diff --git a/migrations/env.py b/migrations/env.py
index 251d615548bd0bc71933d8f180012782a94b6ccb..1625aef4e0acbc6090fdbcdf6cff325299e9f92c 100644
--- a/migrations/env.py
+++ b/migrations/env.py
@@ -53,25 +53,23 @@ def run_migrations_online():
and associate a connection with the context.
"""
- alembic_config = config.get_section(config.config_ini_section)
- alembic_config['sqlalchemy.url'] = settings.DATABASE_URI
-
- engine = engine_from_config(
- alembic_config,
- prefix='sqlalchemy.',
- poolclass=pool.NullPool)
-
- connection = engine.connect()
- context.configure(
- connection=connection,
- target_metadata=target_metadata
- )
-
- try:
+ # Enable reusing an existing connection.
+ # http://alembic.readthedocs.org/en/latest/cookbook.html#sharing-a-connection-with-a-series-of-migration-commands-and-environments
+ connectable = config.attributes.get('connection', None)
+
+ if connectable is None:
+ alembic_config = config.get_section(config.config_ini_section)
+ alembic_config['sqlalchemy.url'] = settings.DATABASE_URI
+ connectable = engine_from_config(alembic_config,
+ prefix='sqlalchemy.',
+ poolclass=pool.NullPool)
+
+ with connectable.connect() as connection:
+ context.configure(connection=connection,
+ target_metadata=target_metadata)
with context.begin_transaction():
context.run_migrations()
- finally:
- connection.close()
+
if context.is_offline_mode():
run_migrations_offline()
diff --git a/migrations/versions/ea660b66f26_initial_schema.py b/migrations/versions/ea660b66f26_initial_schema.py
index eec6ce6af5ee8767be03e99bda445305002394b1..bd3c148349274bb9b966410d4d12b66cfa54bd95 100644
--- a/migrations/versions/ea660b66f26_initial_schema.py
+++ b/migrations/versions/ea660b66f26_initial_schema.py
@@ -1,4 +1,4 @@
-"""initial schema
+"""Initial schema
Revision ID: ea660b66f26
Revises: None
@@ -17,8 +17,132 @@ import sqlalchemy as sa
def upgrade():
- pass
+ ### commands auto generated by Alembic - please adjust! ###
+ op.create_table('assemblies',
+ sa.Column('id', sa.Integer(), nullable=False),
+ sa.Column('name', sa.String(length=30), nullable=False),
+ sa.Column('alias', sa.String(length=10), nullable=True),
+ sa.Column('taxonomy_id', sa.Integer(), nullable=False),
+ sa.Column('taxonomy_common_name', sa.String(length=50), nullable=False),
+ sa.PrimaryKeyConstraint('id'),
+ sa.UniqueConstraint('alias'),
+ sa.UniqueConstraint('name'),
+ mysql_charset='utf8',
+ mysql_engine='InnoDB'
+ )
+ op.create_table('batch_jobs',
+ sa.Column('id', sa.Integer(), nullable=False),
+ sa.Column('email', sa.String(length=200), nullable=True),
+ sa.Column('download_url', sa.String(length=200), nullable=True),
+ sa.Column('job_type', sa.Enum('name-checker', 'syntax-checker', 'position-converter', 'snp-converter', name='job_type'), nullable=False),
+ sa.Column('argument', sa.String(length=20), nullable=True),
+ sa.Column('result_id', sa.String(length=50), nullable=False),
+ sa.Column('added', sa.DateTime(), nullable=True),
+ sa.PrimaryKeyConstraint('id'),
+ mysql_charset='utf8',
+ mysql_engine='InnoDB'
+ )
+ op.create_index(op.f('ix_batch_jobs_result_id'), 'batch_jobs', ['result_id'], unique=True)
+ op.create_table('references',
+ sa.Column('id', sa.Integer(), nullable=False),
+ sa.Column('accession', sa.String(length=20), nullable=False),
+ sa.Column('checksum', sa.String(length=32), nullable=False),
+ sa.Column('geninfo_identifier', sa.String(length=13), nullable=True),
+ sa.Column('slice_accession', sa.String(length=20), nullable=True),
+ sa.Column('slice_start', sa.Integer(), nullable=True),
+ sa.Column('slice_stop', sa.Integer(), nullable=True),
+ sa.Column('slice_orientation', sa.Enum('forward', 'reverse', name='slice_orentation'), nullable=True),
+ sa.Column('download_url', sa.String(length=255), nullable=True),
+ sa.Column('added', sa.DateTime(), nullable=True),
+ sa.PrimaryKeyConstraint('id'),
+ mysql_charset='utf8',
+ mysql_engine='InnoDB'
+ )
+ op.create_index(op.f('ix_references_accession'), 'references', ['accession'], unique=True)
+ op.create_index(op.f('ix_references_checksum'), 'references', ['checksum'], unique=True)
+ op.create_index(op.f('ix_references_download_url'), 'references', ['download_url'], unique=True)
+ op.create_index(op.f('ix_references_geninfo_identifier'), 'references', ['geninfo_identifier'], unique=True)
+ op.create_index('reference_slice', 'references', ['slice_accession', 'slice_start', 'slice_stop', 'slice_orientation'], unique=True)
+ op.create_table('transcript_protein_links',
+ sa.Column('id', sa.Integer(), nullable=False),
+ sa.Column('transcript_accession', sa.String(length=20), nullable=False),
+ sa.Column('protein_accession', sa.String(length=20), nullable=True),
+ sa.Column('added', sa.DateTime(), nullable=True),
+ sa.PrimaryKeyConstraint('id'),
+ mysql_charset='utf8',
+ mysql_engine='InnoDB'
+ )
+ op.create_index(op.f('ix_transcript_protein_links_protein_accession'), 'transcript_protein_links', ['protein_accession'], unique=False)
+ op.create_index(op.f('ix_transcript_protein_links_transcript_accession'), 'transcript_protein_links', ['transcript_accession'], unique=True)
+ op.create_table('batch_queue_items',
+ sa.Column('id', sa.Integer(), nullable=False),
+ sa.Column('batch_job_id', sa.Integer(), nullable=False),
+ sa.Column('item', sa.String(length=200), nullable=False),
+ sa.Column('flags', sa.String(length=20), nullable=False),
+ sa.ForeignKeyConstraint(['batch_job_id'], ['batch_jobs.id'], ondelete='CASCADE'),
+ sa.PrimaryKeyConstraint('id'),
+ mysql_charset='utf8',
+ mysql_engine='InnoDB'
+ )
+ op.create_index('batch_queue_item_with_batch_job', 'batch_queue_items', ['batch_job_id', 'id'], unique=False)
+ op.create_table('chromosomes',
+ sa.Column('id', sa.Integer(), nullable=False),
+ sa.Column('assembly_id', sa.Integer(), nullable=False),
+ sa.Column('name', sa.String(length=30), nullable=False),
+ sa.Column('accession', sa.String(length=30), nullable=False),
+ sa.Column('organelle', sa.Enum('nucleus', 'mitochondrion', name='organelle'), nullable=True),
+ sa.ForeignKeyConstraint(['assembly_id'], ['assemblies.id'], ondelete='CASCADE'),
+ sa.PrimaryKeyConstraint('id'),
+ mysql_charset='utf8',
+ mysql_engine='InnoDB'
+ )
+ op.create_index('chromosome_accession', 'chromosomes', ['assembly_id', 'accession'], unique=True)
+ op.create_index('chromosome_name', 'chromosomes', ['assembly_id', 'name'], unique=True)
+ op.create_table('transcript_mappings',
+ sa.Column('id', sa.Integer(), nullable=False),
+ sa.Column('chromosome_id', sa.Integer(), nullable=False),
+ sa.Column('reference_type', sa.Enum('refseq', 'lrg', name='reference_type'), nullable=False),
+ sa.Column('accession', sa.String(length=20), nullable=False),
+ sa.Column('version', sa.Integer(), nullable=True),
+ sa.Column('gene', sa.String(length=30), nullable=False),
+ sa.Column('transcript', sa.Integer(), nullable=False),
+ sa.Column('orientation', sa.Enum('forward', 'reverse', name='orentation'), nullable=False),
+ sa.Column('start', sa.Integer(), nullable=False),
+ sa.Column('stop', sa.Integer(), nullable=False),
+ sa.Column('cds_start', sa.Integer(), nullable=True),
+ sa.Column('cds_stop', sa.Integer(), nullable=True),
+ sa.Column('exon_starts', sa.Text(), nullable=False),
+ sa.Column('exon_stops', sa.Text(), nullable=False),
+ sa.Column('select_transcript', sa.Boolean(), nullable=False),
+ sa.Column('source', sa.Enum('ucsc', 'ncbi', 'reference', name='source'), nullable=False),
+ sa.ForeignKeyConstraint(['chromosome_id'], ['chromosomes.id'], ondelete='CASCADE'),
+ sa.PrimaryKeyConstraint('id'),
+ mysql_charset='utf8',
+ mysql_engine='InnoDB'
+ )
+ op.create_index('transcript_mapping_transcript', 'transcript_mappings', ['accession', 'version', 'gene', 'transcript', 'chromosome_id'], unique=True)
+ ### end Alembic commands ###
def downgrade():
- pass
+ ### commands auto generated by Alembic - please adjust! ###
+ op.drop_index('transcript_mapping_transcript', table_name='transcript_mappings')
+ op.drop_table('transcript_mappings')
+ op.drop_index('chromosome_name', table_name='chromosomes')
+ op.drop_index('chromosome_accession', table_name='chromosomes')
+ op.drop_table('chromosomes')
+ op.drop_index('batch_queue_item_with_batch_job', table_name='batch_queue_items')
+ op.drop_table('batch_queue_items')
+ op.drop_index(op.f('ix_transcript_protein_links_transcript_accession'), table_name='transcript_protein_links')
+ op.drop_index(op.f('ix_transcript_protein_links_protein_accession'), table_name='transcript_protein_links')
+ op.drop_table('transcript_protein_links')
+ op.drop_index('reference_slice', table_name='references')
+ op.drop_index(op.f('ix_references_geninfo_identifier'), table_name='references')
+ op.drop_index(op.f('ix_references_download_url'), table_name='references')
+ op.drop_index(op.f('ix_references_checksum'), table_name='references')
+ op.drop_index(op.f('ix_references_accession'), table_name='references')
+ op.drop_table('references')
+ op.drop_index(op.f('ix_batch_jobs_result_id'), table_name='batch_jobs')
+ op.drop_table('batch_jobs')
+ op.drop_table('assemblies')
+ ### end Alembic commands ###
diff --git a/tests/test_migrations.py b/tests/test_migrations.py
new file mode 100644
index 0000000000000000000000000000000000000000..c367e43c6159aeabc79a282af4e59441b2ea8cb3
--- /dev/null
+++ b/tests/test_migrations.py
@@ -0,0 +1,184 @@
+"""
+Test database migrations.
+"""
+
+
+from __future__ import unicode_literals
+
+import os
+
+import alembic.autogenerate
+import alembic.command
+import alembic.config
+from alembic.migration import MigrationContext
+import sqlalchemy as sa
+from sqlalchemy import create_engine, sql
+
+from mutalyzer import db
+
+
+def test_migrations():
+ """
+ Run all migrations and assert the result is up to date with the model
+ definitions.
+
+ We don't use `utils.MutalyzerTest` here, or `mutalyzer.db.session` in any
+ way for that matter, since it will bootstrap the database schema.
+ """
+ database_uri = os.getenv('MUTALYZER_TEST_DATABASE_URI', 'sqlite://')
+
+ alembic_config = alembic.config.Config('migrations/alembic.ini')
+ engine = create_engine(database_uri)
+
+ with engine.begin() as connection:
+ # http://alembic.readthedocs.org/en/latest/cookbook.html#sharing-a-connection-with-a-series-of-migration-commands-and-environments
+ alembic_config.attributes['connection'] = connection
+
+ if database_uri != 'sqlite://':
+ db.Base.metadata.drop_all(connection)
+
+ # Create initial schema by running the first migration.
+ alembic.command.upgrade(alembic_config, 'ea660b66f26')
+
+ # Add some database content to run the migrations on.
+ add_database_content(connection)
+
+ # Run the remaining migrations.
+ alembic.command.upgrade(alembic_config, 'head')
+
+ context = MigrationContext.configure(connection)
+ assert not alembic.autogenerate.compare_metadata(
+ context, db.Base.metadata)
+
+ engine.dispose()
+
+
+def add_database_content(connection):
+ """
+ Add some content to the database.
+ """
+ # We only define tables and columns we actually need, so this is not a
+ # complete mapping of the schema.
+
+ assemblies = sql.table(
+ 'assemblies',
+ sql.column('id', sa.Integer),
+ sql.column('name', sa.String(30)),
+ sql.column('alias', sa.String(10)),
+ sql.column('taxonomy_id', sa.Integer),
+ sql.column('taxonomy_common_name', sa.String(50)))
+
+ chromosomes = sql.table(
+ 'chromosomes',
+ sql.column('id', sa.Integer),
+ sql.column('assembly_id', sa.Integer),
+ sql.column('name', sa.String(30)),
+ sql.column('accession', sa.String(30)),
+ sql.column('organelle', sa.Enum('nucleus', 'mitochondrion',
+ name='organelle')))
+
+ transcript_mappings = sql.table(
+ 'transcript_mappings',
+ sql.column('chromosome_id', sa.Integer),
+ sql.column('reference_type', sa.Enum('refseq', 'lrg',
+ name='reference_type')),
+ sql.column('accession', sa.String(20)),
+ sql.column('gene', sa.String(30)),
+ sql.column('transcript', sa.Integer),
+ sql.column('orientation', sa.Enum('forward', 'reverse',
+ name='orentation')),
+ sql.column('start', sa.Integer),
+ sql.column('stop', sa.Integer),
+ sql.column('exon_starts', sa.Text),
+ sql.column('exon_stops', sa.Text),
+ sql.column('select_transcript', sa.Boolean),
+ sql.column('source', sa.Enum('ucsc', 'ncbi', 'reference',
+ name='source')))
+
+ transcript_protein_links = sql.table(
+ 'transcript_protein_links',
+ sql.column('transcript_accession', sa.String(30)),
+ sql.column('protein_accession', sa.String(30)))
+
+ # Add some common data.
+ connection.execute(
+ assemblies.insert(),
+ name='GRCh37',
+ taxonomy_id=9606,
+ taxonomy_common_name='Homo sapiens',
+ alias='hg19')
+ hg19_id = connection.execute(
+ assemblies.select(assemblies.c.alias == 'hg19')
+ .with_only_columns([assemblies.c.id])
+ ).fetchone()[0]
+
+ connection.execute(
+ chromosomes.insert(),
+ assembly_id=hg19_id,
+ name='chr1',
+ accession='NC_000001.10',
+ organelle='nucleus')
+ chr1_id = connection.execute(
+ chromosomes.select(chromosomes.c.name == 'chr1')
+ .with_only_columns([chromosomes.c.id])
+ ).fetchone()[0]
+
+ # Data for migration 402ff01b0d5d:
+ # Fix GRCm38 chromosome accession number versions.
+ connection.execute(
+ chromosomes.insert(),
+ assembly_id=hg19_id,
+ name='chr11',
+ accession='NC_000077.60',
+ organelle='nucleus')
+
+ # Data for migration 2e062969eb54:
+ # Rename GRCh36 assembly to NCBI36.
+ connection.execute(
+ assemblies.insert(),
+ name='GRCh36',
+ taxonomy_id=9606,
+ taxonomy_common_name='Homo sapiens',
+ alias='hg18')
+
+ # Data for migration 4bafcc5086dd:
+ # Fix zero-exon transcript mappings.
+ connection.execute(
+ transcript_mappings.insert(),
+ chromosome_id=chr1_id,
+ reference_type='refseq',
+ accession='NC_001807',
+ gene='ATP6',
+ transcript=1,
+ orientation='forward',
+ start=8528,
+ stop=9208,
+ exon_starts='8528',
+ exon_stops='9208',
+ select_transcript=True,
+ source='ncbi')
+ connection.execute(
+ transcript_mappings.insert(),
+ chromosome_id=chr1_id,
+ reference_type='refseq',
+ accession='NC_001807',
+ gene='ATP8',
+ transcript=1,
+ orientation='forward',
+ start=8367,
+ stop=8573,
+ exon_starts='',
+ exon_stops='',
+ select_transcript=True,
+ source='ncbi')
+
+ # Data for migration 3492d2ee8884:
+ # Transcript protein links have nullable transcript and unique protein.
+ connection.execute(
+ transcript_protein_links.insert(),
+ transcript_accession='NM_052818',
+ protein_accession='NP_438169')
+ connection.execute(
+ transcript_protein_links.insert(),
+ transcript_accession='NM_001079691',
+ protein_accession=None)
diff --git a/tests/utils.py b/tests/utils.py
index aa90a1fd7b8c21bbc9c9ae97a676004b5a11854d..6743804fdb528b75d0ffb3c1c2a90b821c9ae88f 100644
--- a/tests/utils.py
+++ b/tests/utils.py
@@ -48,6 +48,8 @@ class TestEnvironment(object):
"""
Destroy all storage defined in the current environment.
"""
+ db.session.remove()
+
shutil.rmtree(self.cache_dir)
os.unlink(self.log_file)