Merge pull request #96 from quickwit-oss/update_tanti

upgrade to clap 4

Cargo.toml

@@ -21,7 +21,7 @@ serde = "1.0"
 serde_derive = "1.0"
 serde_json = "1.0"
 persistent = "0.4"
-clap = "3"
+clap = "4"
 ansi_term = "0.12"
 urlencoded = "0.6"
 mount = "0.4"

src/commands/bench.rs

@@ -12,10 +12,10 @@ use tantivy::schema::{Field, Schema};
 use tantivy::Index;
 
 pub fn run_bench_cli(matches: &ArgMatches) -> Result<(), String> {
-    let index_path = PathBuf::from(matches.value_of("index").unwrap());
-    let queries_path = PathBuf::from(matches.value_of("queries").unwrap()); // the unwrap is safe as long as it is comming from the main cli.
-    let num_repeat = ArgMatches::value_of_t(matches, "num_repeat")
-        .map_err(|e| format!("Failed to read num_repeat argument as an integer. {:?}", e))?;
+    let index_path = PathBuf::from(matches.get_one::<String>("index").unwrap());
+    let queries_path = PathBuf::from(matches.get_one::<String>("queries").unwrap()); // the unwrap is safe as long as it is comming from the main cli.
+    let num_repeat: usize = *ArgMatches::get_one(matches, "num_repeat")
+        .expect("Failed to read num_repeat argument as an integer");
     run_bench(&index_path, &queries_path, num_repeat).map_err(From::from)
 }
 

src/commands/index.rs

@@ -17,20 +17,20 @@ use time::Instant;
 use crate::commands::merge::run_merge;
 
 pub fn run_index_cli(argmatch: &ArgMatches) -> Result<(), String> {
-    let index_directory = PathBuf::from(argmatch.value_of("index").unwrap());
+    let index_directory = PathBuf::from(argmatch.get_one::<String>("index").unwrap());
     let document_source = argmatch
-        .value_of("file")
+        .get_one::<String>("file")
         .map(|path| DocumentSource::FromFile(PathBuf::from(path)))
         .unwrap_or(DocumentSource::FromPipe);
-    let no_merge = argmatch.is_present("nomerge");
-    let force_merge = argmatch.is_present("forcemerge");
-    let mut num_threads = ArgMatches::value_of_t(argmatch, "num_threads")
-        .map_err(|_| format!("Failed to read num_threads argument as an integer."))?;
+    let no_merge = argmatch.contains_id("nomerge");
+    let force_merge = argmatch.contains_id("forcemerge");
+    let mut num_threads: usize = *ArgMatches::get_one(argmatch, "num_threads")
+        .expect("Failed to read num_threads argument as an integer.");
     if num_threads == 0 {
         num_threads = 1;
     }
-    let buffer_size: usize = ArgMatches::value_of_t(argmatch, "memory_size")
-        .map_err(|_| format!("Failed to read the buffer size argument as an integer."))?;
+    let buffer_size: usize = *ArgMatches::get_one(argmatch, "memory_size")
+        .expect("Failed to read the buffer size argument as an integer.");
     let buffer_size_per_thread = buffer_size / num_threads;
     run_index(
         index_directory,

src/commands/inspect.rs

@@ -1,14 +1,14 @@
 use clap::ArgMatches;
-use tantivy::schema::Schema;
-use tantivy::space_usage::PerFieldSpaceUsage;
 use std::convert::From;
 use std::path::Path;
 use std::path::PathBuf;
 use tantivy;
+use tantivy::schema::Schema;
+use tantivy::space_usage::PerFieldSpaceUsage;
 use tantivy::Index;
 
 pub fn run_inspect_cli(matches: &ArgMatches) -> Result<(), String> {
-    let index_directory = PathBuf::from(matches.value_of("index").unwrap());
+    let index_directory = PathBuf::from(matches.get_one::<String>("index").unwrap());
     run_inspect(&index_directory).map_err(|e| format!("{:?}", e))
 }
 
@@ -28,14 +28,25 @@ fn run_inspect(directory: &Path) -> tantivy::Result<()> {
     let space_usage = searcher.space_usage()?;
     println!("Total bytes: {}", space_usage.total());
     println!("");
-    for (i, (segment_reader, segment_space_usage)) in segments.iter().zip(space_usage.segments().iter()).enumerate() {
+    for (i, (segment_reader, segment_space_usage)) in segments
+        .iter()
+        .zip(space_usage.segments().iter())
+        .enumerate()
+    {
         let section_count = i + 2;
-        println!("{}. Space usage for segment: `{}`", section_count, segment_reader.segment_id().uuid_string());
+        println!(
+            "{}. Space usage for segment: `{}`",
+            section_count,
+            segment_reader.segment_id().uuid_string()
+        );
         println!("==============================================================================");
         println!("Num docs: {}", segment_space_usage.num_docs());
         println!("Store space usage:");
         println!("Total bytes: {}", segment_space_usage.store().total());
-        println!("Offset bytes: {}", segment_space_usage.store().offsets_usage());
+        println!(
+            "Offset bytes: {}",
+            segment_space_usage.store().offsets_usage()
+        );
         println!("");
         println!("{}.1 Term dictionnary space usage", section_count);
         println!("--------------------------------");
@@ -75,6 +86,10 @@ fn print_fields_space_usage(schema: &Schema, per_field_space_usage: &PerFieldSpa
     println!("Total bytes: {}", per_field_space_usage.total());
     for (field, field_space_usage) in per_field_space_usage.fields() {
         let field_name = schema.get_field_name(*field);
-        println!("Field `{}` bytes: {}", field_name, field_space_usage.total());
+        println!(
+            "Field `{}` bytes: {}",
+            field_name,
+            field_space_usage.total()
+        );
     }
 }

src/commands/merge.rs

@@ -9,7 +9,7 @@ fn error_msg(err: tantivy::TantivyError) -> String {
 }
 
 pub fn run_merge_cli(argmatch: &ArgMatches) -> Result<(), String> {
-    let index_directory = PathBuf::from(argmatch.value_of("index").unwrap());
+    let index_directory = PathBuf::from(argmatch.get_one::<String>("index").unwrap());
     run_merge(index_directory).map_err(error_msg)
 
     // we rollback to force a gc.

src/commands/new.rs

@@ -10,7 +10,7 @@ use tantivy::schema::*;
 use tantivy::Index;
 
 pub fn run_new_cli(matches: &ArgMatches) -> Result<(), String> {
-    let index_directory = PathBuf::from(matches.value_of("index").unwrap());
+    let index_directory = PathBuf::from(matches.get_one::<String>("index").unwrap());
     run_new(index_directory).map_err(|e| format!("{:?}", e))
 }
 

src/commands/search.rs

@@ -15,16 +15,16 @@ use tantivy::schema::FieldType;
 use tantivy::{Index, TERMINATED};
 
 pub fn run_search_cli(matches: &ArgMatches) -> Result<(), String> {
-    let index_directory = PathBuf::from(matches.value_of("index").unwrap());
-    let query = matches.value_of("query").unwrap();
-    let agg = matches.value_of("aggregation");
+    let index_directory = PathBuf::from(matches.get_one::<String>("index").unwrap());
+    let query = matches.get_one::<String>("query").unwrap();
+    let agg = matches.get_one::<String>("aggregation");
     run_search(&index_directory, &query, &agg).map_err(|e| format!("{:?}", e))
 }
 
 fn run_search(
     directory: &Path,
     query: &str,
-    agg: &std::option::Option<&str>,
+    agg: &std::option::Option<&String>,
 ) -> tantivy::Result<()> {
     let index = Index::open_in_dir(directory)?;
     let schema = index.schema();

src/commands/serve.rs

@@ -40,9 +40,10 @@ use tantivy::{DocAddress, Score};
 use urlencoded::UrlEncodedQuery;
 
 pub fn run_serve_cli(matches: &ArgMatches) -> Result<(), String> {
-    let index_directory = PathBuf::from(matches.value_of("index").unwrap());
-    let port = ArgMatches::value_of_t(matches, "port").unwrap_or(3000u16);
-    let host_str = matches.value_of("host").unwrap_or("localhost");
+    let index_directory = PathBuf::from(matches.get_one::<String>("index").unwrap());
+    let port = ArgMatches::get_one(matches, "port").unwrap_or(&3000u16);
+    let fallback = "localhost".to_string();
+    let host_str = matches.get_one::<String>("host").unwrap_or(&fallback);
     let host = format!("{}:{}", host_str, port);
     run_serve(index_directory, &host).map_err(|e| format!("{:?}", e))
 }