perf(crates/yt/db/extractor_hash/realize): Allow passing in a `all_hashes`

This avoids having to re-fetch that value for each realized short hash
and massively speeds up the `process_line` code while running `select {file,split}`.

1 files changed, 28 insertions, 23 deletions

diff --git a/crates/yt/src/storage/db/extractor_hash.rs b/crates/yt/src/storage/db/extractor_hash.rs
index abe1f0f..b828348 100644
--- a/crates/yt/src/storage/db/extractor_hash.rs
+++ b/crates/yt/src/storage/db/extractor_hash.rs
@@ -64,8 +64,12 @@ impl FromStr for LazyExtractorHash {
 
 impl LazyExtractorHash {
     /// Turn the [`LazyExtractorHash`] into the [`ExtractorHash`]
-    pub(crate) async fn realize(self, app: &App) -> Result<ExtractorHash> {
-        ExtractorHash::from_short_hash(app, &self.value).await
+    pub(crate) async fn realize(
+        self,
+        app: &App,
+        all_hashes: Option<&[ExtractorHash]>,
+    ) -> Result<ExtractorHash> {
+        ExtractorHash::from_short_hash(app, &self.value, all_hashes).await
     }
 }
 
@@ -75,10 +79,27 @@ impl ExtractorHash {
         Self { hash }
     }
 
-    pub(crate) async fn from_short_hash(app: &App, s: &ShortHash) -> Result<Self> {
-        Ok(Self {
-            hash: Self::short_hash_to_full_hash(app, s).await?.hash,
-        })
+    pub(crate) async fn from_short_hash(
+        app: &App,
+        s: &ShortHash,
+        all_hashes: Option<&[Self]>,
+    ) -> Result<Self> {
+        let all_hashes = if let Some(all) = all_hashes {
+            all
+        } else {
+            &Self::get_all(app)
+                .await
+                .context("Failed to fetch all extractor-hashes from the database")?
+        };
+        let needed_chars = s.0.len();
+        for hash in all_hashes {
+            // PERFORMANCE(@bpeetz): This could avoid the string construction and just use a
+            // numeric equality check instead. <2025-07-15>
+            if hash.hash().to_hex()[..needed_chars] == s.0 {
+                return Ok(*hash);
+            }
+        }
+        bail!("Your shortend hash, does not match a real hash (this is probably a bug)!");
     }
 
     pub(crate) fn from_info_json(entry: &InfoJson) -> Self {
@@ -135,7 +156,7 @@ impl ExtractorHash {
         &self.hash
     }
 
-    pub(crate) async fn into_short_hash(&self, app: &App) -> Result<ShortHash> {
+    pub(crate) async fn as_short_hash(&self, app: &App) -> Result<ShortHash> {
         let needed_chars = if let Some(needed_chars) = EXTRACTOR_HASH_LENGTH.get() {
             *needed_chars
         } else {
@@ -159,22 +180,6 @@ impl ExtractorHash {
         ))
     }
 
-    async fn short_hash_to_full_hash(app: &App, s: &ShortHash) -> Result<Self> {
-        let all_hashes = Self::get_all(app)
-            .await
-            .context("Failed to fetch all extractor -hashesh from database")?;
-
-        let needed_chars = s.0.len();
-
-        for hash in all_hashes {
-            if hash.hash().to_hex()[..needed_chars] == s.0 {
-                return Ok(hash);
-            }
-        }
-
-        bail!("Your shortend hash, does not match a real hash (this is probably a bug)!");
-    }
-
     async fn get_needed_char_len(&self, app: &App) -> Result<usize> {
         debug!("Calculating the needed hash char length");
         let all_hashes = Self::get_all(app)