"""classify_lang_script

Revision ID: e5e21700527d
Revises: cf8b451a9700
Create Date: 2020-06-25 11:32:38.491834

"""

# revision identifiers, used by Alembic.
revision = 'e5e21700527d'
down_revision = 'cf8b451a9700'
branch_labels = None
depends_on = None

from alembic import op
import sqlalchemy as sa


def upgrade():
    op.execute('''
create or replace function classify_lang_script(words text) returns integer
  strict
  language sql
as
$$
select
  case
    when words ~* '[\u0400-\u04FF]'  then 1 -- cyrillic
    when words ~* '[\u0600-\u06FF]' then 2 -- arabic
    when words ~* '[\u4E00-\u9FFF]' then 3 -- cjk
    when words ~* '[\u0E00-\u0E7F]' then 4 -- thai
    else 0
  end;
$$;
    ''')


def downgrade():
    op.execute(''' drop function classify_lang_script(words text) ''')