kudzues/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-04-05 19:21:05 +08:00
Code Issues Packages Projects Releases Wiki Activity

Feat/support parent child chunk (#12092)

Browse Source
This commit is contained in:
Jyong
2024-12-25 19:49:07 +08:00
committed by GitHub
parent 017d7538ae
commit 9231fdbf4c
54 changed files with 2578 additions and 808 deletions
Download Patch File Download Diff File Expand all files Collapse all files

55
api/migrations/versions/2024_11_22_0701-e19037032219_parent_child_index.py Normal file
View File

@@ -0,0 +1,55 @@
"""parent-child-index
Revision ID: e19037032219
Revises: 01d6889832f7
Create Date: 2024-11-22 07:01:17.550037
"""
from alembic import op
import models as models
import sqlalchemy as sa
# revision identifiers, used by Alembic.
revision = 'e19037032219'
down_revision = 'd7999dfa4aae'
branch_labels = None
depends_on = None
def upgrade():
# ### commands auto generated by Alembic - please adjust! ###
op.create_table('child_chunks',
sa.Column('id', models.types.StringUUID(), server_default=sa.text('uuid_generate_v4()'), nullable=False),
sa.Column('tenant_id', models.types.StringUUID(), nullable=False),
sa.Column('dataset_id', models.types.StringUUID(), nullable=False),
sa.Column('document_id', models.types.StringUUID(), nullable=False),
sa.Column('segment_id', models.types.StringUUID(), nullable=False),
sa.Column('position', sa.Integer(), nullable=False),
sa.Column('content', sa.Text(), nullable=False),
sa.Column('word_count', sa.Integer(), nullable=False),
sa.Column('index_node_id', sa.String(length=255), nullable=True),
sa.Column('index_node_hash', sa.String(length=255), nullable=True),
sa.Column('type', sa.String(length=255), server_default=sa.text("'automatic'::character varying"), nullable=False),
sa.Column('created_by', models.types.StringUUID(), nullable=False),
sa.Column('created_at', sa.DateTime(), server_default=sa.text('CURRENT_TIMESTAMP(0)'), nullable=False),
sa.Column('updated_by', models.types.StringUUID(), nullable=True),
sa.Column('updated_at', sa.DateTime(), server_default=sa.text('CURRENT_TIMESTAMP(0)'), nullable=False),
sa.Column('indexing_at', sa.DateTime(), nullable=True),
sa.Column('completed_at', sa.DateTime(), nullable=True),
sa.Column('error', sa.Text(), nullable=True),
sa.PrimaryKeyConstraint('id', name='child_chunk_pkey')
)
with op.batch_alter_table('child_chunks', schema=None) as batch_op:
batch_op.create_index('child_chunk_dataset_id_idx', ['tenant_id', 'dataset_id', 'document_id', 'segment_id', 'index_node_id'], unique=False)
# ### end Alembic commands ###
def downgrade():
# ### commands auto generated by Alembic - please adjust! ###
with op.batch_alter_table('child_chunks', schema=None) as batch_op:
batch_op.drop_index('child_chunk_dataset_id_idx')
op.drop_table('child_chunks')
# ### end Alembic commands ###

47
api/migrations/versions/2024_12_25_1137-923752d42eb6_add_auto_disabled_dataset_logs.py Normal file
View File

@@ -0,0 +1,47 @@
"""add_auto_disabled_dataset_logs
Revision ID: 923752d42eb6
Revises: e19037032219
Create Date: 2024-12-25 11:37:55.467101
"""
from alembic import op
import models as models
import sqlalchemy as sa
from sqlalchemy.dialects import postgresql
# revision identifiers, used by Alembic.
revision = '923752d42eb6'
down_revision = 'e19037032219'
branch_labels = None
depends_on = None
def upgrade():
# ### commands auto generated by Alembic - please adjust! ###
op.create_table('dataset_auto_disable_logs',
sa.Column('id', models.types.StringUUID(), server_default=sa.text('uuid_generate_v4()'), nullable=False),
sa.Column('tenant_id', models.types.StringUUID(), nullable=False),
sa.Column('dataset_id', models.types.StringUUID(), nullable=False),
sa.Column('document_id', models.types.StringUUID(), nullable=False),
sa.Column('notified', sa.Boolean(), server_default=sa.text('false'), nullable=False),
sa.Column('created_at', sa.DateTime(), server_default=sa.text('CURRENT_TIMESTAMP(0)'), nullable=False),
sa.PrimaryKeyConstraint('id', name='dataset_auto_disable_log_pkey')
)
with op.batch_alter_table('dataset_auto_disable_logs', schema=None) as batch_op:
batch_op.create_index('dataset_auto_disable_log_created_atx', ['created_at'], unique=False)
batch_op.create_index('dataset_auto_disable_log_dataset_idx', ['dataset_id'], unique=False)
batch_op.create_index('dataset_auto_disable_log_tenant_idx', ['tenant_id'], unique=False)
# ### end Alembic commands ###
def downgrade():
# ### commands auto generated by Alembic - please adjust! ###
with op.batch_alter_table('dataset_auto_disable_logs', schema=None) as batch_op:
batch_op.drop_index('dataset_auto_disable_log_tenant_idx')
batch_op.drop_index('dataset_auto_disable_log_dataset_idx')
batch_op.drop_index('dataset_auto_disable_log_created_atx')
op.drop_table('dataset_auto_disable_logs')
# ### end Alembic commands ###