rscsv 0.1.1 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 284dad5b774af98eb63d7de4e33f20fe3891fffe
-  data.tar.gz: 8511d9bd4720ff92cf889e3604de0a835ca7d49d
+  metadata.gz: 33297a5689b1597d449a3c0779895ce85b7c764c
+  data.tar.gz: 5c1ae9a22753cc171b73044bec4c11ff8cce924a
 SHA512:
-  metadata.gz: 25d4d11871615a9be86fc32a1f51b309723ea0c95c24d1f02e02c5a6e427b0f5412db68a2585d23a1404e9727df0ece1a8f7772299a042abb78db71926c75fc1
-  data.tar.gz: 8efdbe79abd543ef6128ba868f3ef21c039200a75becddba813968d6c2ee1eb9fbf4fee2593efdd51964cdb4d38dfb97120d0dc8afc90709347c8c90d6bee8b0
+  metadata.gz: f4b8863ca0859d40012f2d595ef7f75ab7405c9d0361ba02776b577814e23a4fc051983143f2b51ec812b29b46ac51c049efe96ae57a8ddc5fad3390a54925e2
+  data.tar.gz: be1e68c1f448be078683b2e29762802ce5d8cbff3f2916b952e002156eac3d15c483df93969b522d85fbd7b5673e0e9ab984b45d8ac47e5fb65456ee03b5671f

data/Cargo.lock

@@ -2,15 +2,10 @@
 name = "rscsv"
 version = "0.1.0"
 dependencies = [
- "csv 0.15.0 (registry+https://github.com/rust-lang/crates.io-index)",
- "helix 0.6.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "csv 1.0.0-beta.1 (registry+https://github.com/rust-lang/crates.io-index)",
+ "helix 0.6.1 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
-[[package]]
-name = "byteorder"
-version = "1.0.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-
 [[package]]
 name = "cslice"
 version = "0.3.0"
@@ -23,23 +18,30 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 
 [[package]]
 name = "csv"
-version = "0.15.0"
+version = "1.0.0-beta.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "csv-core 0.1.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "serde 1.0.7 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
+[[package]]
+name = "csv-core"
+version = "0.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 dependencies = [
- "byteorder 1.0.0 (registry+https://github.com/rust-lang/crates.io-index)",
  "memchr 1.0.1 (registry+https://github.com/rust-lang/crates.io-index)",
- "rustc-serialize 0.3.24 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
 [[package]]
 name = "helix"
-version = "0.6.0"
+version = "0.6.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 dependencies = [
  "cslice 0.3.0 (registry+https://github.com/rust-lang/crates.io-index)",
  "cstr-macro 0.1.0 (registry+https://github.com/rust-lang/crates.io-index)",
  "libc 0.2.22 (registry+https://github.com/rust-lang/crates.io-index)",
- "libcruby-sys 0.6.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "libcruby-sys 0.6.1 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
 [[package]]
@@ -49,7 +51,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 
 [[package]]
 name = "libcruby-sys"
-version = "0.6.0"
+version = "0.6.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 dependencies = [
  "libc 0.2.22 (registry+https://github.com/rust-lang/crates.io-index)",
@@ -64,17 +66,17 @@ dependencies = [
 ]
 
 [[package]]
-name = "rustc-serialize"
-version = "0.3.24"
+name = "serde"
+version = "1.0.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 
 [metadata]
-"checksum byteorder 1.0.0 (registry+https://github.com/rust-lang/crates.io-index)" = "c40977b0ee6b9885c9013cd41d9feffdd22deb3bb4dc3a71d901cc7a77de18c8"
 "checksum cslice 0.3.0 (registry+https://github.com/rust-lang/crates.io-index)" = "0f8cb7306107e4b10e64994de6d3274bd08996a7c1322a27b86482392f96be0a"
 "checksum cstr-macro 0.1.0 (registry+https://github.com/rust-lang/crates.io-index)" = "db53fddba18cdd35477a7213a3ef6acfbfa333c31b42ce019e544c4a1420a06f"
-"checksum csv 0.15.0 (registry+https://github.com/rust-lang/crates.io-index)" = "7ef22b37c7a51c564a365892c012dc0271221fdcc64c69b19ba4d6fa8bd96d9c"
-"checksum helix 0.6.0 (registry+https://github.com/rust-lang/crates.io-index)" = "4b3d0b8579a68e79a629d9c2f4717dbf5e2b467a7d7afb38c737d573d46331ac"
+"checksum csv 1.0.0-beta.1 (registry+https://github.com/rust-lang/crates.io-index)" = "81675dd89651e2aa0989e6a5249dc5c5bcfc13772baec7f9652208e7691e0955"
+"checksum csv-core 0.1.0 (registry+https://github.com/rust-lang/crates.io-index)" = "6a11ab3094dd197341f9d66753789a5cdf29ce35450a7d6e7968024e2d44519c"
+"checksum helix 0.6.1 (registry+https://github.com/rust-lang/crates.io-index)" = "b97612e592273062bb127433a22b201c9b58375c95fc7ace355ee2fc6be61ebe"
 "checksum libc 0.2.22 (registry+https://github.com/rust-lang/crates.io-index)" = "babb8281da88cba992fa1f4ddec7d63ed96280a1a53ec9b919fd37b53d71e502"
-"checksum libcruby-sys 0.6.0 (registry+https://github.com/rust-lang/crates.io-index)" = "3bf276325da1f1b20cd68108b33c02dfcad26ff4f4d7fdac4cd2a5aac8f04ae5"
+"checksum libcruby-sys 0.6.1 (registry+https://github.com/rust-lang/crates.io-index)" = "d9ae575f064157bc55bb543589971959e3a0163fd49be03fef4c1f3440146114"
 "checksum memchr 1.0.1 (registry+https://github.com/rust-lang/crates.io-index)" = "1dbccc0e46f1ea47b9f17e6d67c5a96bd27030519c519c9c91327e31275a47b4"
-"checksum rustc-serialize 0.3.24 (registry+https://github.com/rust-lang/crates.io-index)" = "dcf128d1287d2ea9d80910b5f1120d0b8eede3fbf1abe91c40d39ea7d51e6fda"
+"checksum serde 1.0.7 (registry+https://github.com/rust-lang/crates.io-index)" = "c0c3d79316a6051231925504f6ef893d45088e8823c77a8331a3dcf427ee9087"

data/Cargo.toml

@@ -8,5 +8,5 @@ authors = ["Ville Lautanala <lautis@gmail.com>"]
 crate-type = ["cdylib"]
 
 [dependencies]
-helix = "0.6.0"
-csv = "0.15.0"
+helix = "0.6.1"
+csv = "1.0.0-beta.1"

data/README.md

@@ -1,6 +1,7 @@
 # Rscsv
 
-Fast CSV using Rust extensions. Currently writing CSV is implemented.
+Fast CSV using Rust extensions. Can read arrays of arrays from strings and write
+strings from arrays of arrays.
 
 [![Build Status](https://travis-ci.org/lautis/rscsv.svg?branch=master)](https://travis-ci.org/lautis/rscsv)
 
@@ -22,6 +23,9 @@ Rscsv::Writer.generate_lines([['1', '2', '3'], ['3', '4', '5']])
 # => 1,2,3\n4,5,6\n
 Rscsv::Writer.generate_line(['1', '2', '3'])
 # => 1,2,3\n
+
+Rscsv::Reader.parse("1,2,3\n4,5,6")
+# => [["1", "2", "3"], ["4", "5", "6"]]
 ```
 
-This is 3x faster than using native Ruby `CSV.generate`.
+This is ~3x faster than using native Ruby `CSV.generate` or `CSV.parse`.

data/bin/benchmark

@@ -11,6 +11,10 @@ rows = (0...1000).map do
   (0...10).map { SecureRandom.hex }
 end
 
+csv_string = CSV.generate do |csv|
+  rows.each { |row| csv << row }
+end
+
 Benchmark.ips do |x|
   x.report('Ruby CSV') do |times|
     times.times do
@@ -26,3 +30,17 @@ Benchmark.ips do |x|
 
   x.compare!
 end
+
+Benchmark.ips do |x|
+  x.report('Ruby CSV') do |times|
+    times.times do
+      CSV.parse(csv_string)
+    end
+  end
+
+  x.report('rscsv') do |times|
+    times.times { Rscsv::Reader.parse(csv_string) }
+  end
+
+  x.compare!
+end

data/lib/rscsv/version.rb

@@ -1,3 +1,3 @@
 module Rscsv
-  VERSION = '0.1.1'.freeze
+  VERSION = '0.2.0'.freeze
 end

data/lib/rscsv.rb

@@ -3,5 +3,6 @@ require 'rscsv/native'
 require 'rscsv/version'
 
 module Rscsv
+  Reader = RscsvReader
   Writer = RscsvWriter
 end

data/rscsv.gemspec

@@ -10,8 +10,7 @@ Gem::Specification.new do |spec|
   spec.email = ['lautis@gmail.com']
 
   spec.summary = 'Rust-powered CSV'
-  spec.description = 'Fast CSV using Rust extensions.' \
-    'Currently only writes CSV.'
+  spec.description = 'Fast CSV using Rust extensions.'
   spec.homepage = 'https://github.com/lautis/rscsv'
   spec.license = 'MIT'
 
@@ -23,7 +22,7 @@ Gem::Specification.new do |spec|
   spec.require_paths = ['lib']
   spec.extensions = Dir['ext/extconf.rb']
 
-  spec.add_dependency 'helix_runtime', '~> 0.6.0'
+  spec.add_dependency 'helix_runtime', '0.6.1'
   spec.add_dependency 'rake', '>= 10.0'
   spec.add_development_dependency 'bundler', '~> 1.14'
   spec.add_development_dependency 'rspec', '~> 3.0'

data/src/lib.rs

@@ -2,9 +2,10 @@
 extern crate helix;
 extern crate csv;
 
+use std::error::Error;
 use helix::sys;
 use helix::sys::VALUE;
-use helix::{UncheckedValue, CheckResult, CheckedValue, ToRust};
+use helix::{UncheckedValue, CheckResult, CheckedValue, ToRust, ToRuby};
 
 struct VecWrap<T>(Vec<T>);
 
@@ -64,23 +65,74 @@ impl ToRust<VecWrap<VecWrap<String>>> for CheckedValue<VecWrap<VecWrap<String>>>
     }
 }
 
+#[cfg_attr(windows, link(name="helix-runtime"))]
+extern "C" {
+    pub fn rb_ary_new_capa(capa: isize) -> VALUE;
+    pub fn rb_ary_entry(ary: VALUE, offset: isize) -> VALUE;
+    pub fn rb_ary_push(ary: VALUE, item: VALUE) -> VALUE;
+}
+
+impl ToRuby for VecWrap<csv::StringRecord> {
+    fn to_ruby(self) -> VALUE {
+        let ary = unsafe { rb_ary_new_capa(self.0.len() as isize) };
+        for row in self.0 {
+            let inner_array = unsafe { rb_ary_new_capa(row.len() as isize) };
+            for column in row.iter() {
+                unsafe {
+                    rb_ary_push(inner_array, column.to_ruby());
+                }
+            }
+            unsafe {
+                rb_ary_push(ary, inner_array);
+            }
+        }
+        ary
+    }
+}
+
+fn generate_lines(rows: VecWrap<VecWrap<String>>) -> Result<String, Box<Error>> {
+    let mut wtr = csv::WriterBuilder::new().from_writer(vec![]);
+    for row in rows.0 {
+        wtr.write_record(&(row.0))?;
+    }
+
+    return Ok(String::from_utf8(wtr.into_inner()?)?);
+}
+
+fn parse_csv(data: String) -> Result<Vec<csv::StringRecord>, csv::Error> {
+    let mut reader = csv::ReaderBuilder::new()
+        .has_headers(false)
+        .from_reader(data.as_bytes());
+    let records = reader
+        .records()
+        .collect::<Result<Vec<csv::StringRecord>, csv::Error>>();
+    return records;
+}
+
 ruby! {
+    class RscsvReader {
+        def parse(data: String) -> VecWrap<csv::StringRecord> {
+            match parse_csv(data) {
+                Err(_) => throw!("Error parsing CSV"),
+                Ok(result) => return VecWrap(result)
+            };
+        }
+    }
     class RscsvWriter {
         def generate_line(row: VecWrap<String>) -> String {
-            let mut writer = csv::Writer::from_memory();
-            writer.write(row.0.into_iter()).unwrap();
-            return writer.into_string();
+            let mut wtr = csv::WriterBuilder::new().from_writer(vec![]);
+            let result = wtr.write_record(&(row.0));
+            match result {
+                Err(_) => throw!("Error generating csv"),
+                Ok(_) => return String::from_utf8(wtr.into_inner().unwrap()).unwrap(),
+            };
         }
 
         def generate_lines(rows: VecWrap<VecWrap<String>>) -> String {
-            let mut writer = csv::Writer::from_memory();
-            let vec = rows.0;
-
-            for row in vec {
-                writer.write(row.0.into_iter()).unwrap();
+            match generate_lines(rows) {
+                Err(_) => throw!("Error generating csv"),
+                Ok(csv) => csv
             }
-
-            return writer.into_string();
         }
     }
 }

metadata

@@ -1,29 +1,29 @@
 --- !ruby/object:Gem::Specification
 name: rscsv
 version: !ruby/object:Gem::Version
-  version: 0.1.1
+  version: 0.2.0
 platform: ruby
 authors:
 - Ville Lautanala
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2017-05-17 00:00:00.000000000 Z
+date: 2017-05-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: helix_runtime
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - '='
       - !ruby/object:Gem::Version
-        version: 0.6.0
+        version: 0.6.1
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - '='
       - !ruby/object:Gem::Version
-        version: 0.6.0
+        version: 0.6.1
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
@@ -80,7 +80,7 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '2.7'
-description: Fast CSV using Rust extensions.Currently only writes CSV.
+description: Fast CSV using Rust extensions.
 email:
 - lautis@gmail.com
 executables: []