apply select clippy::pedantic recommendations

src/cmd/jsonl.rs

@@ -98,7 +98,6 @@ fn json_line_to_csv_record(value: &Value, headers: &[Vec<String>]) -> csv::Strin
 
  if let Some(value) = value {
  record.push_field(&match value {
- Value::Null => String::new(),
  Value::Bool(v) => {
  if v {
  String::from("true")

src/cmd/sample.rs

@@ -63,30 +63,28 @@ pub fn run(argv: &[&str]) -> CliResult<()> {
  let mut sample_size = args.arg_sample_size;
 
  let mut wtr = Config::new(&args.flag_output).writer()?;
- let sampled = match rconfig.indexed()? {
- Some(mut idx) => {
- if sample_size < 1.0 {
- sample_size *= idx.count() as f64;
- }
- if args.flag_seed.is_none() && do_random_access(sample_size as u64, idx.count()) {
- rconfig.write_headers(&mut *idx, &mut wtr)?;
- sample_random_access(&mut idx, sample_size as u64)?
- } else {
- let mut rdr = rconfig.reader()?;
- rconfig.write_headers(&mut rdr, &mut wtr)?;
- sample_reservoir(&mut rdr, sample_size as u64, args.flag_seed)?
- }
+ let sampled = if let Some(mut idx) = rconfig.indexed()? {
+ if sample_size < 1.0 {
+ sample_size *= idx.count() as f64;
  }
- _ => {
- debug!("no index");
- if sample_size < 1.0 {
- return fail!("Percentage sampling requires an index.");
- }
+ if args.flag_seed.is_none() && do_random_access(sample_size as u64, idx.count()) {
+ rconfig.write_headers(&mut *idx, &mut wtr)?;
+ sample_random_access(&mut idx, sample_size as u64)?
+ } else {
  let mut rdr = rconfig.reader()?;
  rconfig.write_headers(&mut rdr, &mut wtr)?;
  sample_reservoir(&mut rdr, sample_size as u64, args.flag_seed)?
  }
+ } else {
+ debug!("no index");
+ if sample_size < 1.0 {
+ return fail!("Percentage sampling requires an index.");
+ }
+ let mut rdr = rconfig.reader()?;
+ rconfig.write_headers(&mut rdr, &mut wtr)?;
+ sample_reservoir(&mut rdr, sample_size as u64, args.flag_seed)?
  };
+
  for row in sampled {
  wtr.write_byte_record(&row)?;
  }

src/cmd/schema.rs

@@ -434,15 +434,15 @@ fn get_unique_values(
  _ => freq_args.sequential_ftables(),
  }?;
 
- let unique_values_map = construct_map_of_unique_values(&headers, ftables)?;
+ let unique_values_map = construct_map_of_unique_values(&headers, &ftables)?;
 
  Ok(unique_values_map)
 }
 
 /// construct map of unique values keyed by header
 fn construct_map_of_unique_values(
  freq_csv_fields: &ByteRecord,
- frequency_tables: Vec<Frequencies<Vec<u8>>>,
+ frequency_tables: &[Frequencies<Vec<u8>>],
 ) -> CliResult<AHashMap<String, Vec<String>>> {
  let mut unique_values_map: AHashMap<String, Vec<String>> = AHashMap::new();
 

src/cmd/search.rs

@@ -87,7 +87,8 @@ pub fn run(argv: &[&str]) -> CliResult<()> {
  }
  let mut record = csv::ByteRecord::new();
  let mut flag_rowi: u64 = 1;
- let mut _matched_rows = String::from("");
+ #[allow(unused_assignments)]
+ let mut matched_rows = String::with_capacity(20); // to save on allocs
  while rdr.read_byte_record(&mut record)? {
  let mut m = sel.select(&record).any(|f| pattern.is_match(f));
  if args.flag_invert_match {
@@ -97,8 +98,8 @@ pub fn run(argv: &[&str]) -> CliResult<()> {
  if args.flag_flag.is_some() {
  flag_rowi += 1;
  record.push_field(if m {
- _matched_rows = flag_rowi.to_string();
- _matched_rows.as_bytes()
+ matched_rows = flag_rowi.to_string();
+ matched_rows.as_bytes()
  } else {
  b"0"
  });

src/cmd/searchset.rs

@@ -115,8 +115,11 @@ pub fn run(argv: &[&str]) -> CliResult<()> {
 
  let mut record = csv::ByteRecord::new();
  let mut flag_rowi: u64 = 1;
- let mut _matched_rows = String::new();
- let mut _match_list_with_row = String::new();
+ // to save allocs
+ #[allow(unused_assignments)]
+ let mut matched_rows = String::with_capacity(20);
+ #[allow(unused_assignments)]
+ let mut match_list_with_row = String::with_capacity(20);
  while rdr.read_byte_record(&mut record)? {
  let mut m = sel.select(&record).any(|f| {
  let matched = pattern.is_match(f);
@@ -136,12 +139,12 @@ pub fn run(argv: &[&str]) -> CliResult<()> {
  if do_match_list {
  flag_rowi += 1;
  record.push_field(if m {
- _matched_rows = flag_rowi.to_string();
+ matched_rows = flag_rowi.to_string();
  if args.flag_invert_match {
- _matched_rows.as_bytes()
+ matched_rows.as_bytes()
  } else {
- _match_list_with_row = format!("{_matched_rows};{match_list}");
- _match_list_with_row.as_bytes()
+ match_list_with_row = format!("{matched_rows};{match_list}");
+ match_list_with_row.as_bytes()
  }
  } else {
  b"0"

src/cmd/split.rs

@@ -67,7 +67,7 @@ pub fn run(argv: &[&str]) -> CliResult<()> {
  fs::create_dir_all(&args.arg_outdir)?;
 
  match args.rconfig().indexed()? {
- Some(idx) => args.parallel_split(idx),
+ Some(idx) => args.parallel_split(&idx),
  None => args.sequential_split(),
  }
 }
@@ -93,7 +93,7 @@ impl Args {
  Ok(())
  }
 
- fn parallel_split(&self, idx: Indexed<fs::File, fs::File>) -> CliResult<()> {
+ fn parallel_split(&self, idx: &Indexed<fs::File, fs::File>) -> CliResult<()> {
  let nchunks = util::num_of_chunks(idx.count() as usize, self.flag_size);
  let pool = ThreadPool::new(util::njobs(self.flag_jobs));
  for i in 0..nchunks {

src/cmd/stats.rs

@@ -488,9 +488,9 @@ impl Stats {
  pieces.push(buffer.format(1.5f64.mul_add(iqr, q3)).to_owned());
  // calculate skewnewss using Pearson's median skewness
  // https://en.wikipedia.org/wiki/Skewness#Pearson's_second_skewness_coefficient_(median_skewness)
- let _mean = self.online.unwrap().mean();
- let _stddev = self.online.unwrap().stddev();
- pieces.push(buffer.format((3.0 * (_mean - q2)) / _stddev).to_owned());
+ let mean = self.online.unwrap().mean();
+ let stddev = self.online.unwrap().stddev();
+ pieces.push(buffer.format((3.0 * (mean - q2)) / stddev).to_owned());
  }
  }
  match self.modes.as_mut() {

src/cmd/validate.rs

@@ -129,9 +129,8 @@ pub fn run(argv: &[&str]) -> CliResult<()> {
  }]
  });
  return fail!(header_error.to_string());
- } else {
- return fail!(format!("Cannot read header ({e})."));
  }
+ return fail!(format!("Cannot read header ({e})."));
  }
  }
  }
@@ -147,11 +146,10 @@ pub fn run(argv: &[&str]) -> CliResult<()> {
  }]
  });
  return fail!(validation_error.to_string());
- } else {
- return fail!(format!(
- r#"Validation error: {e}. Try "qsv fixlengths" or "qsv fmt" to fix it."#
- ));
  }
+ return fail!(format!(
+ r#"Validation error: {e}. Try "qsv fixlengths" or "qsv fmt" to fix it."#
+ ));
  }
  record_count += 1;
  }
@@ -516,12 +514,11 @@ fn to_json_instance(headers: &ByteRecord, record: &ByteRecord, schema: &Value) -
 
  // grab the first entry that's not a "null", since it just means value is optional
  for val in vec {
- if *val != null_type {
- return_val = val.as_str().expect("type info should be a JSON string");
- } else {
+ if *val == null_type {
  // keep looking
  continue;
  }
+ return_val = val.as_str().expect("type info should be a JSON string");
  }
 
  return_val

src/util.rs

@@ -132,17 +132,18 @@ pub fn show_env_vars() {
 }
 
 pub fn count_rows(conf: &Config) -> u64 {
- if let Some(idx) = conf.indexed().unwrap() {
- idx.count()
- } else {
- let mut rdr = conf.reader().unwrap();
- let mut count = 0u64;
- let mut record = csv::ByteRecord::new();
- while rdr.read_byte_record(&mut record).unwrap() {
- count += 1;
- }
- count
- }
+ conf.indexed().unwrap().map_or_else(
+ || {
+ let mut rdr = conf.reader().unwrap();
+ let mut count = 0u64;
+ let mut record = csv::ByteRecord::new();
+ while rdr.read_byte_record(&mut record).unwrap() {
+ count += 1;
+ }
+ count
+ },
+ |idx| idx.count(),
+ )
 }
 
 #[cfg(any(feature = "full", feature = "lite"))]