feat: Force RocksDB compaction

meilidb-data/src/database/documents_addition.rs

@@ -122,6 +122,7 @@ impl<'a> DocumentsAddition<'a> {
         let ranked_map = self.ranked_map;
         let schema = lease_inner.schema.clone();
         let raw = lease_inner.raw.clone();
+        lease_inner.raw.compact();
 
         let inner = InnerIndex { words, schema, ranked_map, raw };
         self.inner.0.store(Arc::new(inner));

meilidb-data/src/database/documents_deletion.rs

@@ -121,6 +121,7 @@ impl<'a> DocumentsDeletion<'a> {
         let ranked_map = lease_inner.ranked_map.clone();
         let schema = lease_inner.schema.clone();
         let raw = lease_inner.raw.clone();
+        lease_inner.raw.compact();
 
         let inner = InnerIndex { words, schema, ranked_map, raw };
         self.inner.0.store(Arc::new(inner));

meilidb-data/src/database/raw_index.rs

@@ -10,6 +10,16 @@ pub struct RawIndex {
     pub custom: CustomSettings,
 }
 
+impl RawIndex {
+    pub(crate) fn compact(&self) {
+        self.main.0.compact_range(None::<&[u8]>, None::<&[u8]>);
+        self.words.0.compact_range(None::<&[u8]>, None::<&[u8]>);
+        self.docs_words.0.compact_range(None::<&[u8]>, None::<&[u8]>);
+        self.documents.0.compact_range(None::<&[u8]>, None::<&[u8]>);
+        self.custom.0.compact_range(None::<&[u8]>, None::<&[u8]>);
+    }
+}
+
 #[derive(Clone)]
 pub struct InnerRawIndex {
     database: Arc<rocksdb::DB>,
@@ -65,4 +75,12 @@ impl InnerRawIndex {
 
         self.database.write(batch)
     }
+
+    pub fn compact_range<S, E>(&self, start: Option<S>, end: Option<E>)
+    where S: AsRef<[u8]>,
+          E: AsRef<[u8]>,
+    {
+        let cf = self.database.cf_handle(&self.name).expect("cf not found");
+        self.database.compact_range_cf(cf, start, end)
+    }
 }

meilidb-data/src/database/words_index.rs

@@ -12,9 +12,24 @@ impl WordsIndex {
         // we must force an allocation to make the memory aligned
         match self.0.get(word)? {
             Some(bytes) => {
-                let layout = LayoutVerified::new_slice(bytes.as_ref()).expect("invalid layout");
-                let slice = layout.into_slice();
-                let setbuf = SetBuf::new_unchecked(slice.to_vec());
+                let vec = match LayoutVerified::new_slice(bytes.as_ref()) {
+                    Some(layout) => layout.into_slice().to_vec(),
+                    None => {
+                        let len = bytes.as_ref().len();
+                        let count = len / std::mem::size_of::<DocIndex>();
+                        let mut buf: Vec<DocIndex> = Vec::with_capacity(count);
+                        unsafe {
+                            let src = bytes.as_ref().as_ptr();
+                            let dst = buf.as_mut_ptr() as *mut u8;
+                            std::ptr::copy_nonoverlapping(src, dst, len);
+                            buf.set_len(count);
+                        }
+                        buf
+                    }
+                };
+
+                let setbuf = SetBuf::new_unchecked(vec);
+
                 Ok(Some(setbuf))
             },
             None => Ok(None),