Packages

paasaa

0.1.1
1.0.0 0.6.0 0.5.1 0.5.0 0.4.0 0.3.1 0.3.0 0.2.2 0.2.1 0.2.0 0.1.1 0.1.0

Natural language detection

HexDocs

Current section

12 Versions

Jump to
Readme
12 Versions
1 Dependency
Dependants
Files
Activity
Hide whitespace

Compare versions

4 files changed
+31 additions
-26 deletions
  @@ -1,5 +1,10 @@
1 1 # Paasaa
2 2
3 + [![Travis](https://img.shields.io/travis/minibikini/paasaa.svg?maxAge=2592000)](https://travis-ci.org/minibikini/paasaa)
4 + [![Coverage Status](https://coveralls.io/repos/github/minibikini/paasaa/badge.svg?branch=master)](https://coveralls.io/github/minibikini/paasaa?branch=master)
5 + [![Hex.pm](https://img.shields.io/hexpm/v/paasaa.svg?maxAge=2592000)](https://hex.pm/packages/paasaa)
6 + [![Hex.pm](https://img.shields.io/hexpm/l/paasaa.svg?maxAge=2592000)](https://hex.pm/packages/paasaa)
7 +
3 8 Language detection for Elixir
4 9
5 10 [Api Documentation] | [Hex Package]
  @@ -53,4 +58,4 @@ iex> Paasaa.all("Detect this!")
53 58 [Franc]: https://github.com/wooorm/franc/
54 59 [Titus Wormer]: http://wooorm.com/
55 60 [mit]: LICENSE
56 - [Egor Kislitsyn]: https://github/minibikini
61 + [Egor Kislitsyn]: https://github.com/minibikini
  @@ -1,6 +1,6 @@
1 1 {<<"app">>,<<"paasaa">>}.
2 2 {<<"build_tools">>,[<<"mix">>]}.
3 - {<<"description">>,<<"Provides language detection functionss">>}.
3 + {<<"description">>,<<"Natural language detection">>}.
4 4 {<<"elixir">>,<<"~> 1.3">>}.
5 5 {<<"files">>,
6 6 [<<"lib/paasaa.ex">>,<<"priv/languages.json">>,<<"priv/scripts.json">>,
  @@ -14,4 +14,4 @@
14 14 {<<"name">>,<<"exjsx">>},
15 15 {<<"optional">>,false},
16 16 {<<"requirement">>,<<"~> 3.2">>}]]}.
17 - {<<"version">>,<<"0.1.0">>}.
17 + {<<"version">>,<<"0.1.1">>}.
  @@ -21,7 +21,7 @@ defmodule Paasaa do
21 21 langs = Enum.map(langs, fn {lang, trigrams} ->
22 22 trigrams = trigrams
23 23 |> String.split("|")
24 - |> Enum.with_index()
24 + |> Enum.with_index
25 25 |> Enum.into(%{})
26 26
27 27 {lang, trigrams}
  @@ -33,7 +33,7 @@ defmodule Paasaa do
33 33
34 34 @max_difference 300
35 35
36 - @type result :: [{language :: Strinsg.t, score :: number}]
36 + @type result :: [{language :: String.t, score :: number}]
37 37
38 38 @type options :: [
39 39 min_length: integer,
  @@ -50,7 +50,7 @@ defmodule Paasaa do
50 50 ]
51 51
52 52 @doc """
53 - Detects a language. Returns a string with ISO6393 language code (i.g. "eng").
53 + Detects a language. Returns a string with ISO6393 language code (e.g. "eng").
54 54
55 55 ## Parameters
56 56
  @@ -88,7 +88,7 @@ defmodule Paasaa do
88 88 def detect(str, options \\ @default_options) do
89 89 str
90 90 |> all(options)
91 - |> List.first()
91 + |> List.first
92 92 |> elem(0)
93 93 end
94 94
  @@ -139,7 +139,7 @@ defmodule Paasaa do
139 139 count == 0 -> und
140 140 Map.has_key?(@languages, script) ->
141 141 str
142 - |> get_clean_trigrams()
142 + |> get_clean_trigrams
143 143 |> get_distances(@languages[script], options)
144 144 |> normalize(str)
145 145
  @@ -158,13 +158,9 @@ defmodule Paasaa do
158 158 |> Enum.max_by(fn {_, count} -> count end)
159 159 end
160 160
161 - @spec get_occurrence(str :: String.t, re :: Regex.t, str_len :: number) :: number
161 + @spec get_occurrence(str :: String.t, re :: Regex.t, str_len :: non_neg_integer) :: float
162 162 defp get_occurrence(str, re, str_len) do
163 - if found = Regex.scan(re, str) do
164 - Enum.count(found) / str_len
165 - else
166 - 0
167 - end
163 + Enum.count(Regex.scan(re, str)) / str_len
168 164 end
169 165
170 166 @spec get_distances([String.t], Enumerable.t, options) :: result
  @@ -203,7 +199,7 @@ defmodule Paasaa do
203 199 @spec normalize(result, String.t) :: result
204 200 defp normalize([], _str), do: und
205 201 defp normalize(distances, str) do
206 - min = distances |> List.first() |> elem(1)
202 + min = distances |> List.first |> elem(1)
207 203 max = String.length(str) * @max_difference - min
208 204
209 205 Enum.map distances, fn({lang, dist}) ->
  @@ -215,12 +211,12 @@ defmodule Paasaa do
215 211
216 212 # trigram stuff
217 213
218 - @spec get_clean_trigrams(String.t) :: [String.t]
214 + @spec get_clean_trigrams(String.t) :: result
219 215 defp get_clean_trigrams(str) do
220 216 str
221 - |> clean()
222 - |> pad()
223 - |> n_grams()
217 + |> clean
218 + |> pad
219 + |> n_grams
224 220 |> Enum.reduce(%{}, fn(trigram, acc) ->
225 221 count = acc[trigram] && acc[trigram] + 1 || 1
226 222 Map.put(acc, trigram, count)
  @@ -235,8 +231,8 @@ defmodule Paasaa do
235 231 str
236 232 |> String.replace(expression_symbols, " ")
237 233 |> String.replace(~r/\s+/, " ")
238 - |> String.trim()
239 - |> String.downcase()
234 + |> String.trim
235 + |> String.downcase
240 236 end
241 237
242 238 defp pad(str), do: " #{str} "
  @@ -244,7 +240,7 @@ defmodule Paasaa do
244 240 @spec n_grams(str :: String.t, n :: number) :: [String.t]
245 241 defp n_grams(str, n \\ 3) do
246 242 str
247 - |> String.graphemes()
243 + |> String.graphemes
248 244 |> Enum.chunk(n, 1)
249 245 |> Enum.map(&Enum.join/1)
250 246 end
  @@ -4,11 +4,13 @@ defmodule Paasaa.Mixfile do
4 4 def project do
5 5 [
6 6 app: :paasaa,
7 - version: "0.1.0",
7 + version: "0.1.1",
8 8 elixir: "~> 1.3",
9 9 description: description(),
10 10 package: package(),
11 - deps: deps()
11 + deps: deps(),
12 + test_coverage: [tool: ExCoveralls],
13 + preferred_cli_env: ["coveralls": :test, "coveralls.detail": :test, "coveralls.post": :test, "coveralls.html": :test]
12 14 ]
13 15 end
14 16
  @@ -22,13 +24,15 @@ defmodule Paasaa.Mixfile do
22 24 {:ex_doc, "~> 0.12", only: :dev},
23 25 {:mix_test_watch, "~> 0.2", only: :dev},
24 26 {:benchfella, "~> 0.3.0", only: :dev},
25 - {:credo, "~> 0.4", only: [:dev, :test]}
27 + {:credo, "~> 0.4", only: [:dev, :test]},
28 + {:excoveralls, "~> 0.5", only: :test},
29 + {:dialyxir, "~> 0.3.5", only: [:dev]}
26 30 ]
27 31 end
28 32
29 33 defp description do
30 34 """
31 - Provides language detection functionss
35 + Natural language detection
32 36 """
33 37 end