let path_clone = self.path.clone();
let temp_dump_dir =
tokio::task::spawn_blocking(|| tempfile::TempDir::new_in(path_clone)).await??;
let temp_dump_path = temp_dump_dir.path().to_owned();
let meta = Metadata::new_v2(self.index_db_size, self.update_db_size);
let meta_path = temp_dump_path.join(META_FILE_NAME);
let mut meta_file = File::create(&meta_path)?;
serde_json::to_writer(&mut meta_file, &meta)?;
let uuids = self.uuid_resolver.dump(temp_dump_path.clone()).await?;
self.update_handle
.dump(uuids, temp_dump_path.clone())
.await?;
let dump_path = tokio::task::spawn_blocking(move || -> Result<PathBuf> {
let temp_dump_file = tempfile::NamedTempFile::new_in(&self.path)?;
crate::to_tar_gz(temp_dump_path, temp_dump_file.path())
.map_err(|e| DumpActorError::Internal(e.into()))?;
let dump_path = self.path.join(self.uid).with_extension("dump");
temp_dump_file.persist(&dump_path)?;
Ok(dump_path)
})
.await??;
info!("Created dump in {:?}.", dump_path);
Ok(())
}
}