Packages
text_stemmer
0.1.0
Pre-compiled Snowball stemmers for Elixir. Ships 36 stemming algorithms covering a wide range of natural languages, accessible through a single `Text.Stemmer.stem/2` entry point.
Current section
Files
Jump to
Current section
Files
lib/text/stemmer/stemmers/danish.ex
# Generated by Snowball.Generator — do not edit by hand.
defmodule Text.Stemmer.Stemmers.Danish do
@moduledoc """
Snowball stemmer for danish.
Generated from the canonical Snowball algorithm source.
"""
alias Snowball.Runtime
alias Snowball.Grouping
# Groupings
@g_c Grouping.from_string("bcdfghjklmnpqrstvwxz")
@g_s_ending Grouping.from_string("abcdfghjklmnoprtvyzå")
@g_v Grouping.from_string("aeiouyæåø")
# Among tables
@a_0 [
{"ig", -1, 1, nil},
{"lig", 0, 1, nil},
{"elig", 1, 1, nil},
{"els", -1, 1, nil},
{"løst", -1, 2, nil}
]
@a_1 [
{"gd", -1, -1, nil},
{"dt", -1, -1, nil},
{"gt", -1, -1, nil},
{"kt", -1, -1, nil}
]
@a_2 [
{"hed", -1, 1, nil},
{"ethed", 0, 1, nil},
{"ered", -1, 1, nil},
{"e", -1, 1, nil},
{"erede", 3, 1, nil},
{"ende", 3, 1, nil},
{"erende", 5, 1, nil},
{"ene", 3, 1, nil},
{"erne", 3, 1, nil},
{"ere", 3, 1, nil},
{"en", -1, 1, nil},
{"heden", 10, 1, nil},
{"eren", 10, 1, nil},
{"er", -1, 1, nil},
{"heder", 13, 1, nil},
{"erer", 13, 1, nil},
{"s", -1, 2, nil},
{"heds", 16, 1, nil},
{"es", 16, 1, nil},
{"endes", 18, 1, nil},
{"erendes", 19, 1, nil},
{"enes", 18, 1, nil},
{"ernes", 18, 1, nil},
{"eres", 18, 1, nil},
{"ens", 16, 1, nil},
{"hedens", 24, 1, nil},
{"erens", 24, 1, nil},
{"ers", 16, 1, nil},
{"ets", 16, 1, nil},
{"erets", 28, 1, nil},
{"et", -1, 1, nil},
{"eret", 30, 1, nil}
]
@doc """
Stem a word.
### Arguments
* `word` is a UTF-8 binary.
### Returns
* The stemmed UTF-8 binary.
"""
@spec stem(binary()) :: binary()
def stem(word) when is_binary(word) do
state = Runtime.new(word) |> init_vars()
state = run_stem(state)
Runtime.assign_to(state)
end
defp init_vars(state), do: %{state | vars: %{p1: 0, x: 0, ch: ""}}
defp run_stem(%Runtime{} = state) do
{_, state} = r_stem(state)
state
end
# Snowball runtime helpers.
defp snowball_do_f(state, fun) do
saved_c = state.cursor
{_, s} = fun.(state)
%{s | cursor: saved_c}
end
defp snowball_do_b(state, fun) do
rel = state.limit - state.cursor
{_, s} = fun.(state)
%{s | cursor: s.limit - rel}
end
defp snowball_test_f(state, fun) do
saved_c = state.cursor
case fun.(state) do
{:ok, s} -> {:ok, %{s | cursor: saved_c}}
{:fail, s} -> {:fail, %{s | cursor: saved_c}}
end
end
defp snowball_test_b(state, fun) do
rel = state.limit - state.cursor
case fun.(state) do
{:ok, s} -> {:ok, %{s | cursor: s.limit - rel}}
{:fail, s} -> {:fail, %{s | cursor: s.limit - rel}}
end
end
defp snowball_try(state, fun) do
rel = state.limit - state.cursor
case fun.(state) do
{:ok, s} -> s
{:fail, s} -> %{s | cursor: s.limit - rel}
end
end
defp lift(state, :fail), do: {:fail, state}
defp lift(_state, %Runtime{} = s), do: {:ok, s}
defp next_codepoint(%Runtime{cursor: c, limit: lim, current: cur} = state) do
case Runtime.codepoint_at(cur, c, lim) do
{_cp, size} -> {:ok, %{state | cursor: c + size}}
:error -> {:fail, state}
end
end
defp next_codepoint_b(%Runtime{cursor: c, limit_backward: lb, current: cur} = state) do
case Runtime.codepoint_before(cur, c, lb) do
{_cp, size} -> {:ok, %{state | cursor: c - size}}
:error -> {:fail, state}
end
end
defp r_undouble(%Runtime{} = state) do
case (fn state ->
(fn ->
case (fn state ->
(fn ->
target = state.vars[:p1]
if target < state.limit_backward or target > state.limit do
{:fail, state}
else
{:ok, %{state | cursor: target}}
end
end).()
end).(state) do
{:fail, _} -> {:fail, state}
{:ok, limit_state} ->
old_lb = state.limit_backward
state = %{state | limit_backward: limit_state.cursor}
case (fn state ->
case (fn state ->
{:ok, %{state | ket: state.cursor}}
end).(state) do
{:ok, state} ->
case (fn state ->
lift(state, Runtime.in_grouping_b(state, @g_c))
end).(state) do
{:ok, state} ->
case (fn state ->
{:ok, %{state | bra: state.cursor}}
end).(state) do
{:ok, state} ->
{:ok, put_in(state.vars[:ch], Runtime.slice_to(state))}
end
r -> r
end
end
end).(state) do
{:ok, s} -> {:ok, %{s | limit_backward: old_lb}}
{:fail, s} -> {:fail, %{s | limit_backward: old_lb}}
end
end
end).()
end).(state) do
{:ok, state} ->
case (fn state ->
lift(state, Runtime.eq_s_b(state, state.vars[:ch]))
end).(state) do
{:ok, state} ->
{:ok, Runtime.slice_del(state)}
r -> r
end
r -> r
end
end
defp r_other_suffix(%Runtime{} = state) do
case (fn state ->
{:ok, snowball_do_b(state, fn state ->
case (fn state ->
{:ok, %{state | ket: state.cursor}}
end).(state) do
{:ok, state} ->
case (fn state ->
lift(state, Runtime.eq_s_b(state, "st"))
end).(state) do
{:ok, state} ->
case (fn state ->
{:ok, %{state | bra: state.cursor}}
end).(state) do
{:ok, state} ->
case (fn state ->
lift(state, Runtime.eq_s_b(state, "ig"))
end).(state) do
{:ok, state} ->
{:ok, Runtime.slice_del(state)}
r -> r
end
end
r -> r
end
end
end)}
end).(state) do
{:ok, state} ->
(fn ->
old_limit = state.limit_backward
case (fn state ->
(fn ->
target = state.vars[:p1]
if target < state.limit_backward or target > state.limit do
{:fail, state}
else
{:ok, %{state | cursor: target}}
end
end).()
end).(state) do
{:fail, _} -> {:fail, state}
{:ok, limit_state} ->
state = %{state | ket: state.cursor, limit_backward: limit_state.cursor}
case Runtime.find_among_b(state, @a_0) do
:fail -> {:fail, %{state | limit_backward: old_limit}}
{%Runtime{} = s, result} ->
state = %{s | bra: s.cursor, limit_backward: old_limit}
case result do
1 ->
case (fn state ->
{:ok, Runtime.slice_del(state)}
end).(state) do
{:ok, state} ->
{:ok, snowball_do_b(state, fn state ->
r_consonant_pair(state)
end)}
end
2 ->
{:ok, Runtime.slice_from(state, "løs")}
_ -> {:ok, state}
end
end
end
end).()
end
end
defp r_consonant_pair(%Runtime{} = state) do
case (fn state ->
snowball_test_b(state, fn state ->
(fn ->
old_limit = state.limit_backward
case (fn state ->
(fn ->
target = state.vars[:p1]
if target < state.limit_backward or target > state.limit do
{:fail, state}
else
{:ok, %{state | cursor: target}}
end
end).()
end).(state) do
{:fail, _} -> {:fail, state}
{:ok, limit_state} ->
state = %{state | ket: state.cursor, limit_backward: limit_state.cursor}
case Runtime.find_among_b(state, @a_1) do
:fail -> {:fail, %{state | limit_backward: old_limit}}
{%Runtime{} = s, _result} ->
state = %{s | bra: s.cursor, limit_backward: old_limit}
{:ok, state}
end
end
end).()
end)
end).(state) do
{:ok, state} ->
case (fn state ->
next_codepoint_b(state)
end).(state) do
{:ok, state} ->
case (fn state ->
{:ok, %{state | bra: state.cursor}}
end).(state) do
{:ok, state} ->
{:ok, Runtime.slice_del(state)}
end
r -> r
end
r -> r
end
end
defp r_main_suffix(%Runtime{} = state) do
(fn ->
old_limit = state.limit_backward
case (fn state ->
(fn ->
target = state.vars[:p1]
if target < state.limit_backward or target > state.limit do
{:fail, state}
else
{:ok, %{state | cursor: target}}
end
end).()
end).(state) do
{:fail, _} -> {:fail, state}
{:ok, limit_state} ->
state = %{state | ket: state.cursor, limit_backward: limit_state.cursor}
case Runtime.find_among_b(state, @a_2) do
:fail -> {:fail, %{state | limit_backward: old_limit}}
{%Runtime{} = s, result} ->
state = %{s | bra: s.cursor, limit_backward: old_limit}
case result do
1 ->
{:ok, Runtime.slice_del(state)}
2 ->
case (fn state ->
lift(state, Runtime.in_grouping_b(state, @g_s_ending))
end).(state) do
{:ok, state} ->
{:ok, Runtime.slice_del(state)}
r -> r
end
_ -> {:ok, state}
end
end
end
end).()
end
defp r_mark_regions(%Runtime{} = state) do
case (fn state ->
{:ok, put_in(state.vars[:p1], state.limit)}
end).(state) do
{:ok, state} ->
case (fn state ->
snowball_test_f(state, fn state ->
case (fn state ->
(fn ->
n = 3
Enum.reduce_while(1..max(n, 0)//1, {:ok, state}, fn _, {:ok, s} ->
case next_codepoint(s) do
{:ok, s2} -> {:cont, {:ok, s2}}
{:fail, _} -> {:halt, {:fail, state}}
end
end)
end).()
end).(state) do
{:ok, state} ->
{:ok, put_in(state.vars[:x], state.cursor)}
r -> r
end
end)
end).(state) do
{:ok, state} ->
case (fn state ->
case Runtime.go_out_grouping(state, @g_v) do
:fail -> {:fail, state}
%Runtime{} = s -> next_codepoint(s)
end
end).(state) do
{:ok, state} ->
case (fn state ->
case Runtime.go_in_grouping(state, @g_v) do
:fail -> {:fail, state}
%Runtime{} = s -> next_codepoint(s)
end
end).(state) do
{:ok, state} ->
case (fn state ->
{:ok, put_in(state.vars[:p1], state.cursor)}
end).(state) do
{:ok, state} ->
{:ok, snowball_try(state, fn state ->
case (fn state ->
if state.vars[:p1] < state.vars[:x], do: {:ok, state}, else: {:fail, state}
end).(state) do
{:ok, state} ->
{:ok, put_in(state.vars[:p1], state.vars[:x])}
r -> r
end
end)}
end
r -> r
end
r -> r
end
r -> r
end
end
end
defp r_stem(%Runtime{} = state) do
case (fn state ->
{:ok, snowball_do_f(state, fn state ->
r_mark_regions(state)
end)}
end).(state) do
{:ok, state} ->
(fn state ->
old_cursor = state.cursor
old_lb = state.limit_backward
state = %{state | cursor: state.limit, limit_backward: old_cursor}
{tag, s} = (fn state ->
case (fn state ->
{:ok, snowball_do_b(state, fn state ->
r_main_suffix(state)
end)}
end).(state) do
{:ok, state} ->
case (fn state ->
{:ok, snowball_do_b(state, fn state ->
r_consonant_pair(state)
end)}
end).(state) do
{:ok, state} ->
case (fn state ->
{:ok, snowball_do_b(state, fn state ->
r_other_suffix(state)
end)}
end).(state) do
{:ok, state} ->
{:ok, snowball_do_b(state, fn state ->
r_undouble(state)
end)}
end
end
end
end).(state)
{tag, %{s | cursor: old_cursor, limit_backward: old_lb}}
end).(state)
end
end
end