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/czech.ex
# Generated by Snowball.Generator — do not edit by hand.
defmodule Text.Stemmer.Stemmers.Czech do
@moduledoc """
Snowball stemmer for czech.
Generated from the canonical Snowball algorithm source.
"""
alias Snowball.Runtime
alias Snowball.Grouping
# Groupings
@g_env_ending Grouping.from_string("bcčdhkprsštvzž")
@g_ev_ending Grouping.from_string("hknrtz")
@g_v Grouping.from_string("aeiouyáěéíóúůý")
@g_v_or_syllabic_c Grouping.from_string("aeiouyáěéíóúůýlr")
# Among tables
@a_0 [
{"", -1, 2, nil},
{"l", 0, 1, nil},
{"tl", 1, 2, nil},
{"s", 0, 1, nil},
{"es", 3, 2, nil},
{"č", 0, 1, nil},
{"eč", 5, 2, nil},
{"ř", 0, 1, nil},
{"ž", 0, 1, nil}
]
@a_1 [
{"obl", -1, -1, nil},
{"sn", -1, -1, nil},
{"dot", -1, -1, nil}
]
@a_2 [
{"uc", -1, -1, nil},
{"h", -1, -1, nil},
{"ok", -1, -1, nil},
{"kar", -1, -1, nil},
{"č", -1, -1, nil}
]
@a_3 [
{"a", -1, 1, nil},
{"ama", 0, 1, nil},
{"ata", 0, 1, nil},
{"eb", -1, 5, nil},
{"ec", -1, 6, nil},
{"e", -1, 3, nil},
{"ete", 5, 4, nil},
{"ěte", 5, 2, nil},
{"ech", -1, 3, nil},
{"atech", 8, 1, nil},
{"ách", -1, 1, nil},
{"ích", -1, 13, nil},
{"ých", -1, 1, nil},
{"i", -1, 13, nil},
{"mi", 13, 1, nil},
{"ami", 14, 1, nil},
{"emi", 14, 3, nil},
{"ěmi", 14, 2, nil},
{"ťmi", 14, 12, nil},
{"ími", 14, 13, nil},
{"ými", 14, 1, nil},
{"eti", 13, 4, nil},
{"ěti", 13, 2, nil},
{"ovi", 13, 1, nil},
{"ek", -1, 7, nil},
{"ěk", -1, 8, nil},
{"em", -1, 3, nil},
{"etem", 26, 4, nil},
{"ětem", 26, 2, nil},
{"ěm", -1, 2, nil},
{"ám", -1, 1, nil},
{"ém", -1, 1, nil},
{"ím", -1, 13, nil},
{"ům", -1, 1, nil},
{"atům", 33, 1, nil},
{"ým", -1, 1, nil},
{"o", -1, 1, nil},
{"ého", 36, 1, nil},
{"ího", 36, 13, nil},
{"us", -1, 1, nil},
{"at", -1, 1, nil},
{"et", -1, 10, nil},
{"u", -1, 1, nil},
{"ému", 42, 1, nil},
{"ímu", 42, 13, nil},
{"ou", 42, 1, nil},
{"ev", -1, 11, nil},
{"y", -1, 1, nil},
{"aty", 47, 1, nil},
{"eň", -1, 9, nil},
{"ě", -1, 2, nil},
{"á", -1, 1, nil},
{"ť", -1, 12, nil},
{"é", -1, 1, nil},
{"ové", 53, 1, nil},
{"í", -1, 13, nil},
{"ů", -1, 1, nil},
{"ý", -1, 1, nil}
]
@a_4 [
{"in", -1, 2, nil},
{"ov", -1, 1, nil},
{"ův", -1, 1, nil}
]
@a_5 [
{"c", -1, 1, nil},
{"nc", 0, -1, nil},
{"ínc", 1, 2, nil},
{"avc", 0, -1, nil},
{"ovc", 0, -1, nil},
{"čt", -1, 3, nil},
{"št", -1, 4, nil},
{"dešt", 6, -1, nil},
{"lešt", 6, -1, nil},
{"išt", 6, -1, nil},
{"poušt", 6, -1, nil},
{"ášt", 6, -1, nil},
{"íšt", 6, -1, nil}
]
@a_6 [
{"c", -1, 1, nil},
{"nc", 0, -1, nil},
{"ínc", 1, 2, nil},
{"avc", 0, -1, nil},
{"ovc", 0, -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}}
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 snowball_or(state, fun1, fun2) do
rel = state.limit - state.cursor
case fun1.(state) do
{:ok, s} -> {:ok, s}
{:fail, s} -> fun2.(%{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 r_case_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_3) 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 ->
{:ok, Runtime.slice_del(state)}
3 ->
case (fn state ->
{:ok, Runtime.slice_del(state)}
end).(state) do
{:ok, state} ->
{:ok, snowball_try(state, fn state ->
r_palatalise_e(state)
end)}
end
4 ->
(fn ->
case Runtime.find_among_b(state, @a_0) do
:fail -> {:fail, state}
{%Runtime{} = s, result} ->
state = s
case result do
1 ->
{:ok, Runtime.slice_del(state)}
2 ->
{:ok, Runtime.slice_from(state, "et")}
_ -> {:ok, s}
end
end
end).()
5 ->
case (fn state ->
snowball_test_b(state, fn state ->
lift(state, Runtime.out_grouping_b(state, @g_v))
end)
end).(state) do
{:ok, state} ->
case (fn state ->
(fn state ->
saved_c = state.cursor
case (fn state ->
lift(state, Runtime.eq_s_b(state, "tř"))
end).(state) do
{:ok, s} -> {:fail, %{s | cursor: saved_c}}
{:fail, s} -> {:ok, %{s | cursor: saved_c}}
end
end).(state)
end).(state) do
{:ok, state} ->
{:ok, Runtime.slice_from(state, "b")}
r -> r
end
r -> r
end
6 ->
case (fn state ->
snowball_test_b(state, fn state ->
lift(state, Runtime.out_grouping_b(state, @g_v))
end)
end).(state) do
{:ok, state} ->
case (fn state ->
{:ok, Runtime.slice_del(state)}
end).(state) do
{:ok, state} ->
case (fn state ->
(fn ->
saved_c = state.cursor
new_state = Runtime.insert(state, saved_c, saved_c, "c")
{:ok, %{new_state | cursor: saved_c + byte_size("c")}}
end).()
end).(state) do
{:ok, state} ->
{:ok, snowball_try(state, fn state ->
r_palatalise_e(state)
end)}
end
end
r -> r
end
7 ->
case (fn state ->
snowball_test_b(state, fn state ->
lift(state, Runtime.out_grouping_b(state, @g_v))
end)
end).(state) do
{:ok, state} ->
case (fn state ->
(fn state ->
saved_c = state.cursor
case (fn state ->
(fn ->
case Runtime.find_among_b(state, @a_1) do
:fail -> {:fail, state}
{%Runtime{} = s, _} -> {:ok, s}
end
end).()
end).(state) do
{:ok, s} -> {:fail, %{s | cursor: saved_c}}
{:fail, s} -> {:ok, %{s | cursor: saved_c}}
end
end).(state)
end).(state) do
{:ok, state} ->
{:ok, Runtime.slice_from(state, "k")}
r -> r
end
r -> r
end
8 ->
case (fn state ->
lift(state, Runtime.eq_s_b(state, "n"))
end).(state) do
{:ok, state} ->
case (fn state ->
{:ok, %{state | bra: state.cursor}}
end).(state) do
{:ok, state} ->
{:ok, Runtime.slice_from(state, "ňk")}
end
r -> r
end
9 ->
case (fn state ->
snowball_test_b(state, fn state ->
lift(state, Runtime.in_grouping_b(state, @g_env_ending))
end)
end).(state) do
{:ok, state} ->
{:ok, Runtime.slice_from(state, "n")}
r -> r
end
10 ->
case (fn state ->
(fn ->
case Runtime.find_among_b(state, @a_2) do
:fail -> {:fail, state}
{%Runtime{} = s, _} -> {:ok, s}
end
end).()
end).(state) do
{:ok, state} ->
{:ok, Runtime.slice_from(state, "t")}
r -> r
end
11 ->
case (fn state ->
lift(state, Runtime.in_grouping_b(state, @g_ev_ending))
end).(state) do
{:ok, state} ->
{:ok, Runtime.slice_from(state, "v")}
r -> r
end
12 ->
{:ok, Runtime.slice_from(state, "t")}
13 ->
case (fn state ->
{:ok, Runtime.slice_del(state)}
end).(state) do
{:ok, state} ->
{:ok, snowball_try(state, fn state ->
r_palatalise_i(state)
end)}
end
_ -> {:ok, state}
end
end
end
end).()
end
defp r_possessive_suffix(%Runtime{} = state) do
(fn ->
state = %{state | ket: state.cursor}
case Runtime.find_among_b(state, @a_4) do
:fail -> {:fail, state}
{%Runtime{} = s, result} ->
state = %{s | bra: s.cursor}
case (fn state ->
r_R1(state)
end).(state) do
{:ok, state} ->
case result do
1 ->
{:ok, Runtime.slice_del(state)}
2 ->
case (fn state ->
{:ok, Runtime.slice_del(state)}
end).(state) do
{:ok, state} ->
{:ok, snowball_try(state, fn state ->
r_palatalise_i(state)
end)}
end
_ -> {:ok, state}
end
r -> r
end
end
end).()
end
defp r_palatalise_i(%Runtime{} = state) do
(fn ->
state = %{state | ket: state.cursor}
case Runtime.find_among_b(state, @a_5) do
:fail -> {:fail, state}
{%Runtime{} = s, result} ->
state = %{s | bra: s.cursor}
case result do
1 ->
{:ok, Runtime.slice_from(state, "k")}
2 ->
{:ok, Runtime.slice_from(state, "ínk")}
3 ->
{:ok, Runtime.slice_from(state, "ck")}
4 ->
{:ok, Runtime.slice_from(state, "sk")}
_ -> {:ok, state}
end
end
end).()
end
defp r_palatalise_e(%Runtime{} = state) do
(fn ->
state = %{state | ket: state.cursor}
case Runtime.find_among_b(state, @a_6) do
:fail -> {:fail, state}
{%Runtime{} = s, result} ->
state = %{s | bra: s.cursor}
case result do
1 ->
{:ok, Runtime.slice_from(state, "k")}
2 ->
{:ok, Runtime.slice_from(state, "ínk")}
_ -> {:ok, state}
end
end
end).()
end
defp r_R1(%Runtime{} = state) do
if state.vars[:p1] <= state.cursor, do: {:ok, state}, else: {:fail, state}
end
defp r_mark_regions(%Runtime{} = state) do
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 ->
{:ok, put_in(state.vars[:p1], state.limit)}
end).(state) do
{:ok, state} ->
{:ok, snowball_do_f(state, fn state ->
case (fn state ->
snowball_or(state,
fn state ->
lift(state, Runtime.in_grouping(state, @g_v))
end,
fn state ->
case (fn state ->
next_codepoint(state)
end).(state) do
{:ok, state} ->
case Runtime.go_out_grouping(state, @g_v_or_syllabic_c) do
:fail -> {:fail, state}
%Runtime{} = s -> next_codepoint(s)
end
r -> r
end
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
end)}
end
r -> r
end
end
defp r_stem(%Runtime{} = state) do
case (fn state ->
r_mark_regions(state)
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_case_suffix(state)
end)}
end).(state) do
{:ok, state} ->
{:ok, snowball_do_b(state, fn state ->
r_possessive_suffix(state)
end)}
end
end).(state)
{tag, %{s | cursor: old_cursor, limit_backward: old_lb}}
end).(state)
r -> r
end
end
end