clippy (#2661)

* clippy

* use readable version

---------

Co-authored-by: Pascal Seitz <pascal.seitz@datadoghq.com>

src/aggregation/bucket/term_agg.rs

@@ -518,7 +518,7 @@ impl SegmentTermCollector {
                 |term| {
                     let entry = entries[idx];
                     let intermediate_entry = into_intermediate_bucket_entry(entry.0, entry.1)
-                        .map_err(|err| io::Error::new(io::ErrorKind::Other, err))?;
+                        .map_err(io::Error::other)?;
                     dict.insert(
                         IntermediateKey::Str(
                             String::from_utf8(term.to_vec()).expect("could not convert to String"),

src/collector/facet_collector.rs

@@ -739,7 +739,7 @@ mod tests {
                 .flat_map(|(c, count)| {
                     let facet = Facet::from(&format!("/facet/{c}"));
                     let doc = doc!(facet_field => facet);
-                    iter::repeat(doc).take(count)
+                    std::iter::repeat_n(doc, count)
                 })
                 .map(|mut doc| {
                     doc.add_facet(
@@ -787,7 +787,7 @@ mod tests {
             .flat_map(|(c, count)| {
                 let facet = Facet::from(&format!("/facet/{c}"));
                 let doc = doc!(facet_field => facet);
-                iter::repeat(doc).take(count)
+                std::iter::repeat_n(doc, count)
             })
             .collect();
 

src/compat_tests.rs

@@ -30,7 +30,7 @@ fn create_format() {
 }
 
 fn path_for_version(version: &str) -> String {
-    format!("./tests/compat_tests_data/index_v{}/", version)
+    format!("./tests/compat_tests_data/index_v{version}/")
 }
 
 /// feature flag quickwit uses a different dictionary type

src/core/executor.rs

@@ -65,8 +65,7 @@ impl Executor {
                                 if let Err(err) = fruit_sender_ref.send((idx, fruit)) {
                                     error!(
                                         "Failed to send search task. It probably means all search \
-                                         threads have panicked. {:?}",
-                                        err
+                                         threads have panicked. {err:?}"
                                     );
                                 }
                             });

src/directory/directory.rs

@@ -56,7 +56,7 @@ impl<T: Send + Sync + 'static> From<Box<T>> for DirectoryLock {
 impl Drop for DirectoryLockGuard {
     fn drop(&mut self) {
         if let Err(e) = self.directory.delete(&self.path) {
-            error!("Failed to remove the lock file. {:?}", e);
+            error!("Failed to remove the lock file. {e:?}");
         }
     }
 }

src/directory/file_watcher.rs

@@ -51,7 +51,7 @@ impl FileWatcher {
                             .map(|current_checksum| current_checksum != checksum)
                             .unwrap_or(true);
                         if metafile_has_changed {
-                            info!("Meta file {:?} was modified", path);
+                            info!("Meta file {path:?} was modified");
                             current_checksum_opt = Some(checksum);
                             // We actually ignore callbacks failing here.
                             // We just wait for the end of their execution.
@@ -75,7 +75,7 @@ impl FileWatcher {
         let reader = match fs::File::open(path) {
             Ok(f) => io::BufReader::new(f),
             Err(e) => {
-                warn!("Failed to open meta file {:?}: {:?}", path, e);
+                warn!("Failed to open meta file {path:?}: {e:?}");
                 return Err(e);
             }
         };

src/directory/managed_directory.rs

@@ -157,7 +157,7 @@ impl ManagedDirectory {
         for file_to_delete in files_to_delete {
             match self.delete(&file_to_delete) {
                 Ok(_) => {
-                    info!("Deleted {:?}", file_to_delete);
+                    info!("Deleted {file_to_delete:?}");
                     deleted_files.push(file_to_delete);
                 }
                 Err(file_error) => {
@@ -170,7 +170,7 @@ impl ManagedDirectory {
                             if !cfg!(target_os = "windows") {
                                 // On windows, delete is expected to fail if the file
                                 // is mmapped.
-                                error!("Failed to delete {:?}", file_to_delete);
+                                error!("Failed to delete {file_to_delete:?}");
                             }
                         }
                     }

src/directory/mmap_directory.rs

@@ -29,7 +29,7 @@ pub type WeakArcBytes = Weak<dyn Deref<Target = [u8]> + Send + Sync + 'static>;
 
 /// Create a default io error given a string.
 pub(crate) fn make_io_err(msg: String) -> io::Error {
-    io::Error::new(io::ErrorKind::Other, msg)
+    io::Error::other(msg)
 }
 
 /// Returns `None` iff the file exists, can be read, but is empty (and hence
@@ -369,7 +369,7 @@ pub(crate) fn atomic_write(path: &Path, content: &[u8]) -> io::Result<()> {
 
 impl Directory for MmapDirectory {
     fn get_file_handle(&self, path: &Path) -> Result<Arc<dyn FileHandle>, OpenReadError> {
-        debug!("Open Read {:?}", path);
+        debug!("Open Read {path:?}");
         let full_path = self.resolve_path(path);
 
         let mut mmap_cache = self.inner.mmap_cache.write().map_err(|_| {
@@ -414,7 +414,7 @@ impl Directory for MmapDirectory {
     }
 
     fn open_write(&self, path: &Path) -> Result<WritePtr, OpenWriteError> {
-        debug!("Open Write {:?}", path);
+        debug!("Open Write {path:?}");
         let full_path = self.resolve_path(path);
 
         let open_res = OpenOptions::new()
@@ -467,7 +467,7 @@ impl Directory for MmapDirectory {
     }
 
     fn atomic_write(&self, path: &Path, content: &[u8]) -> io::Result<()> {
-        debug!("Atomic Write {:?}", path);
+        debug!("Atomic Write {path:?}");
         let full_path = self.resolve_path(path);
         atomic_write(&full_path, content)?;
         Ok(())

src/directory/ram_directory.rs

@@ -191,7 +191,7 @@ impl Directory for RamDirectory {
             .fs
             .read()
             .map_err(|e| OpenReadError::IoError {
-                io_error: Arc::new(io::Error::new(io::ErrorKind::Other, e.to_string())),
+                io_error: Arc::new(io::Error::other(e.to_string())),
                 filepath: path.to_path_buf(),
             })?
             .exists(path))

src/directory/watch_event_router.rs

@@ -90,10 +90,7 @@ impl WatchCallbackList {
                 let _ = sender.send(Ok(()));
             });
         if let Err(err) = spawn_res {
-            error!(
-                "Failed to spawn thread to call watch callbacks. Cause: {:?}",
-                err
-            );
+            error!("Failed to spawn thread to call watch callbacks. Cause: {err:?}");
         }
         result
     }

src/indexer/index_writer.rs

@@ -370,7 +370,7 @@ impl<D: Document> IndexWriter<D> {
             .map_err(|_| error_in_index_worker_thread("Failed to join merging thread."));
 
         if let Err(ref e) = result {
-            error!("Some merging thread failed {:?}", e);
+            error!("Some merging thread failed {e:?}");
         }
 
         result
@@ -644,7 +644,7 @@ impl<D: Document> IndexWriter<D> {
 
         let commit_opstamp = self.stamper.stamp();
         let prepared_commit = PreparedCommit::new(self, commit_opstamp);
-        info!("Prepared commit {}", commit_opstamp);
+        info!("Prepared commit {commit_opstamp}");
         Ok(prepared_commit)
     }
 

src/indexer/segment_updater.rs

@@ -501,8 +501,7 @@ impl SegmentUpdater {
             Ok(segment_entries) => segment_entries,
             Err(err) => {
                 warn!(
-                    "Starting the merge failed for the following reason. This is not fatal. {}",
-                    err
+                    "Starting the merge failed for the following reason. This is not fatal. {err}"
                 );
                 return err.into();
             }

src/indexer/segment_writer.rs

@@ -873,8 +873,8 @@ mod tests {
 
         fn assert_type(reader: &SegmentReader, field: &str, typ: ColumnType) {
             let cols = reader.fast_fields().dynamic_column_handles(field).unwrap();
-            assert_eq!(cols.len(), 1, "{}", field);
-            assert_eq!(cols[0].column_type(), typ, "{}", field);
+            assert_eq!(cols.len(), 1, "{field}");
+            assert_eq!(cols[0].column_type(), typ, "{field}");
         }
         assert_type(segment_reader, "json.toto", ColumnType::Str);
         assert_type(segment_reader, "json.float", ColumnType::F64);

src/positions/mod.rs

@@ -206,7 +206,7 @@ pub(crate) mod tests {
     #[test]
     fn test_position() -> crate::Result<()> {
         const CONST_VAL: u32 = 9u32;
-        let positions_delta: Vec<u32> = iter::repeat(CONST_VAL).take(2_000_000).collect();
+        let positions_delta: Vec<u32> = std::iter::repeat_n(CONST_VAL, 2_000_000).collect();
         let positions_data = create_positions_data(&positions_delta[..])?;
         assert_eq!(positions_data.len(), 1_015_627);
         let mut position_reader = PositionReader::open(positions_data)?;

src/query/boolean_query/block_wand.rs

@@ -436,7 +436,7 @@ mod tests {
         let fieldnorms_expanded = fieldnorms
             .iter()
             .cloned()
-            .flat_map(|fieldnorm| iter::repeat(fieldnorm).take(REPEAT))
+            .flat_map(|fieldnorm| std::iter::repeat_n(fieldnorm, REPEAT))
             .collect::<Vec<u32>>();
 
         let postings_lists_expanded: Vec<Vec<(DocId, u32)>> = posting_lists

src/query/exist_query.rs

@@ -384,7 +384,7 @@ mod tests {
                 .search(&ExistsQuery::new(field.to_string(), json_subpaths), &Count)
                 .unwrap_err()
                 .to_string(),
-            format!("The field does not exist: '{}'", field)
+            format!("The field does not exist: '{field}'")
         );
     }
 }

src/query/phrase_query/regex_phrase_query.rs

@@ -124,8 +124,7 @@ impl RegexPhraseQuery {
         if field_type != Type::Str {
             return Err(crate::TantivyError::SchemaError(format!(
                 "RegexPhraseQuery can only be used with a field of type text currently, but got \
-                 {:?}",
-                field_type
+                 {field_type:?}"
             )));
         }
 

src/query/phrase_query/regex_phrase_weight.rs

@@ -81,8 +81,7 @@ impl RegexPhraseWeight {
             num_terms += term_infos.len();
             if num_terms > self.max_expansions as usize {
                 return Err(crate::TantivyError::InvalidArgument(format!(
-                    "Phrase query exceeded max expansions {}",
-                    num_terms
+                    "Phrase query exceeded max expansions {num_terms}"
                 )));
             }
             let union = Self::get_union_from_term_infos(&term_infos, reader, &inverted_index)?;

src/query/range_query/range_query_fastfield.rs

@@ -177,8 +177,7 @@ impl Weight for FastFieldRangeWeight {
         } else {
             assert!(
                 maps_to_u64_fastfield(field_type.value_type()),
-                "{:?}",
-                field_type
+                "{field_type:?}"
             );
 
             let bounds = self.bounds.map_bound_res(|term| {
@@ -193,8 +192,7 @@ impl Weight for FastFieldRangeWeight {
                     val.to_u64()
                 } else {
                     return Err(TantivyError::InvalidArgument(format!(
-                        "Expected term with u64, i64, f64 or date, but got {:?}",
-                        term
+                        "Expected term with u64, i64, f64 or date, but got {term:?}"
                     )));
                 };
                 Ok(val)
@@ -254,12 +252,9 @@ fn search_on_json_numerical_field(
     else {
         return Ok(Box::new(EmptyScorer));
     };
-    let actual_column_type: NumericalType = col_type.numerical_type().unwrap_or_else(|| {
-        panic!(
-            "internal error: couldn't cast to numerical_type: {:?}",
-            col_type
-        )
-    });
+    let actual_column_type: NumericalType = col_type
+        .numerical_type()
+        .unwrap_or_else(|| panic!("internal error: couldn't cast to numerical_type: {col_type:?}"));
 
     let bounds = match typ.numerical_type().unwrap() {
         NumericalType::I64 => {
@@ -1058,7 +1053,7 @@ mod tests {
     fn doc_from_id_1(id: u64) -> Doc {
         let id = id * 1000;
         Doc {
-            id_name: format!("id_name{:010}", id),
+            id_name: format!("id_name{id:010}"),
             id,
         }
     }
@@ -1242,7 +1237,7 @@ mod tests {
 
         let field_path = |field: &str| {
             if json {
-                format!("json.{}", field)
+                format!("json.{field}")
             } else {
                 field.to_string()
             }

src/query/set_query.rs

@@ -52,7 +52,7 @@ impl TermSetQuery {
                     .iter()
                     .map(|key| (key.serialized_value_bytes(), 0)),
             )
-            .map_err(|e| std::io::Error::new(std::io::ErrorKind::Other, e))?;
+            .map_err(std::io::Error::other)?;
 
             sub_queries.push((
                 Occur::Should,

src/query/term_query/term_scorer.rs

@@ -172,7 +172,7 @@ mod tests {
             let doc = i * 10;
             doc_and_tfs.push((doc, 1u32 + doc % 3u32));
         }
-        let fieldnorms: Vec<u32> = std::iter::repeat(10u32).take(3_000).collect();
+        let fieldnorms: Vec<u32> = std::iter::repeat_n(10u32, 3_000).collect();
         let mut term_scorer = TermScorer::create_for_test(&doc_and_tfs, &fieldnorms, bm25_weight);
         assert_eq!(term_scorer.doc(), 0u32);
         term_scorer.shallow_seek(1289);
@@ -238,7 +238,7 @@ mod tests {
         doc_tfs.push((257, 3u32));
         doc_tfs.push((258, 1u32));
 
-        let fieldnorms: Vec<u32> = std::iter::repeat(20u32).take(300).collect();
+        let fieldnorms: Vec<u32> = std::iter::repeat_n(20u32, 300).collect();
         let bm25_weight = Bm25Weight::for_one_term(10, 129, 20.0);
         let mut docs = TermScorer::create_for_test(&doc_tfs[..], &fieldnorms[..], bm25_weight);
         assert_nearly_equals!(docs.block_max_score(), 2.5161593);
@@ -304,7 +304,7 @@ mod tests {
         writer.set_merge_policy(Box::new(NoMergePolicy));
         for _ in 0..3_000 {
             let term_freq = rng.gen_range(1..10000);
-            let words: Vec<&str> = std::iter::repeat("bbbb").take(term_freq).collect();
+            let words: Vec<&str> = std::iter::repeat_n("bbbb", term_freq).collect();
             let text = words.join(" ");
             writer.add_document(doc!(text_field=>text))?;
         }

src/reader/mod.rs

@@ -86,10 +86,7 @@ impl IndexReaderBuilder {
                 let inner_reader_arc_clone = inner_reader_arc.clone();
                 let callback = move || {
                     if let Err(err) = inner_reader_arc_clone.reload() {
-                        error!(
-                            "Error while loading searcher after commit was detected. {:?}",
-                            err
-                        );
+                        error!("Error while loading searcher after commit was detected. {err:?}");
                     }
                 };
                 let watch_handle = inner_reader_arc

src/reader/warming.rs

@@ -152,7 +152,7 @@ impl WarmingStateInner {
                 // rely on deterministic gc in tests
                 #[cfg(not(test))]
                 if let Err(err) = std::panic::catch_unwind(|| inner.lock().unwrap().gc_maybe()) {
-                    error!("Panic in Warmer GC {:?}", err);
+                    error!("Panic in Warmer GC {err:?}");
                 }
                 // avoid unused var warning in tests
                 #[cfg(test)]

src/schema/document/se.rs

@@ -55,13 +55,10 @@ where W: Write
         }
 
         if num_field_values != actual_length {
-            return Err(io::Error::new(
-                io::ErrorKind::Other,
-                format!(
-                    "Unexpected number of entries written to serializer, expected \
-                     {num_field_values} entries, got {actual_length} entries",
-                ),
-            ));
+            return Err(io::Error::other(format!(
+                "Unexpected number of entries written to serializer, expected {num_field_values} \
+                 entries, got {actual_length} entries",
+            )));
         }
 
         Ok(())
@@ -215,14 +212,11 @@ where W: Write
     /// Finishes writing the array to the writer and validates it.
     pub(crate) fn end(self) -> io::Result<()> {
         if self.expected_length != self.actual_length {
-            return Err(io::Error::new(
-                io::ErrorKind::Other,
-                format!(
-                    "Unexpected number of entries written to serializer, expected {} entries, got \
-                     {} entries",
-                    self.expected_length, self.actual_length,
-                ),
-            ));
+            return Err(io::Error::other(format!(
+                "Unexpected number of entries written to serializer, expected {} entries, got {} \
+                 entries",
+                self.expected_length, self.actual_length,
+            )));
         }
         Ok(())
     }
@@ -276,14 +270,11 @@ where W: Write
     /// Finishes writing the array to the writer and validates it.
     pub(crate) fn end(self) -> io::Result<()> {
         if self.expected_length != self.actual_length {
-            return Err(io::Error::new(
-                io::ErrorKind::Other,
-                format!(
-                    "Unexpected number of entries written to serializer, expected {} entries, got \
-                     {} entries",
-                    self.expected_length, self.actual_length,
-                ),
-            ));
+            return Err(io::Error::other(format!(
+                "Unexpected number of entries written to serializer, expected {} entries, got {} \
+                 entries",
+                self.expected_length, self.actual_length,
+            )));
         }
 
         // This should never fail if the above statement is valid.

src/store/reader.rs

@@ -52,7 +52,7 @@ impl BinarySerializable for DocStoreVersion {
             v => {
                 return Err(io::Error::new(
                     io::ErrorKind::InvalidData,
-                    format!("Invalid doc store version {}", v),
+                    format!("Invalid doc store version {v}"),
                 ))
             }
         })
@@ -320,7 +320,9 @@ impl StoreReader {
                     doc_pos = 0;
                 }
 
-                let alive = alive_bitset.map_or(true, |bitset| bitset.is_alive(doc_id));
+                let alive = alive_bitset
+                    .map(|bitset| bitset.is_alive(doc_id))
+                    .unwrap_or(true);
                 let res = if alive {
                     Some((curr_block.clone(), doc_pos))
                 } else {

src/store/store_compressor.rs

@@ -215,7 +215,7 @@ impl DedicatedThreadBlockCompressorImpl {
     fn send(&mut self, msg: BlockCompressorMessage) -> io::Result<()> {
         if self.tx.send(msg).is_err() {
             harvest_thread_result(self.join_handle.take())?;
-            return Err(io::Error::new(io::ErrorKind::Other, "Unidentified error."));
+            return Err(io::Error::other("Unidentified error."));
         }
         Ok(())
     }
@@ -231,11 +231,10 @@ impl DedicatedThreadBlockCompressorImpl {
 /// If the thread panicked, or if the result has already been harvested,
 /// returns an explicit error.
 fn harvest_thread_result(join_handle_opt: Option<JoinHandle<io::Result<()>>>) -> io::Result<()> {
-    let join_handle = join_handle_opt
-        .ok_or_else(|| io::Error::new(io::ErrorKind::Other, "Thread already joined."))?;
+    let join_handle = join_handle_opt.ok_or_else(|| io::Error::other("Thread already joined."))?;
     join_handle
         .join()
-        .map_err(|_err| io::Error::new(io::ErrorKind::Other, "Compressing thread panicked."))?
+        .map_err(|_err| io::Error::other("Compressing thread panicked."))?
 }
 
 #[cfg(test)]

src/termdict/fst_termdict/termdict.rs

@@ -13,7 +13,7 @@ use crate::postings::TermInfo;
 use crate::termdict::TermOrdinal;
 
 fn convert_fst_error(e: tantivy_fst::Error) -> io::Error {
-    io::Error::new(io::ErrorKind::Other, e)
+    io::Error::other(e)
 }
 
 const FST_VERSION: u32 = 1;
@@ -128,10 +128,9 @@ impl TermDictionary {
         let footer_size = u64::deserialize(&mut footer_len_bytes)?;
         let version = u32::deserialize(&mut footer_len_bytes)?;
         if version != FST_VERSION {
-            return Err(io::Error::new(
-                io::ErrorKind::Other,
-                format!("Unsupported fst version, expected {version}, found {FST_VERSION}",),
-            ));
+            return Err(io::Error::other(format!(
+                "Unsupported fst version, expected {version}, found {FST_VERSION}",
+            )));
         }
 
         let (fst_file_slice, values_file_slice) = main_slice.split_from_end(footer_size as usize);

src/termdict/mod.rs

@@ -84,20 +84,14 @@ impl TermDictionary {
         let mut dict_type = dict_type.read_bytes()?;
         let dict_type = u32::deserialize(&mut dict_type)?;
         let dict_type = DictionaryType::try_from(dict_type).map_err(|_| {
-            io::Error::new(
-                io::ErrorKind::Other,
-                format!("Unsupported dictionary type, found {dict_type}"),
-            )
+            io::Error::other(format!("Unsupported dictionary type, found {dict_type}"))
         })?;
 
         if dict_type != CURRENT_TYPE {
-            return Err(io::Error::new(
-                io::ErrorKind::Other,
-                format!(
-                    "Unsupported dictionary type, compiled tantivy with {CURRENT_TYPE:?}, but got \
-                     {dict_type:?}",
-                ),
-            ));
+            return Err(io::Error::other(format!(
+                "Unsupported dictionary type, compiled tantivy with {CURRENT_TYPE:?}, but got \
+                 {dict_type:?}",
+            )));
         }
 
         InnerTermDict::open(main_slice).map(TermDictionary)

src/tokenizer/ascii_folding_filter.rs

@@ -1605,35 +1605,35 @@ mod tests {
                    Ò Ó Ô Õ Ö Ø Œ Þ Ù Ú Û Ü Ý Ÿ à á â ã ä å æ ç è é ê ë ì í î ï ij
                    ð ñ ò ó ô õ ö ø œ ß þ ù ú û ü ý ÿ fi fl";
         let mut vec: Vec<&str> = vec!["Des", "mot", "cles", "A", "LA", "CHAINE"];
-        vec.extend(iter::repeat("A").take(6));
-        vec.extend(iter::repeat("AE").take(1));
-        vec.extend(iter::repeat("C").take(1));
-        vec.extend(iter::repeat("E").take(4));
-        vec.extend(iter::repeat("I").take(4));
-        vec.extend(iter::repeat("IJ").take(1));
-        vec.extend(iter::repeat("D").take(1));
-        vec.extend(iter::repeat("N").take(1));
-        vec.extend(iter::repeat("O").take(6));
-        vec.extend(iter::repeat("OE").take(1));
-        vec.extend(iter::repeat("TH").take(1));
-        vec.extend(iter::repeat("U").take(4));
-        vec.extend(iter::repeat("Y").take(2));
-        vec.extend(iter::repeat("a").take(6));
-        vec.extend(iter::repeat("ae").take(1));
-        vec.extend(iter::repeat("c").take(1));
-        vec.extend(iter::repeat("e").take(4));
-        vec.extend(iter::repeat("i").take(4));
-        vec.extend(iter::repeat("ij").take(1));
-        vec.extend(iter::repeat("d").take(1));
-        vec.extend(iter::repeat("n").take(1));
-        vec.extend(iter::repeat("o").take(6));
-        vec.extend(iter::repeat("oe").take(1));
-        vec.extend(iter::repeat("ss").take(1));
-        vec.extend(iter::repeat("th").take(1));
-        vec.extend(iter::repeat("u").take(4));
-        vec.extend(iter::repeat("y").take(2));
-        vec.extend(iter::repeat("fi").take(1));
-        vec.extend(iter::repeat("fl").take(1));
+        vec.extend(std::iter::repeat_n("A", 6));
+        vec.extend(std::iter::repeat_n("AE", 1));
+        vec.extend(std::iter::repeat_n("C", 1));
+        vec.extend(std::iter::repeat_n("E", 4));
+        vec.extend(std::iter::repeat_n("I", 4));
+        vec.extend(std::iter::repeat_n("IJ", 1));
+        vec.extend(std::iter::repeat_n("D", 1));
+        vec.extend(std::iter::repeat_n("N", 1));
+        vec.extend(std::iter::repeat_n("O", 6));
+        vec.extend(std::iter::repeat_n("OE", 1));
+        vec.extend(std::iter::repeat_n("TH", 1));
+        vec.extend(std::iter::repeat_n("U", 4));
+        vec.extend(std::iter::repeat_n("Y", 2));
+        vec.extend(std::iter::repeat_n("a", 6));
+        vec.extend(std::iter::repeat_n("ae", 1));
+        vec.extend(std::iter::repeat_n("c", 1));
+        vec.extend(std::iter::repeat_n("e", 4));
+        vec.extend(std::iter::repeat_n("i", 4));
+        vec.extend(std::iter::repeat_n("ij", 1));
+        vec.extend(std::iter::repeat_n("d", 1));
+        vec.extend(std::iter::repeat_n("n", 1));
+        vec.extend(std::iter::repeat_n("o", 6));
+        vec.extend(std::iter::repeat_n("oe", 1));
+        vec.extend(std::iter::repeat_n("ss", 1));
+        vec.extend(std::iter::repeat_n("th", 1));
+        vec.extend(std::iter::repeat_n("u", 4));
+        vec.extend(std::iter::repeat_n("y", 2));
+        vec.extend(std::iter::repeat_n("fi", 1));
+        vec.extend(std::iter::repeat_n("fl", 1));
         assert_eq!(folding_helper(latin1_string), vec);
     }
 

src/tokenizer/ngram_tokenizer.rs

@@ -287,7 +287,7 @@ impl Iterator for CodepointFrontiers<'_> {
     type Item = usize;
 
     fn next(&mut self) -> Option<usize> {
-        self.next_el.map(|offset| {
+        self.next_el.inspect(|&offset| {
             if self.s.is_empty() {
                 self.next_el = None;
             } else {
@@ -295,7 +295,6 @@ impl Iterator for CodepointFrontiers<'_> {
                 self.s = &self.s[first_codepoint_width..];
                 self.next_el = Some(offset + first_codepoint_width);
             }
-            offset
         })
     }
 }

src/tokenizer/tokenized_string.rs

@@ -32,8 +32,7 @@ impl BinarySerializable for PreTokenizedString {
         if let Ok(text) = serde_json::to_string(self) {
             <String as BinarySerializable>::serialize(&text, writer)
         } else {
-            Err(io::Error::new(
-                io::ErrorKind::Other,
+            Err(io::Error::other(
                 "Failed to dump PreTokenizedString to json.",
             ))
         }
@@ -45,8 +44,7 @@ impl BinarySerializable for PreTokenizedString {
         if let Ok(value) = serde_json::from_str(&json_text) {
             Ok(value)
         } else {
-            Err(io::Error::new(
-                io::ErrorKind::Other,
+            Err(io::Error::other(
                 "Failed to parse string data as PreTokenizedString.",
             ))
         }