zhoujl/rag_agent
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

init. project

Browse Source
This commit is contained in:
junlan
2026-04-13 11:34:23 +08:00
commit c7c0659a85
202 changed files with 31196 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

36
rag-web-ui/backend/alembic/versions/3580c0dcd005_increase_api_key_length.py Normal file
View File

@@ -0,0 +1,36 @@
"""increase_api_key_length
Revision ID: 3580c0dcd005
Revises: e214adf7fb66
Create Date: 2024-01-20 14:25:00.000000
"""
from typing import Sequence, Union
from alembic import op
import sqlalchemy as sa
# revision identifiers, used by Alembic.
revision: str = '3580c0dcd005'
down_revision: Union[str, None] = 'e214adf7fb66'
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
# ### commands auto generated by Alembic - please adjust! ###
op.alter_column('api_keys', 'key',
existing_type=sa.String(length=64),
type_=sa.String(length=128),
existing_nullable=False)
# ### end Alembic commands ###
def downgrade() -> None:
# ### commands auto generated by Alembic - please adjust! ###
op.alter_column('api_keys', 'key',
existing_type=sa.String(length=128),
type_=sa.String(length=64),
existing_nullable=False)
# ### end Alembic commands ###

116
rag-web-ui/backend/alembic/versions/59cfa0f1361d_rename_metadata_to_chunk_metadata.py Normal file
View File

@@ -0,0 +1,116 @@
"""rename_metadata_to_chunk_metadata
Revision ID: 59cfa0f1361d
Revises: initial_schema
Create Date: 2025-01-13 23:26:38.232326
"""
from typing import Sequence, Union
from alembic import op
import sqlalchemy as sa
from sqlalchemy.dialects import mysql
# revision identifiers, used by Alembic.
revision: str = '59cfa0f1361d'
down_revision: Union[str, None] = 'initial_schema'
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
# ### commands auto generated by Alembic - please adjust! ###
op.create_index(op.f('ix_chats_id'), 'chats', ['id'], unique=False)
op.add_column('document_chunks', sa.Column('document_id', sa.Integer(), nullable=False))
op.add_column('document_chunks', sa.Column('chunk_metadata', sa.JSON(), nullable=True))
op.alter_column('document_chunks', 'created_at',
existing_type=mysql.TIMESTAMP(),
type_=sa.DateTime(),
nullable=False,
existing_server_default=sa.text('CURRENT_TIMESTAMP'))
op.alter_column('document_chunks', 'updated_at',
existing_type=mysql.TIMESTAMP(),
type_=sa.DateTime(),
nullable=False,
existing_server_default=sa.text('CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP'))
op.drop_index('idx_hash', table_name='document_chunks')
op.create_index(op.f('ix_document_chunks_hash'), 'document_chunks', ['hash'], unique=False)
op.create_foreign_key(None, 'document_chunks', 'knowledge_bases', ['kb_id'], ['id'])
op.create_foreign_key(None, 'document_chunks', 'documents', ['document_id'], ['id'])
op.drop_column('document_chunks', 'metadata')
op.drop_index('idx_file_hash', table_name='documents')
op.create_index(op.f('ix_documents_file_hash'), 'documents', ['file_hash'], unique=False)
op.create_index(op.f('ix_documents_id'), 'documents', ['id'], unique=False)
op.create_index(op.f('ix_knowledge_bases_id'), 'knowledge_bases', ['id'], unique=False)
op.create_index(op.f('ix_messages_id'), 'messages', ['id'], unique=False)
op.alter_column('processing_tasks', 'knowledge_base_id',
existing_type=mysql.INTEGER(),
nullable=True)
op.alter_column('processing_tasks', 'document_id',
existing_type=mysql.INTEGER(),
nullable=True)
op.alter_column('processing_tasks', 'status',
existing_type=mysql.VARCHAR(length=50),
nullable=True)
op.alter_column('processing_tasks', 'created_at',
existing_type=mysql.DATETIME(),
nullable=True)
op.alter_column('processing_tasks', 'updated_at',
existing_type=mysql.DATETIME(),
nullable=True)
op.create_index(op.f('ix_processing_tasks_id'), 'processing_tasks', ['id'], unique=False)
op.drop_index('email', table_name='users')
op.drop_index('username', table_name='users')
op.create_index(op.f('ix_users_email'), 'users', ['email'], unique=True)
op.create_index(op.f('ix_users_id'), 'users', ['id'], unique=False)
op.create_index(op.f('ix_users_username'), 'users', ['username'], unique=True)
# ### end Alembic commands ###
def downgrade() -> None:
# ### commands auto generated by Alembic - please adjust! ###
op.drop_index(op.f('ix_users_username'), table_name='users')
op.drop_index(op.f('ix_users_id'), table_name='users')
op.drop_index(op.f('ix_users_email'), table_name='users')
op.create_index('username', 'users', ['username'], unique=True)
op.create_index('email', 'users', ['email'], unique=True)
op.drop_index(op.f('ix_processing_tasks_id'), table_name='processing_tasks')
op.alter_column('processing_tasks', 'updated_at',
existing_type=mysql.DATETIME(),
nullable=False)
op.alter_column('processing_tasks', 'created_at',
existing_type=mysql.DATETIME(),
nullable=False)
op.alter_column('processing_tasks', 'status',
existing_type=mysql.VARCHAR(length=50),
nullable=False)
op.alter_column('processing_tasks', 'document_id',
existing_type=mysql.INTEGER(),
nullable=False)
op.alter_column('processing_tasks', 'knowledge_base_id',
existing_type=mysql.INTEGER(),
nullable=False)
op.drop_index(op.f('ix_messages_id'), table_name='messages')
op.drop_index(op.f('ix_knowledge_bases_id'), table_name='knowledge_bases')
op.drop_index(op.f('ix_documents_id'), table_name='documents')
op.drop_index(op.f('ix_documents_file_hash'), table_name='documents')
op.create_index('idx_file_hash', 'documents', ['file_hash'], unique=False)
op.add_column('document_chunks', sa.Column('metadata', mysql.JSON(), nullable=True))
op.drop_constraint(None, 'document_chunks', type_='foreignkey')
op.drop_constraint(None, 'document_chunks', type_='foreignkey')
op.drop_index(op.f('ix_document_chunks_hash'), table_name='document_chunks')
op.create_index('idx_hash', 'document_chunks', ['hash'], unique=False)
op.alter_column('document_chunks', 'updated_at',
existing_type=sa.DateTime(),
type_=mysql.TIMESTAMP(),
nullable=True,
existing_server_default=sa.text('CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP'))
op.alter_column('document_chunks', 'created_at',
existing_type=sa.DateTime(),
type_=mysql.TIMESTAMP(),
nullable=True,
existing_server_default=sa.text('CURRENT_TIMESTAMP'))
op.drop_column('document_chunks', 'chunk_metadata')
op.drop_column('document_chunks', 'document_id')
op.drop_index(op.f('ix_chats_id'), table_name='chats')
# ### end Alembic commands ###

37
rag-web-ui/backend/alembic/versions/5be054bd6587_add_document_upload_id_to_processing_.py Normal file
View File

@@ -0,0 +1,37 @@
"""add_document_upload_id_to_processing_tasks
Revision ID: 5be054bd6587
Revises: fd73eebc87c1
Create Date: 2025-01-14 01:17:24.164593
"""
from typing import Sequence, Union
from alembic import op
import sqlalchemy as sa
# revision identifiers, used by Alembic.
revision: str = '5be054bd6587'
down_revision: Union[str, None] = 'fd73eebc87c1'
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
# 1. 添加 document_upload_id 字段
op.execute("""
ALTER TABLE processing_tasks
ADD COLUMN document_upload_id INT,
ADD CONSTRAINT processing_tasks_document_upload_id_fkey
FOREIGN KEY (document_upload_id) REFERENCES document_uploads(id)
""")
def downgrade() -> None:
# 1. 删除外键约束和字段
op.execute("""
ALTER TABLE processing_tasks
DROP FOREIGN KEY processing_tasks_document_upload_id_fkey,
DROP COLUMN document_upload_id
""")

102
rag-web-ui/backend/alembic/versions/a4f9c89b7d11_add_tool_jobs_and_srs_tables.py Normal file
View File

@@ -0,0 +1,102 @@
"""add tool jobs and srs tables
Revision ID: a4f9c89b7d11
Revises: 3580c0dcd005
Create Date: 2026-04-12 18:00:00.000000
"""
from typing import Sequence, Union
from alembic import op
import sqlalchemy as sa
from sqlalchemy.dialects import mysql
# revision identifiers, used by Alembic.
revision: str = "a4f9c89b7d11"
down_revision: Union[str, None] = "3580c0dcd005"
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
op.create_table(
"tool_jobs",
sa.Column("id", sa.Integer(), nullable=False),
sa.Column("user_id", sa.Integer(), nullable=False),
sa.Column("tool_name", sa.String(length=128), nullable=False),
sa.Column("status", sa.String(length=32), nullable=False, server_default="pending"),
sa.Column("input_file_name", sa.String(length=255), nullable=False),
sa.Column("input_file_path", sa.String(length=512), nullable=False),
sa.Column("error_message", sa.Text(), nullable=True),
sa.Column("started_at", sa.DateTime(), nullable=True),
sa.Column("completed_at", sa.DateTime(), nullable=True),
sa.Column("output_summary", sa.JSON(), nullable=True),
sa.Column("created_at", sa.DateTime(), nullable=False),
sa.Column("updated_at", sa.DateTime(), nullable=False),
sa.ForeignKeyConstraint(["user_id"], ["users.id"]),
sa.PrimaryKeyConstraint("id"),
)
op.create_index(op.f("ix_tool_jobs_id"), "tool_jobs", ["id"], unique=False)
op.create_index(op.f("ix_tool_jobs_tool_name"), "tool_jobs", ["tool_name"], unique=False)
op.create_index(op.f("ix_tool_jobs_user_id"), "tool_jobs", ["user_id"], unique=False)
op.create_table(
"srs_extractions",
sa.Column("id", sa.Integer(), nullable=False),
sa.Column("job_id", sa.Integer(), nullable=False),
sa.Column("document_name", sa.String(length=255), nullable=False),
sa.Column("document_title", sa.String(length=255), nullable=False),
sa.Column("generated_at", sa.DateTime(), nullable=False),
sa.Column("total_requirements", sa.Integer(), nullable=False),
sa.Column("statistics", sa.JSON(), nullable=True),
sa.Column("raw_output", sa.JSON(), nullable=True),
sa.Column("created_at", sa.DateTime(), nullable=False),
sa.Column("updated_at", sa.DateTime(), nullable=False),
sa.ForeignKeyConstraint(["job_id"], ["tool_jobs.id"], ondelete="CASCADE"),
sa.PrimaryKeyConstraint("id"),
sa.UniqueConstraint("job_id"),
)
op.create_index(op.f("ix_srs_extractions_id"), "srs_extractions", ["id"], unique=False)
op.create_table(
"srs_requirements",
sa.Column("id", sa.Integer(), nullable=False),
sa.Column("extraction_id", sa.Integer(), nullable=False),
sa.Column("requirement_uid", sa.String(length=64), nullable=False),
sa.Column("title", sa.String(length=255), nullable=False),
sa.Column("description", mysql.LONGTEXT(), nullable=False),
sa.Column("priority", sa.String(length=16), nullable=False),
sa.Column("acceptance_criteria", sa.JSON(), nullable=False),
sa.Column("source_field", sa.String(length=255), nullable=False),
sa.Column("section_number", sa.String(length=64), nullable=True),
sa.Column("section_title", sa.String(length=255), nullable=True),
sa.Column("requirement_type", sa.String(length=64), nullable=True),
sa.Column("sort_order", sa.Integer(), nullable=False),
sa.Column("created_at", sa.DateTime(), nullable=False),
sa.Column("updated_at", sa.DateTime(), nullable=False),
sa.ForeignKeyConstraint(["extraction_id"], ["srs_extractions.id"], ondelete="CASCADE"),
sa.PrimaryKeyConstraint("id"),
sa.UniqueConstraint("extraction_id", "requirement_uid", name="uq_srs_extraction_requirement_uid"),
)
op.create_index(op.f("ix_srs_requirements_id"), "srs_requirements", ["id"], unique=False)
op.create_index(
"idx_srs_requirements_extraction_sort",
"srs_requirements",
["extraction_id", "sort_order"],
unique=False,
)
def downgrade() -> None:
op.drop_index("idx_srs_requirements_extraction_sort", table_name="srs_requirements")
op.drop_index(op.f("ix_srs_requirements_id"), table_name="srs_requirements")
op.drop_table("srs_requirements")
op.drop_index(op.f("ix_srs_extractions_id"), table_name="srs_extractions")
op.drop_table("srs_extractions")
op.drop_index(op.f("ix_tool_jobs_user_id"), table_name="tool_jobs")
op.drop_index(op.f("ix_tool_jobs_tool_name"), table_name="tool_jobs")
op.drop_index(op.f("ix_tool_jobs_id"), table_name="tool_jobs")
op.drop_table("tool_jobs")

49
rag-web-ui/backend/alembic/versions/e214adf7fb66_add_api_keys_table.py Normal file
View File

@@ -0,0 +1,49 @@
"""add_api_keys_table
Revision ID: e214adf7fb66
Revises: 5be054bd6587
Create Date: 2024-01-20 13:24:00.000000
"""
from typing import Sequence, Union
from alembic import op
import sqlalchemy as sa
# revision identifiers, used by Alembic.
revision: str = 'e214adf7fb66'
down_revision: Union[str, None] = '5be054bd6587'
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
# ### commands auto generated by Alembic - please adjust! ###
op.create_table(
'api_keys',
sa.Column('id', sa.Integer(), nullable=False),
sa.Column('key', sa.String(length=64), nullable=False),
sa.Column('name', sa.String(length=255), nullable=False),
sa.Column('user_id', sa.Integer(), nullable=False),
sa.Column('is_active', sa.Boolean(), nullable=False, default=True),
sa.Column('last_used_at', sa.DateTime(), nullable=True),
sa.Column('created_at', sa.DateTime(), nullable=False, server_default=sa.text('CURRENT_TIMESTAMP')),
sa.Column('updated_at', sa.DateTime(), nullable=False, server_default=sa.text('CURRENT_TIMESTAMP'), onupdate=sa.text('CURRENT_TIMESTAMP')),
sa.ForeignKeyConstraint(['user_id'], ['users.id'], ),
sa.PrimaryKeyConstraint('id'),
sa.UniqueConstraint('key')
)
op.create_index(op.f('ix_api_keys_id'), 'api_keys', ['id'], unique=False)
op.create_index(op.f('ix_api_keys_key'), 'api_keys', ['key'], unique=True)
op.create_index(op.f('ix_api_keys_name'), 'api_keys', ['name'], unique=False)
# ### end Alembic commands ###
def downgrade() -> None:
# ### commands auto generated by Alembic - please adjust! ###
op.drop_index(op.f('ix_api_keys_name'), table_name='api_keys')
op.drop_index(op.f('ix_api_keys_key'), table_name='api_keys')
op.drop_index(op.f('ix_api_keys_id'), table_name='api_keys')
op.drop_table('api_keys')
# ### end Alembic commands ###

44
rag-web-ui/backend/alembic/versions/fd73eebc87c1_add_document_uploads_table.py Normal file
View File

@@ -0,0 +1,44 @@
"""add document uploads table
Revision ID: fd73eebc87c1
Revises: 59cfa0f1361d
Create Date: 2024-01-13 16:24:07.182834
"""
from alembic import op
import sqlalchemy as sa
# revision identifiers, used by Alembic.
revision = 'fd73eebc87c1'
down_revision = '59cfa0f1361d'
branch_labels = None
depends_on = None
def upgrade() -> None:
op.create_table(
'document_uploads',
sa.Column('id', sa.Integer(), nullable=False),
sa.Column('knowledge_base_id', sa.Integer(), nullable=False),
sa.Column('file_name', sa.String(255), nullable=False),
sa.Column('file_hash', sa.String(64), nullable=False),
sa.Column('file_size', sa.BigInteger(), nullable=False),
sa.Column('content_type', sa.String(100), nullable=False),
sa.Column('temp_path', sa.String(255), nullable=False),
sa.Column('created_at', sa.TIMESTAMP(), server_default=sa.text('now()'), nullable=False),
sa.Column('status', sa.String(50), nullable=False, server_default='pending'),
sa.Column('error_message', sa.Text(), nullable=True),
sa.PrimaryKeyConstraint('id'),
sa.ForeignKeyConstraint(['knowledge_base_id'], ['knowledge_bases.id'], ondelete='CASCADE')
)
# 添加索引以加速查询
op.create_index('ix_document_uploads_created_at', 'document_uploads', ['created_at'])
op.create_index('ix_document_uploads_status', 'document_uploads', ['status'])
def downgrade() -> None:
op.drop_index('ix_document_uploads_status')
op.drop_index('ix_document_uploads_created_at')
op.drop_table('document_uploads')

148
rag-web-ui/backend/alembic/versions/initial_schema.py Normal file
View File

@@ -0,0 +1,148 @@
"""initial schema
Revision ID: initial_schema
Revises:
Create Date: 2024-01-13 15:00:00.000000
"""
from typing import Sequence, Union
from alembic import op
import sqlalchemy as sa
from sqlalchemy.dialects import mysql
# revision identifiers, used by Alembic.
revision: str = 'initial_schema'
down_revision: Union[str, None] = None
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
# Create users table
op.create_table(
'users',
sa.Column('id', sa.Integer(), nullable=False),
sa.Column('email', sa.String(255), nullable=False),
sa.Column('username', sa.String(255), nullable=False),
sa.Column('hashed_password', sa.String(255), nullable=False),
sa.Column('is_active', sa.Boolean(), nullable=True, default=True),
sa.Column('is_superuser', sa.Boolean(), nullable=True, default=False),
sa.Column('created_at', sa.DateTime(), nullable=False),
sa.Column('updated_at', sa.DateTime(), nullable=False),
sa.PrimaryKeyConstraint('id'),
sa.UniqueConstraint('email'),
sa.UniqueConstraint('username')
)
# Create knowledge_bases table
op.create_table(
'knowledge_bases',
sa.Column('id', sa.Integer(), nullable=False),
sa.Column('name', sa.String(255), nullable=False),
sa.Column('description', mysql.LONGTEXT(), nullable=True),
sa.Column('user_id', sa.Integer(), nullable=False),
sa.Column('created_at', sa.DateTime(), nullable=False),
sa.Column('updated_at', sa.DateTime(), nullable=False),
sa.ForeignKeyConstraint(['user_id'], ['users.id'], ),
sa.PrimaryKeyConstraint('id')
)
# Create documents table
op.create_table(
'documents',
sa.Column('id', sa.Integer(), nullable=False),
sa.Column('file_path', sa.String(255), nullable=False),
sa.Column('file_name', sa.String(255), nullable=False),
sa.Column('file_size', sa.Integer(), nullable=True),
sa.Column('content_type', sa.String(100), nullable=True),
sa.Column('file_hash', sa.String(64), nullable=True),
sa.Column('knowledge_base_id', sa.Integer(), nullable=False),
sa.Column('created_at', sa.DateTime(), nullable=False),
sa.Column('updated_at', sa.DateTime(), nullable=False),
sa.ForeignKeyConstraint(['knowledge_base_id'], ['knowledge_bases.id'], ),
sa.PrimaryKeyConstraint('id'),
sa.UniqueConstraint('knowledge_base_id', 'file_name', name='uq_kb_file_name')
)
# Create document_chunks table
op.create_table(
'document_chunks',
sa.Column('id', sa.String(64), nullable=False),
sa.Column('kb_id', sa.Integer(), nullable=False),
sa.Column('file_name', sa.String(255), nullable=False),
sa.Column('metadata', sa.JSON(), nullable=True),
sa.Column('hash', sa.String(64), nullable=False),
sa.Column('created_at', sa.TIMESTAMP(), server_default=sa.text('CURRENT_TIMESTAMP')),
sa.Column('updated_at', sa.TIMESTAMP(), server_default=sa.text('CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP')),
sa.PrimaryKeyConstraint('id')
)
# Create chats table
op.create_table(
'chats',
sa.Column('id', sa.Integer(), nullable=False),
sa.Column('title', sa.String(255), nullable=False),
sa.Column('user_id', sa.Integer(), nullable=False),
sa.Column('created_at', sa.DateTime(), nullable=False),
sa.Column('updated_at', sa.DateTime(), nullable=False),
sa.ForeignKeyConstraint(['user_id'], ['users.id'], ),
sa.PrimaryKeyConstraint('id')
)
# Create chat_knowledge_bases table (association table)
op.create_table(
'chat_knowledge_bases',
sa.Column('chat_id', sa.Integer(), nullable=False),
sa.Column('knowledge_base_id', sa.Integer(), nullable=False),
sa.ForeignKeyConstraint(['chat_id'], ['chats.id'], ),
sa.ForeignKeyConstraint(['knowledge_base_id'], ['knowledge_bases.id'], ),
sa.PrimaryKeyConstraint('chat_id', 'knowledge_base_id')
)
# Create messages table
op.create_table(
'messages',
sa.Column('id', sa.Integer(), nullable=False),
sa.Column('chat_id', sa.Integer(), nullable=False),
sa.Column('role', sa.String(50), nullable=False),
sa.Column('content', mysql.LONGTEXT(), nullable=False),
sa.Column('created_at', sa.DateTime(), nullable=False),
sa.Column('updated_at', sa.DateTime(), nullable=False),
sa.ForeignKeyConstraint(['chat_id'], ['chats.id'], ),
sa.PrimaryKeyConstraint('id')
)
# Create processing_tasks table
op.create_table(
'processing_tasks',
sa.Column('id', sa.Integer(), nullable=False),
sa.Column('knowledge_base_id', sa.Integer(), nullable=False),
sa.Column('document_id', sa.Integer(), nullable=False),
sa.Column('status', sa.String(50), nullable=False, default='pending'),
sa.Column('error_message', sa.Text(), nullable=True),
sa.Column('created_at', sa.DateTime(), nullable=False),
sa.Column('updated_at', sa.DateTime(), nullable=False),
sa.ForeignKeyConstraint(['document_id'], ['documents.id'], ),
sa.ForeignKeyConstraint(['knowledge_base_id'], ['knowledge_bases.id'], ),
sa.PrimaryKeyConstraint('id')
)
# Create indexes
op.create_index('idx_kb_file_name', 'document_chunks', ['kb_id', 'file_name'])
op.create_index('idx_hash', 'document_chunks', ['hash'])
op.create_index('idx_file_hash', 'documents', ['file_hash'])
def downgrade() -> None:
# Drop indexes
op.drop_index('idx_hash', table_name='document_chunks')
op.drop_index('idx_kb_file_name', table_name='document_chunks')
op.drop_index('idx_file_hash', table_name='documents')
# Drop tables in reverse order
op.drop_table('processing_tasks')
op.drop_table('messages')
op.drop_table('chat_knowledge_bases')
op.drop_table('chats')
op.drop_table('document_chunks')
op.drop_table('documents')
op.drop_table('knowledge_bases')
op.drop_table('users')