xre2 0.1.0 → 0.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 55ca051b3f858164be0f5e79831c1ca4b7df267d4e3597f8c128db79524b4f0a
-  data.tar.gz: 0a2d988389268e3cac55d5970075111f74c110479bf1bfc0213b7ee989424872
+  metadata.gz: 6401693cee9fee608b66f07638eb684c95a0ee5d16519aca6d272920cc8bbe3a
+  data.tar.gz: 9c6a061de9b45aeb7acbc5404dd0b21c3f0f394a3f54632e993a335b6a6da5de
 SHA512:
-  metadata.gz: 0fc6f00f3d0283128e09b16bb61210f8de3e9f74f4605b730a36d2790f8a78a6adb2e461f089a585324663b03b98545bcffedfdf59f872f285a423f1cf9d08c0
-  data.tar.gz: ed746fbdf6a0da543a08a5b8259e34541d6feb92f46605a509a931b2419d0d6cefe70fff0cdea2163d4c0e1e153fa2833d3fa6e772160ce0e9d7306093459648
+  metadata.gz: 8feee07ba3c54a1ffda829a7b4903dd9ef9dc909dd8ba87b1c5cfffa3ee06a75ae712ae71da69ade4cb14f193e7d8f9f78a8d0696cb3e72bef3e583b07c85677
+  data.tar.gz: 757d878581016a0f3299b2d4c677e4c4db129ff864469fba7f24b91b147066c461e1caa1c607e8981487768e0cf538cbe63618108c9d77fc21defca8109d22f1

data/Cargo.lock

@@ -106,10 +106,6 @@ name = "cc"
 version = "1.0.90"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "8cd6604a82acf3039f1144f54b8eb34e91ffba622051189e71b781822d5ee1f5"
-dependencies = [
- "jobserver",
- "libc",
-]
 
 [[package]]
 name = "cexpr"
@@ -231,15 +227,6 @@ dependencies = [
  "either",
 ]
 
-[[package]]
-name = "jobserver"
-version = "0.1.28"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "ab46a6e9526ddef3ae7f787c06f0f2600639ba80ea3eade3d8e670a2230f51d6"
-dependencies = [
- "libc",
-]
-
 [[package]]
 name = "lazy_static"
 version = "1.4.0"
@@ -282,9 +269,9 @@ checksum = "90ed8c1e510134f979dbc4f070f87d4313098b704861a105fe34231c70a3901c"
 
 [[package]]
 name = "magnus"
-version = "0.6.2"
+version = "0.6.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4778544796676e8428e9c622460ebf284bea52d8b10db3aeb449d8b5e61b3a13"
+checksum = "0fc7a31fb0b64761e3cd09a6975577601fccc5f08b8fc9245064fc4f71ed6a9d"
 dependencies = [
  "magnus-macros",
  "rb-sys",
@@ -354,28 +341,6 @@ dependencies = [
  "pkg-config",
 ]
 
-[[package]]
-name = "pcre2"
-version = "0.2.7"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5ea92ff5eabd27703ab12cefe01b08b2809ec3dc75fdc69d4e6b75fbce0cbd67"
-dependencies = [
- "libc",
- "log",
- "pcre2-sys",
-]
-
-[[package]]
-name = "pcre2-sys"
-version = "0.2.9"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "550f5d18fb1b90c20b87e161852c10cde77858c3900c5059b5ad2a1449f11d8a"
-dependencies = [
- "cc",
- "libc",
- "pkg-config",
-]
-
 [[package]]
 name = "peeking_take_while"
 version = "0.1.2"
@@ -510,9 +475,9 @@ checksum = "8ea5119cdb4c55b55d432abb513a0429384878c15dde60cc77b1c99de1a95a6a"
 
 [[package]]
 name = "syn"
-version = "2.0.55"
+version = "2.0.57"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "002a1b3dbf967edfafc32655d0f377ab0bb7b994aa1d32c8cc7e9b8bf3ebb8f0"
+checksum = "11a6ae1e52eb25aab8f3fb9fca13be982a373b8f1157ca14b897a825ba4a2d35"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -543,6 +508,12 @@ version = "1.0.12"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "3354b9ac3fae1ff6755cb6db53683adb661634f67557942dea4facebec0fee4b"
 
+[[package]]
+name = "unicode-segmentation"
+version = "1.11.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d4c87d22b6e3f4a18d4d40ef354e97c90fcb14dd91d7dc0aa9d8a1172ebf7202"
+
 [[package]]
 name = "unicode-width"
 version = "0.1.11"
@@ -671,5 +642,6 @@ dependencies = [
  "fancy-regex",
  "magnus",
  "onig",
- "pcre2",
+ "regex",
+ "unicode-segmentation",
 ]

data/README.md

@@ -1,24 +1,20 @@
 # Xre2
 
-TODO: Delete this and the text below, and describe your gem
-
-Welcome to your new gem! In this directory, you'll find the files you need to be able to package up your Ruby library into a gem. Put your Ruby code in the file `lib/xre2`. To experiment with that code, run `bin/console` for an interactive prompt.
+some rust regexes for ruby
 
 ## Installation
 
-TODO: Replace `UPDATE_WITH_YOUR_GEM_NAME_IMMEDIATELY_AFTER_RELEASE_TO_RUBYGEMS_ORG` with your gem name right after releasing it to RubyGems.org. Please do not do it earlier due to security reasons. Alternatively, replace this section with instructions to install your gem from git if you don't plan to release to RubyGems.org.
-
 Install the gem and add to the application's Gemfile by executing:
 
-    $ bundle add UPDATE_WITH_YOUR_GEM_NAME_IMMEDIATELY_AFTER_RELEASE_TO_RUBYGEMS_ORG
+    $ bundle add xre2
 
 If bundler is not being used to manage dependencies, install the gem by executing:
 
-    $ gem install UPDATE_WITH_YOUR_GEM_NAME_IMMEDIATELY_AFTER_RELEASE_TO_RUBYGEMS_ORG
+    $ gem install xre2
 
 ## Usage
 
-TODO: Write usage instructions here
+todo
 
 ## Development
 
@@ -28,7 +24,7 @@ To install this gem onto your local machine, run `bundle exec rake install`. To
 
 ## Contributing
 
-Bug reports and pull requests are welcome on GitHub at https://github.com/[USERNAME]/xre2. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the [code of conduct](https://github.com/[USERNAME]/xre2/blob/main/CODE_OF_CONDUCT.md).
+Bug reports and pull requests are welcome on GitHub at https://github.com/vagab/xre2. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the [code of conduct](https://github.com/vagab/xre2/blob/main/CODE_OF_CONDUCT.md).
 
 ## License
 
@@ -36,4 +32,4 @@ The gem is available as open source under the terms of the [MIT License](https:/
 
 ## Code of Conduct
 
-Everyone interacting in the Xre2 project's codebases, issue trackers, chat rooms and mailing lists is expected to follow the [code of conduct](https://github.com/[USERNAME]/xre2/blob/main/CODE_OF_CONDUCT.md).
+Everyone interacting in the Xre2 project's codebases, issue trackers, chat rooms and mailing lists is expected to follow the [code of conduct](https://github.com/vagab/xre2/blob/main/CODE_OF_CONDUCT.md).

data/ext/xre2/Cargo.toml

@@ -11,6 +11,7 @@ crate-type = ["cdylib"]
 
 [dependencies]
 fancy-regex = "0.13.0"
-magnus = { version = "0.6.2" }
+magnus = "0.6.2"
 onig = "6.4.0"
-pcre2 = "0.2.7"
+regex = "1.10.4"
+unicode-segmentation = "1.11.0"

data/ext/xre2/src/lib.rs

@@ -1,34 +1,196 @@
 use magnus::{function, prelude::*, Error, Ruby};
-use pcre2::bytes::*;
+use std::iter::Enumerate;
+use std::ops::Range;
+use std::str::CharIndices;
+use unicode_segmentation::UnicodeSegmentation;
+
+#[magnus::wrap(class = "Regexx")]
+struct Regex {
+    regex: regex::Regex,
+}
+
+#[magnus::wrap(class = "RegexxList")]
+struct RegexList {
+    regexes: Vec<regex::Regex>,
+}
+
+impl RegexList {
+    fn new(regexes: Vec<String>) -> Self {
+        Self {
+            regexes: regexes.iter().map(|r| regex(r.to_string()).regex).collect(),
+        }
+    }
+}
+
+fn regex(target: String) -> Regex {
+    let regex = regex::RegexBuilder::new(&target)
+        .multi_line(true)
+        .build()
+        .unwrap();
+
+    Regex { regex }
+}
+
+fn reg_array(target: &RegexList, subject: String) -> Vec<(String, usize)> {
+    target
+        .regexes
+        .iter()
+        .map(|t| reg_compiled_array(t, &subject))
+        .flatten()
+        .collect()
+}
+
+fn reg_array_with_context(target: &RegexList, subject: String) -> Vec<(String, usize, String)> {
+    target
+        .regexes
+        .iter()
+        .map(|t| reg_compiled_array_with_context(t, &subject))
+        .flatten()
+        .collect()
+}
+
+fn reg_compiled_array(target: &regex::Regex, subject: &str) -> Vec<(String, usize)> {
+    let mut chars_iterator = subject.char_indices().enumerate();
+
+    target
+        .captures_iter(subject)
+        .map(|m| m.get(1).unwrap())
+        .map(|m| {
+            (
+                m.as_str().trim().to_string(),
+                find_char_index(&mut chars_iterator, m.start()),
+            )
+        })
+        .collect()
+}
+
+fn reg_compiled_array_with_context(
+    target: &regex::Regex,
+    subject: &str,
+) -> Vec<(String, usize, String)> {
+    let mut chars_iterator = subject.char_indices().enumerate();
+
+    target
+        .captures_iter(subject)
+        .map(|m| m.get(1).unwrap())
+        .map(|m| {
+            (
+                m.as_str().trim().to_string(),
+                find_char_index(&mut chars_iterator, m.start()),
+                extract_from_str(&subject, m.start()..m.end(), 10),
+            )
+        })
+        .collect()
+}
+
+fn reg_compiled(target: &Regex, subject: String) -> Vec<(String, usize)> {
+    let mut chars_iterator = subject.char_indices().enumerate();
+
+    target
+        .regex
+        .captures_iter(&subject)
+        .map(|m| m.get(1).unwrap())
+        .map(|m| {
+            (
+                m.as_str().trim().to_string(),
+                find_char_index(&mut chars_iterator, m.start()),
+            )
+        })
+        .collect()
+}
+
+fn reg_compiled_with_context(target: &Regex, subject: String) -> Vec<(String, usize, String)> {
+    let mut chars_iterator = subject.char_indices().enumerate();
+
+    target
+        .regex
+        .captures_iter(&subject)
+        .map(|m| m.get(1).unwrap())
+        .map(|m| {
+            (
+                m.as_str().trim().to_string(),
+                find_char_index(&mut chars_iterator, m.start()),
+                extract_from_str(&subject, m.start()..m.end(), 10),
+            )
+        })
+        .collect()
+}
+
+fn find_char_index(iterator: &mut Enumerate<CharIndices>, byte_start: usize) -> usize {
+    iterator
+        .find_map(|(idx, (byte_index, _))| (byte_index >= byte_start).then_some(idx))
+        .unwrap_or(0)
+}
+
+fn extract_from_vec(vec: &Vec<char>, range: Range<usize>, radius: usize) -> String {
+    vec[range.start.saturating_sub(radius)..range.end.saturating_add(radius).min(vec.len() - 1)]
+        .iter()
+        .collect()
+}
+
+fn extract_from_str(subject: &str, range: Range<usize>, radius: usize) -> String {
+    let l_text = subject[..range.start]
+        .graphemes(true)
+        .rev()
+        .take(radius)
+        .collect::<String>()
+        .chars()
+        .rev()
+        .collect::<String>();
+    let r_text = subject[range.end..]
+        .graphemes(true)
+        .skip(1)
+        .take(radius)
+        .collect::<String>();
+    format!("{}{}{}", l_text, &subject[range], r_text)
+}
+
+fn onig(target: String, subject: String) -> Vec<(String, usize)> {
+    let mut chars_iterator = subject.char_indices().enumerate();
 
-fn onig(target: String, subject: String) -> Vec<(usize, usize)> {
     onig::Regex::new(&target)
         .unwrap()
         .find_iter(&subject)
+        .map(|m| {
+            (
+                subject[m.0..m.1].to_string(),
+                find_char_index(&mut chars_iterator, m.0),
+            )
+        })
         .collect()
 }
 
-fn fancy(target: String, subject: String) -> Vec<(String, usize)> {
-    fancy_regex::RegexBuilder::new(&target)
-        .backtrack_limit(10_000_000)
+fn reg(target: String, subject: String) -> Vec<(String, usize)> {
+    let mut chars_iterator = subject.char_indices().enumerate();
+
+    regex::RegexBuilder::new(&target)
+        // .multi_line(true)
         .build()
         .unwrap()
-        .find_iter(&subject)
-        .map(|m| m.unwrap())
-        .map(|m| (m.as_str().to_string(), m.start() + 1))
+        .captures_iter(&subject)
+        .map(|m| m.get(1).unwrap())
+        .map(|m| {
+            (
+                m.as_str().trim().to_string(),
+                find_char_index(&mut chars_iterator, m.start()),
+            )
+        })
         .collect()
 }
 
-fn pcre2(target: String, subject: String) -> Vec<(String, usize)> {
-    RegexBuilder::new()
-        .build(&target)
+fn fancy(target: String, subject: String) -> Vec<(String, usize)> {
+    let mut chars_iterator = subject.char_indices().enumerate();
+
+    fancy_regex::RegexBuilder::new(&target)
+        .backtrack_limit(100_000_000)
+        .build()
         .unwrap()
-        .find_iter(subject.as_bytes())
-        .map(|m| m.unwrap())
+        .captures_iter(&subject)
+        .map(|m| m.unwrap().iter().nth(1).unwrap().unwrap())
         .map(|m| {
             (
-                std::str::from_utf8(m.as_bytes()).unwrap().to_string(),
-                m.start() + 1,
+                m.as_str().trim().to_string(),
+                find_char_index(&mut chars_iterator, m.start()),
             )
         })
         .collect()
@@ -37,8 +199,22 @@ fn pcre2(target: String, subject: String) -> Vec<(String, usize)> {
 #[magnus::init]
 fn init(ruby: &Ruby) -> Result<(), Error> {
     let module = ruby.define_module("Xre2")?;
+    ruby.define_class("Regexx", ruby.class_object())?;
+    let regexxlist = ruby.define_class("RegexxList", ruby.class_object())?;
+    regexxlist.define_singleton_method("new", function!(RegexList::new, 1))?;
+    module.define_singleton_method("regex", function!(regex, 1))?;
     module.define_singleton_method("fancy", function!(fancy, 2))?;
     module.define_singleton_method("onig", function!(onig, 2))?;
-    module.define_singleton_method("pcre2", function!(pcre2, 2))?;
+    module.define_singleton_method("reg", function!(reg, 2))?;
+    module.define_singleton_method("reg_array", function!(reg_array, 2))?;
+    module.define_singleton_method("reg_compiled", function!(reg_compiled, 2))?;
+    module.define_singleton_method(
+        "reg_compiled_with_context",
+        function!(reg_compiled_with_context, 2),
+    )?;
+    module.define_singleton_method(
+        "reg_array_with_context",
+        function!(reg_array_with_context, 2),
+    )?;
     Ok(())
 }

data/lib/xre2/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Xre2
-  VERSION = "0.1.0"
+  VERSION = "0.4.0"
 end