Add RE2 and TCL to the benchmark harness.

This also adds a new utility, regex-run-one, to the benchmark suite.
This utility is a CLI tool that lets one count the number of regex
matches for any of the regex engines in the benchmark harness. e.g.,

  regex-run-one tcl '\w{5}z\w{5}' my-file

Will count the number of times the regex '\w{5}z\w{5}' matches in
my-file. Supported engines are: pcre1, pcre2, onig, re2, rust,
rust-bytes and tcl.

Author: BurntSushi

.travis.yml

@@ -20,8 +20,7 @@ script:
   - if [ "$TRAVIS_RUST_VERSION" = "nightly" ]; then
       travis_wait ./run-bench rust;
       travis_wait ./run-bench rust-bytes --no-run;
-      travis_wait ./run-bench rust-plugin --no-run;
-      travis_wait ./run-bench pcre --no-run;
+      travis_wait ./run-bench pcre1 --no-run;
       travis_wait ./run-bench onig --no-run;
       travis_wait cargo test --verbose --manifest-path=regex_macros/Cargo.toml;
     fi

benches/Cargo.toml

@@ -8,74 +8,58 @@ repository = "https://github.com/rust-lang/regex"
 documentation = "http://doc.rust-lang.org/regex/regex_syntax/index.html"
 homepage = "https://github.com/rust-lang/regex"
 description = "Regex benchmarks for Rust's and other engines."
-
 build = "build.rs"
 
 [dependencies]
+docopt = "0.6"
 lazy_static = "0.1"
 libc = "0.2"
 onig = { version = "0.4", optional = true }
 libpcre-sys = { version = "0.2", optional = true }
+memmap = "0.2"
 regex = { version = "0.1", path = ".." }
 regex_macros = { version = "0.1", path = "../regex_macros", optional = true }
 regex-syntax = { version = "0.3", path = "../regex-syntax" }
+rustc-serialize = "0.3"
 
 [build-dependencies]
+gcc = "0.3"
 pkg-config = "0.3"
 
+[[bin]]
+name = "regex-run-one"
+path = "src/main.rs"
+bench = false
+
 # Use features to conditionally compile benchmarked regexes, since not every
 # regex works on every engine. Additionally, it is useful to be able to build
 # each benchmark individually, so that not all dependencies are required to
 # run only one benchmark.
+#
+# Note that when running benchmarks, only ONE feature should be set at a time.
+# Doing anything else will probably result in weird "duplicate definition"
+# compiler errors.
+#
+# Tip: use the run-bench script in the root of this repository to run
+# benchmarks.
 [features]
-re-pcre = ["libpcre-sys"]
+re-pcre1 = ["libpcre-sys"]
 re-pcre2 = []
 re-onig = ["onig"]
+re-re2 = []
 re-rust = []
 re-rust-bytes = []
 re-rust-plugin = ["regex_macros"]
+re-tcl = []
 
-# Run the benchmarks on the default behavior of Regex::new.
-[[bench]]
-name = "rust"
-path = "src/bench_rust.rs"
-test = false
-bench = true
-
-# Run the benchmarks on the default behavior of bytes::Regex::new.
-[[bench]]
-name = "rust-bytes"
-path = "src/bench_rust_bytes.rs"
-test = false
-bench = true
-
-# Run the benchmarks on the default behavior of the `regex!` compiler plugin.
-[[bench]]
-name = "rust-plugin"
-path = "src/bench_rust_plugin.rs"
-test = false
-bench = true
-
-# Run the benchmarks on PCRE.
 [[bench]]
-name = "pcre"
-path = "src/bench_pcre.rs"
+name = "bench"
+path = "src/bench.rs"
 test = false
 bench = true
 
-# Run the benchmarks on PCRE2.
-[[bench]]
-name = "pcre2"
-path = "src/bench_pcre2.rs"
-test = false
-bench = true
-
-# Run the benchmarks on Oniguruma.
-[[bench]]
-name = "onig"
-path = "src/bench_onig.rs"
-test = false
-bench = true
+[profile.release]
+debug = true
 
 [profile.bench]
 debug = true

benches/build.rs

@@ -8,13 +8,12 @@
 // option. This file may not be copied, modified, or distributed
 // except according to those terms.
 
+extern crate gcc;
 extern crate pkg_config;
 
 use std::env;
 use std::process;
 
-use pkg_config::Config;
-
 macro_rules! we {
     ($($tt:tt)*) => {{
         use std::io::Write;
@@ -23,11 +22,37 @@ macro_rules! we {
 }
 
 fn main() {
+    // We only need to look for PCRE2 and RE2 because we roll the FFI bindings
+    // for those libraries ourselves from scratch. For PCRE1 and Oniguruma, we
+    // rely on other crates that do something similar to the dance below for
+    // us.
+
     let wants_pcre2 = env::var("CARGO_FEATURE_RE_PCRE2").is_ok();
-    let has_pcre2 =
-        Config::new().atleast_version("10.21").find("libpcre2-8").is_ok();
+    let has_pcre2 = pkg_config::Config::new().find("libpcre2-8").is_ok();
     if wants_pcre2 && !has_pcre2 {
         we!("pcre2 cannot be found by pkg-config");
         process::exit(1);
     }
+
+    let wants_re2 = env::var("CARGO_FEATURE_RE_RE2").is_ok();
+    let has_re2 = pkg_config::Config::new().find("re2").is_ok();
+    if wants_re2 {
+        if !has_re2 {
+            we!("re2 cannot be found by pkg-config");
+            process::exit(1);
+        }
+        gcc::Config::new()
+            .cpp(true)
+            .flag("-std=c++11")
+            .file("src/ffi/re2.cpp")
+            .compile("libcre2.a");
+        println!("cargo:rustc-link-lib=re2");
+    }
+
+    let wants_tcl = env::var("CARGO_FEATURE_RE_TCL").is_ok();
+    let has_tcl = pkg_config::Config::new().find("tcl").is_ok();
+    if wants_tcl && !has_tcl {
+        we!("tcl cannot be found by pkg-config");
+        process::exit(1);
+    }
 }

benches/compile

@@ -0,0 +1,6 @@
+#!/bin/sh
+
+exec cargo build \
+  --release \
+  --features 're-onig re-pcre1 re-pcre2 re-re2 re-rust re-rust-bytes re-tcl' \
+  "$@"

benches/log/.gitignore

@@ -0,0 +1 @@
+tmp

benches/run

@@ -0,0 +1,42 @@
+#!/bin/bash
+
+usage() {
+  echo "Usage: $(basename $0) [rust | rust-bytes | rust-plugin | pcre1 | pcre2 | re2 | onig | tcl ]" >&2
+  exit 1
+}
+
+if [ $# = 0 ] || [ $1 = '-h' ] || [ $1 = '--help' ]; then
+  usage
+fi
+
+which="$1"
+shift
+case $which in
+  rust)
+    exec cargo bench --bench bench --features re-rust "$@"
+    ;;
+  rust-bytes)
+    exec cargo bench --bench bench --features re-rust-bytes "$@"
+    ;;
+  rust-plugin)
+    exec cargo bench --bench bench --features re-rust-plugin "$@"
+    ;;
+  re2)
+    exec cargo bench --bench bench --features re-re2 "$@"
+    ;;
+  pcre1)
+    exec cargo bench --bench bench --features re-pcre1 "$@"
+    ;;
+  pcre2)
+    exec cargo bench --bench bench --features re-pcre2 "$@"
+    ;;
+  onig)
+    exec cargo bench --bench bench --features re-onig "$@"
+    ;;
+  tcl)
+    exec cargo bench --bench bench --features re-tcl "$@"
+    ;;
+  *)
+    usage
+    ;;
+esac