clippy clean

jean-pierreBoth · Jul 5, 2024 · a7fb9a9 · a7fb9a9
1 parent 00e8fb5
commit a7fb9a9
Show file tree

Hide file tree

Showing 13 changed files with 77 additions and 101 deletions.
diff --git a/Cargo.toml b/Cargo.toml
@@ -1,6 +1,6 @@
 [package]
 name = "hnsw_rs"
-version = "0.3.0"
+version = "0.3.1"
 authors = ["[email protected]"]
 description = "Ann based on Hierarchical Navigable Small World Graphs from Yu.A. Malkov and D.A Yashunin"
 license = "MIT/Apache-2.0"
@@ -95,7 +95,7 @@ anndists = { version = "0.1.2" }
 hdf5 = { version = "0.8" }
 ndarray = { version = "0.15" }
 skiplist = { version = "0.5" }
-tempfile = "3"
+tempfile = { version = "3" }
 
 
 [features]

diff --git a/examples/ann-glove25-angular.rs b/examples/ann-glove25-angular.rs
@@ -1,3 +1,5 @@
+#![allow(clippy::needless_range_loop)]
+
 use cpu_time::ProcessTime;
 use std::time::{Duration, SystemTime};
 
@@ -180,7 +182,7 @@ pub fn search<Dist>(
         let recall = knn_neighbours_d.iter().filter(|d| *d <= &max_dist).count();
         recalls.push(recall);
         let mut ratio = 0.;
-        if knn_neighbours_d.len() >= 1 {
+        if !knn_neighbours_d.is_empty() {
             ratio = knn_neighbours_d[knn_neighbours_d.len() - 1] / max_dist;
         }
         last_distances_ratio.push(ratio);

diff --git a/examples/ann-mnist-784-euclidean.rs b/examples/ann-mnist-784-euclidean.rs
@@ -1,3 +1,5 @@
+#![allow(clippy::needless_range_loop)]
+
 use cpu_time::ProcessTime;
 use std::time::{Duration, SystemTime};
 
@@ -132,7 +134,7 @@ pub fn main() {
             .count();
         recalls.push(recall);
         let mut ratio = 0.;
-        if knn_neighbours_dist.len() >= 1 {
+        if !knn_neighbours_dist.is_empty() {
             ratio = knn_neighbours_dist[knn_neighbours_dist.len() - 1] / max_dist;
         }
         last_distances_ratio.push(ratio);

diff --git a/examples/ann-sift1m-128-euclidean.rs b/examples/ann-sift1m-128-euclidean.rs
@@ -1,3 +1,5 @@
+#![allow(clippy::needless_range_loop)]
+
 use cpu_time::ProcessTime;
 use env_logger::Builder;
 use std::time::{Duration, SystemTime};
@@ -159,7 +161,7 @@ pub fn search<Dist>(
         let recall = knn_neighbours_d.iter().filter(|d| *d <= &max_dist).count();
         recalls.push(recall);
         let mut ratio = 0.;
-        if knn_neighbours_d.len() >= 1 {
+        if !knn_neighbours_d.is_empty() {
             ratio = knn_neighbours_d[knn_neighbours_d.len() - 1] / max_dist;
         }
         last_distances_ratio.push(ratio);

diff --git a/examples/levensthein.rs b/examples/levensthein.rs
@@ -31,11 +31,9 @@ fn main() {
     }
     words.push(String::from("abcdj"));
     //
-    let mut i = 0;
-    for w in &words {
+    for (i, w) in words.iter().enumerate() {
         let vec: Vec<u16> = w.chars().map(|c| c as u16).collect();
         hns.insert((&vec, i));
-        i = i + 1;
     }
     // create a filter
     let mut filter: Vec<usize> = Vec::new();

diff --git a/examples/random.rs b/examples/random.rs
@@ -1,3 +1,6 @@
+#![allow(clippy::needless_range_loop)]
+#![allow(clippy::range_zip_with_len)]
+
 use cpu_time::ProcessTime;
 use rand::distributions::Uniform;
 use rand::prelude::*;
@@ -16,10 +19,7 @@ fn main() {
     let unif = Uniform::<f32>::new(0., 1.);
     let mut data = Vec::with_capacity(nb_elem);
     for _ in 0..nb_elem {
-        let column = (0..dim)
-            .into_iter()
-            .map(|_| rng.sample(unif))
-            .collect::<Vec<f32>>();
+        let column = (0..dim).map(|_| rng.sample(unif)).collect::<Vec<f32>>();
         data.push(column);
     }
     // give an id to each data

diff --git a/src/datamap.rs b/src/datamap.rs
@@ -358,8 +358,8 @@ mod tests {
         let ef_construct = 25;
         let nb_connection = 10;
         let hnsw = Hnsw::<f32, DistL1>::new(nb_connection, nbcolumn, 16, ef_construct, DistL1 {});
-        for i in 0..data.len() {
-            hnsw.insert((&data[i], i));
+        for (i, d) in data.iter().enumerate() {
+            hnsw.insert((d, i));
         }
         // some loggin info
         hnsw.dump_layer_info();
@@ -424,28 +424,25 @@ mod tests {
         let ef_construct = 25;
         let nb_connection = 10;
         let hnsw = Hnsw::<u32, DistL1>::new(nb_connection, nbcolumn, 16, ef_construct, DistL1 {});
-        for i in 0..data.len() {
-            hnsw.insert((&data[i], i));
+        for (i, d) in data.iter().enumerate() {
+            hnsw.insert((d, i));
         }
         // some loggin info
         hnsw.dump_layer_info();
         // dump in a file.  Must take care of name as tests runs in // !!!
         let fname = "mmap_order_test";
         let directory = tempfile::tempdir().unwrap();
-        let _res = hnsw.file_dump(directory.path(), &fname);
+        let _res = hnsw.file_dump(directory.path(), fname);
         // now we have check that datamap seems  ok, test reload of hnsw with mmap
-        let datamap: DataMap =
-            DataMap::from_hnswdump::<u32>(directory.path(), &fname.to_string()).unwrap();
+        let datamap: DataMap = DataMap::from_hnswdump::<u32>(directory.path(), fname).unwrap();
         // testing type check
         assert!(datamap.check_data_type::<u32>());
         assert!(!datamap.check_data_type::<f32>());
         info!("Datamap iteration order checking");
         let keys = datamap.get_dataid_iter();
-        let mut ukey = 0usize;
-        for dataid in keys {
+        for (i, dataid) in keys.enumerate() {
             let v = datamap.get_data::<u32>(dataid).unwrap();
-            assert_eq!(v, &data[*dataid], "dataid = {}, ukey = {}", dataid, ukey);
-            ukey += 1;
+            assert_eq!(v, &data[*dataid], "dataid = {}, ukey = {}", dataid, i);
         }
         // rm files generated!
         let _ = std::fs::remove_file("mmap_order_test.hnsw.data");

diff --git a/src/flatten.rs b/src/flatten.rs
@@ -167,8 +167,8 @@ mod tests {
         let ef_construct = 25;
         let nb_connection = 10;
         let hnsw = Hnsw::<f32, DistL1>::new(nb_connection, nbcolumn, 16, ef_construct, DistL1 {});
-        for i in 0..data.len() {
-            hnsw.insert((&data[i], i));
+        for (i, d) in data.iter().enumerate() {
+            hnsw.insert((d, i));
         }
         // some loggin info
         hnsw.dump_layer_info();

diff --git a/src/hnsw.rs b/src/hnsw.rs
@@ -1701,25 +1701,21 @@ mod tests {
             ef_construct,
             dist::DistL1 {},
         );
-        for i in 0..data.len() {
-            hns.insert((&data[i], i));
+        for (i, d) in data.iter().enumerate() {
+            hns.insert((d, i));
         }
         let cpu_time = start.elapsed();
         println!(" test_insert_iter_point time inserting {:?}", cpu_time);
 
         hns.dump_layer_info();
         // now check iteration
-        let mut ptiter = hns.get_point_indexation().into_iter();
+        let ptiter = hns.get_point_indexation().into_iter();
         let mut nb_dumped = 0;
-        loop {
-            if let Some(_point) = ptiter.next() {
-                //    println!("point : {:?}", _point.p_id);
-                nb_dumped += 1;
-            } else {
-                break;
-            }
-        } // end while
-          //
+        for _point in ptiter {
+            //    println!("point : {:?}", _point.p_id);
+            nb_dumped += 1;
+        }
+        //
         assert_eq!(nb_dumped, nbcolumn);
     } // end of test_iter_point
 
@@ -1753,8 +1749,8 @@ mod tests {
             ef_construct,
             dist::DistL1 {},
         );
-        for i in 0..data.len() {
-            hns.insert((&data[i], i));
+        for (i, d) in data.iter().enumerate() {
+            hns.insert((d, i));
         }
         let cpu_time = start.elapsed();
         println!(" test_insert_iter_point time inserting {:?}", cpu_time);
@@ -1763,17 +1759,13 @@ mod tests {
         // now check iteration
         let layer_num = 0;
         let nbpl = hns.get_point_indexation().get_layer_nb_point(layer_num);
-        let mut layer_iter = hns.get_point_indexation().get_layer_iterator(layer_num);
+        let layer_iter = hns.get_point_indexation().get_layer_iterator(layer_num);
         //
         let mut nb_dumped = 0;
-        loop {
-            if let Some(_point) = layer_iter.next() {
-                //    println!("point : {:?}", _point.p_id);
-                nb_dumped += 1;
-            } else {
-                break;
-            }
-        } // end while
+        for _point in layer_iter {
+            //    println!("point : {:?}", _point.p_id);
+            nb_dumped += 1;
+        }
         println!(
             "test_iter_layerpoint : nb point in layer {} , nb found {}",
             nbpl, nb_dumped

diff --git a/src/hnswio.rs b/src/hnswio.rs
@@ -1372,7 +1372,7 @@ mod tests {
 
     fn my_fn(v1: &[f32], v2: &[f32]) -> f32 {
         let norm_l1: f32 = v1.iter().zip(v2.iter()).map(|t| (*t.0 - *t.1).abs()).sum();
-        norm_l1 as f32
+        norm_l1
     }
 
     #[test]
@@ -1404,8 +1404,8 @@ mod tests {
             ef_construct,
             dist::DistL1 {},
         );
-        for i in 0..data.len() {
-            hnsw.insert((&data[i], i));
+        for (i, d) in data.iter().enumerate() {
+            hnsw.insert((d, i));
         }
         // some loggin info
         hnsw.dump_layer_info();
@@ -1454,8 +1454,8 @@ mod tests {
             ef_construct,
             mydist,
         );
-        for i in 0..data.len() {
-            hnsw.insert((&data[i], i));
+        for (i, d) in data.iter().enumerate() {
+            hnsw.insert((d, i));
         }
         // some loggin info
         hnsw.dump_layer_info();
@@ -1502,8 +1502,8 @@ mod tests {
             ef_construct,
             dist::DistL1 {},
         );
-        for i in 0..data.len() {
-            hnsw.insert((&data[i], i));
+        for (i, d) in data.iter().enumerate() {
+            hnsw.insert((d, i));
         }
         // some loggin info
         hnsw.dump_layer_info();
@@ -1554,8 +1554,8 @@ mod tests {
             ef_construct,
             dist::DistL1 {},
         );
-        for i in 0..data.len() {
-            hnsw.insert((&data[i], i));
+        for (i, d) in data.iter().enumerate() {
+            hnsw.insert((d, i));
         }
         // some loggin info
         hnsw.dump_layer_info();
@@ -1593,8 +1593,8 @@ mod tests {
             first_with_mmap
         );
         let nb_in = hnsw.get_nb_point();
-        for i in 0..data.len() {
-            hnsw.insert((&data[i], i + nb_in));
+        for (i, d) in data.iter().enumerate() {
+            hnsw.insert((d, i + nb_in));
         }
         //
         let search_res = hnsw.search(&first, 5, ef_construct);
@@ -1625,7 +1625,7 @@ mod tests {
         //
         // TODO: redump  and care about mmapped file, so we do not overwrite
         //
-        let dump_init = DumpInit::new(directory.path(), &fname, false);
+        let dump_init = DumpInit::new(directory.path(), fname, false);
         info!("will use basename : {}", dump_init.get_basename());
         let res = hnsw.file_dump(directory.path(), dump_init.get_basename());
         if res.is_err() {
@@ -1663,8 +1663,8 @@ mod tests {
             Hnsw::<f32, dist::DistL1>::new(nb_connection, 0, 16, ef_construct, dist::DistL1 {});
         let fname = "empty_db";
         let directory = tempfile::tempdir()?;
-        let _res = hnsw.file_dump(directory.path(), &fname);
-        let mut reloader = HnswIo::new(directory.path(), &fname);
+        let _res = hnsw.file_dump(directory.path(), fname);
+        let mut reloader = HnswIo::new(directory.path(), fname);
         let hnsw_loaded_res = reloader.load_hnsw::<f32, DistL1>();
         assert!(hnsw_loaded_res.is_err());
         Ok(())

diff --git a/tests/deallocation_test.rs b/tests/deallocation_test.rs
@@ -29,6 +29,6 @@ fn main() {
         if counter % 1_000_000 == 0 {
             println!("counter : {}", counter)
         }
-        counter = counter + 1;
+        counter += 1;
     }
 }
diff --git a/tests/filtertest.rs b/tests/filtertest.rs
@@ -1,3 +1,6 @@
+#![allow(clippy::needless_range_loop)]
+#![allow(clippy::range_zip_with_len)]
+
 use anndists::dist::*;
 use hnsw_rs::prelude::*;
 use rand::{distributions::Uniform, Rng};
@@ -19,19 +22,14 @@ fn generate_random_string(len: usize) -> String {
 fn search_closure_filter(
     word: &str,
     hns: &Hnsw<u16, DistLevenshtein>,
-    words: &Vec<String>,
-    filter_vector: &Vec<usize>,
+    words: &[String],
+    filter_vector: &[usize],
 ) {
     // transform string to u16 values
     let vec: Vec<u16> = word.chars().map(|c| c as u16).collect();
     // now create a closure using this filter_vector
     // here we can off course implement more advanced filter logic
-    let filter = |id: &usize| -> bool {
-        match filter_vector.binary_search(id) {
-            Ok(_) => true,
-            Err(_) => false,
-        }
-    };
+    let filter = |id: &usize| -> bool { filter_vector.binary_search(id).is_ok() };
 
     // Now let us do the search by using the defined clojure, which in turn uses our vector
     // ids not in the vector will not be indluced in the search results
@@ -65,11 +63,9 @@ fn filter_levenstein() {
         words.push(tw);
     }
 
-    let mut i = 0;
-    for w in &words {
+    for (i, w) in words.iter().enumerate() {
         let vec: Vec<u16> = w.chars().map(|c| c as u16).collect();
         hns.insert((&vec, i));
-        i = i + 1;
         if i % 1000 == 0 {
             println!("Inserting: {:?}", i);
         }
@@ -158,10 +154,7 @@ fn filter_l2() {
     let unif = Uniform::<f32>::new(0., 1.);
     let mut data = Vec::with_capacity(nb_elem);
     for _ in 0..nb_elem {
-        let column = (0..dim)
-            .into_iter()
-            .map(|_| rng.sample(unif))
-            .collect::<Vec<f32>>();
+        let column = (0..dim).map(|_| rng.sample(unif)).collect::<Vec<f32>>();
         data.push(column);
     }
     // give an id to each data
@@ -176,10 +169,7 @@ fn filter_l2() {
     //
     let ef_search = 30;
     let knbn = 10;
-    let vec_tosearch = (0..dim)
-        .into_iter()
-        .map(|_| rng.sample(unif))
-        .collect::<Vec<f32>>();
+    let vec_tosearch = (0..dim).map(|_| rng.sample(unif)).collect::<Vec<f32>>();
     //
     // Create a sorted vector of ids
     // the ids in the vector will be used as a filter