perf: tune translate_cds implementation (#80) (#81)

.gitattributes

@@ -1,3 +1,4 @@
+*.fasta filter=lfs diff=lfs merge=lfs -text
 tests/data/*.gz filter=lfs diff=lfs merge=lfs -text
 tests/data/*/*.gz filter=lfs diff=lfs merge=lfs -text
 tests/data/*.fasta filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -1,3 +1,5 @@
+perf.*
+
 /target
 /Cargo.lock
 

Cargo.toml

@@ -24,10 +24,10 @@ linked-hash-map = { version = "0.5.6", features = ["serde", "serde_impl"] }
 log = "0.4.17"
 md-5 = "0.10.5"
 nom = "7.1.3"
-phf = { version = "0.11.1", features = ["macros"] }
 postgres = { version = "0.19.4", features = ["with-chrono-0_4"] }
 quick_cache = "0.2.2"
 regex = "1.7.1"
+rustc-hash = "1.1.0"
 seqrepo = { version = "0.3.0" }
 serde = { version = "1.0.152", features = ["derive"] }
 serde_json = "1.0.93"
@@ -38,3 +38,8 @@ env_logger = "0.10.0"
 pretty_assertions = "1.3.0"
 rstest = "0.17.0"
 test-log = "0.2.11"
+criterion = "0.3"
+
+[[bench]]
+name = "translate_cds"
+harness = false

benches/translate_cds.rs

@@ -0,0 +1,27 @@
+use criterion::{criterion_group, criterion_main, Criterion};
+use hgvs::sequences::{translate_cds, TranslationTable};
+
+/// TTN FASTA string from https://www.ncbi.nlm.nih.gov/nuccore/NM_001126114.1
+static TTN_FASTA: &str = include_str!("TTN.fasta");
+
+lazy_static::lazy_static! {
+    /// Raw TTN sequence.
+    static ref SEQ_TTN: String = {
+        let mut seq = String::new();
+        for line in TTN_FASTA.lines() {
+            if !line.starts_with('>') {
+                seq.push_str(line);
+            }
+        }
+        seq
+    };
+}
+
+fn criterion_benchmark(c: &mut Criterion) {
+    c.bench_function("translate_cds TTN", |b| {
+        b.iter(|| translate_cds(&SEQ_TTN, true, "*", TranslationTable::Standard).unwrap())
+    });
+}
+
+criterion_group!(benches, criterion_benchmark);
+criterion_main!(benches);