[please do not merge] Rewrite one method in rust

.circleci/config.yml

@@ -26,7 +26,7 @@ commands:
     description: "Install correct version of bundler and load from cache or bundle install"
     steps:
       - run: sudo apt-get --allow-releaseinfo-change update
-      - run: sudo apt-get install cifs-utils unixodbc unixodbc-dev
+      - run: sudo apt-get install cifs-utils unixodbc unixodbc-dev cargo
       - run:
           name: Configure Bundler
           command: |

.gitignore

@@ -55,3 +55,4 @@ node_modules
 
 # SVN working copy
 /subversion_eads/*
+lib_jobs_rs/target

Gemfile

@@ -17,6 +17,7 @@ gem 'capistrano-yarn', '~> 2.0'
 gem 'devise', '>= 4.6.0'
 gem 'ed25519', '~> 1.3'
 gem 'faraday', '~> 2.7'
+gem 'ffi'
 gem "flipflop", git: "https://github.com/voormedia/flipflop.git", ref: "0d70d8e33483a9c0282ed8d6bca9c5ccd61e61e8"
 gem 'foreman'
 gem 'honeybadger'

Gemfile.lock

@@ -564,6 +564,7 @@ DEPENDENCIES
   ed25519 (~> 1.3)
   factory_bot_rails
   faraday (~> 2.7)
+  ffi
   flipflop!
   foreman
   honeybadger

app/models/oclc/lc_call_slips/keyword_field.rb

@@ -5,6 +5,11 @@ module LcCallSlips
     # Marc::DataField for the keywords that a selector
     # is interested in.
     class KeywordField
+      extend FFI::Library
+      c_lib_extension = /darwin/.match?(RUBY_PLATFORM) ? 'dylib' : 'so'
+      ffi_lib Rails.root.join('lib_jobs_rs', 'target', 'release', "liblib_jobs.#{c_lib_extension}")
+      attach_function :normalize_keyword, [:string], :strptr
+
       def initialize(field:, keywords:)
         @field = field
         @keywords = keywords
@@ -41,7 +46,7 @@ def word_is_keyword?(word)
       end
 
       def normalize(word)
-        word.sub(/[[:punct:]]?$/, '')
+        normalize_keyword(word)[0]
       end
     end
   end

config/initializers/cargo.rb

@@ -0,0 +1,4 @@
+# frozen_string_literal: true
+Dir.chdir(Rails.root.join("lib_jobs_rs")) do
+  `cargo build --release`
+end

lib_jobs_rs/Cargo.toml

@@ -0,0 +1,9 @@
+[package]
+name = "lib_jobs"
+version = "0.1.0"
+edition = "2021"
+
+[lib]
+crate-type = ["cdylib"]
+
+[dependencies]

lib_jobs_rs/src/keyword_field.rs

@@ -0,0 +1,50 @@
+use std::ffi::{CStr, CString};
+
+/// Make a keyword suitable for comparing to other keywords
+///
+/// # Examples
+/// ```
+/// use std::ffi::{CStr, CString};
+///
+/// let original = CString::new("My title /").unwrap().into_raw();
+/// let result = unsafe { lib_jobs::normalize_keyword(original) };
+/// assert_eq!(unsafe { CStr::from_ptr(result) }.to_str().unwrap(), "My title ");
+/// ```
+///
+/// # Safety
+///
+/// Since this function reads arbitrary memory from a C pointer,
+/// there are some safety considerations described in
+/// <https://doc.rust-lang.org/std/ffi/struct.CStr.html#method.from_ptr>
+#[no_mangle]
+pub unsafe extern "C" fn normalize_keyword(raw_string_ptr: *const i8) -> *const i8 {
+    let original_string = match unsafe { CStr::from_ptr(raw_string_ptr) }.to_str() {
+        Ok(string) => string,
+        Err(_) => ""
+    };
+    CString::new(normalize_string(original_string))
+        .expect("Could not create a CString, check for 0 byte errors")
+        .into_raw()
+}
+
+fn normalize_string(original_string: &str) -> &str {
+    match original_string.strip_suffix(|last_character: char| last_character.is_ascii_punctuation())
+    {
+        Some(cleaned) => cleaned,
+        None => original_string,
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use std::ffi::{CStr, CString};
+    #[test]
+    fn normalize_keyword() {
+        let original = CString::new("31 pages.").unwrap().into_raw();
+        let result = unsafe { super::normalize_keyword(original) };
+        assert_eq!(
+            unsafe { CStr::from_ptr(result) }.to_str().unwrap(),
+            "31 pages"
+        );
+    }
+}

lib_jobs_rs/src/lib.rs

@@ -0,0 +1,2 @@
+pub mod keyword_field;
+pub use keyword_field::normalize_keyword;