Add artifact_size DB table to store artifact component sizes

Author: Kobzol

database/schema.md

@@ -61,6 +61,22 @@ id          name        date        type
 1           LOCAL_TEST              release
 ```
 
+### artifact size
+
+Records the size of individual components (like `librustc_driver.so` or `libLLVM.so`) of a single
+artifact.
+
+This description includes:
+* component: normalized name of the component (hashes are removed)
+* size: size of the component in bytes
+
+```
+sqlite> select * from artifact_size limit 1;
+aid         component   size
+----------  ----------  ----------
+1           libLLVM.so  177892352
+```
+
 ### collection
 
 A "collection" of benchmarks tied only differing by the statistic collected.

database/src/pool.rs

@@ -86,6 +86,11 @@ pub trait Connection: Send + Sync {
         krate: &str,
         value: Duration,
     );
+
+    /// Records the size of an artifact component (like `librustc_driver.so` or `libLLVM.so`) in
+    /// bytes.
+    async fn record_artifact_size(&self, artifact: ArtifactIdNumber, component: &str, size: u64);
+
     /// Returns vector of bootstrap build times for the given artifacts. The kth
     /// element is the minimum build time for the kth artifact in `aids`, across
     /// all collections for the artifact, or none if there is no bootstrap data

database/src/pool/postgres.rs

@@ -8,7 +8,7 @@ use chrono::{DateTime, TimeZone, Utc};
 use hashbrown::HashMap;
 use native_tls::{Certificate, TlsConnector};
 use postgres_native_tls::MakeTlsConnector;
-use std::convert::TryFrom;
+use std::convert::{TryFrom, TryInto};
 use std::str::FromStr;
 use std::sync::Arc;
 use std::time::Duration;
@@ -246,6 +246,14 @@ static MIGRATIONS: &[&str] = &[
     drop table error_series;
     alter table error_new rename to error;
 "#,
+    r#"
+    create table artifact_size(
+        aid integer references artifact(id) on delete cascade on update cascade,
+        component text not null,
+        size integer not null,
+        UNIQUE(aid, component)
+    );
+    "#,
 ];
 
 #[async_trait::async_trait]
@@ -328,6 +336,7 @@ pub struct CachedStatements {
     select_runtime_pstat_series: Statement,
     insert_runtime_pstat: Statement,
     get_runtime_pstat: Statement,
+    record_artifact_size: Statement,
 }
 
 pub struct PostgresTransaction<'a> {
@@ -513,7 +522,14 @@ impl PostgresConnection {
                          order by sids.idx
                      ")
                     .await
-                    .unwrap()
+                    .unwrap(),
+                record_artifact_size: conn.prepare("
+                    insert into artifact_size (aid, component, size)
+                    values ($1, $2, $3)
+                    on conflict (aid, component)
+                    do update
+                    set size = excluded.size
+                ").await.unwrap(),
             }),
             conn,
         }
@@ -903,6 +919,17 @@ where
             .unwrap();
     }
 
+    async fn record_artifact_size(&self, artifact: ArtifactIdNumber, component: &str, size: u64) {
+        let size: i32 = size.try_into().expect("Too large artifact");
+        self.conn()
+            .execute(
+                &self.statements().record_artifact_size,
+                &[&(artifact.0 as i32), &component, &size],
+            )
+            .await
+            .unwrap();
+    }
+
     async fn artifact_id(&self, artifact: &ArtifactId) -> ArtifactIdNumber {
         let (name, date, ty) = match artifact {
             ArtifactId::Commit(commit) => (

database/src/pool/sqlite.rs

@@ -357,6 +357,16 @@ static MIGRATIONS: &[Migration] = &[
         alter table error_new rename to error;
     "#,
     ),
+    Migration::new(
+        r#"
+        create table artifact_size(
+            aid integer references artifact(id) on delete cascade on update cascade,
+            component text not null,
+            size integer not null,
+            UNIQUE(aid, component)
+        );
+    "#,
+    ),
 ];
 
 #[async_trait::async_trait]
@@ -791,6 +801,16 @@ impl Connection for SqliteConnection {
             .unwrap();
     }
 
+    async fn record_artifact_size(&self, artifact: ArtifactIdNumber, component: &str, size: u64) {
+        self.raw_ref()
+            .execute(
+                "insert or replace into artifact_size (aid, component, size)\
+                values (?, ?, ?)",
+                params![&artifact.0, &component, &size],
+            )
+            .unwrap();
+    }
+
     async fn get_bootstrap(&self, aids: &[ArtifactIdNumber]) -> Vec<Option<Duration>> {
         aids.iter()
             .map(|aid| {