"""add_hebrew_classify

Revision ID: d442e6552298
Revises: 249d4700d767
Create Date: 2020-07-17 11:10:19.248456

"""

# revision identifiers, used by Alembic.
revision = 'd442e6552298'
down_revision = '249d4700d767'
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
$$
select
  case
    when words ~* '[\u0400-\u04FF]'  then 1 -- cyrillic
    when words ~* '[\u0600-\u06FF]' then 2 -- arabic
    when words ~* '[\u4E00-\u9FFF]' then 3 -- cjk
    when words ~* '[\u0E00-\u0E7F]' then 4 -- thai
    when words ~* '[\u0590-\u05FF]' then 5 -- hebrew
    else 0
  end;
$$;
    ''')


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