Age | Commit message (Collapse) | Author |
|
* gn3/auth/authorisation/groups.py: new `user_group` function
* tests/unit/auth/test_groups.py: test `user_group` function
|
|
* gn3/api/metadata.py (jsonify_dataset_metadata): Rewrite metadata
end-point to use a dataset's name instead of it's accession_id.
* gn3/db/rdf.py (get_dataset_metadata): Replace accession_id with
name. Use one single RDF query instead of multiple queries.
|
|
|
|
* gn3/computations/correlations.py: Return `None` for missing literature
correlations rather than zero. Update the sorting key to avoid issues with
comparing floats to NoneType objects
|
|
|
|
|
|
* gn3/api/metadata.py: Import URLError.
(jsonify_dataset_metadata): Add URLError in except block.
|
|
* gn3/api/general.py: Delete rdf import. Delete trailing white
spaces.
* gn3/api/metadata.py: Delete trailing white spaces.
|
|
* gn3/api/metadata.py: import RemoteDisconnected.
(jsonify_dataset_metadata): Wrap get_dataset_metadata in try block.
|
|
* gn3/auth/authorisation/groups.py: base `MembershipError` on new
`AuthorisationError` base exception. Use new authentication checking
decorator.
* gn3/auth/authorisation/privileges.py: Change argument to User object rather
than UUID object
* gn3/auth/authorisation/roles.py: Use new authentication checking decorator.
* tests/unit/auth/test_groups.py: use `conftest.TEST_USER`
* tests/unit/auth/test_privileges.py: use `conftest.TEST_USER`
* tests/unit/auth/test_roles.py: use `conftest.TEST_USER`
|
|
* gn3/auth/authentication/checks.py: new `authenticated_p` decorator to apply
on any function that requires the user to be authenticated before it runs.
* gn3/auth/authorisation/checks.py: use a `auth.authentication.users.User`
object rather than a UUID object in the global `g`.
* gn3/auth/authorisation/groups.py: Implement the `authenticated_user_group`
function to get the group(s) in which the currently authenticated user
belongs.
* gn3/auth/authorisation/resources.py: Implement the `create_resource`
function correctly.
* tests/unit/auth/conftest.py: extract the User objects into a global variable
for reusability with the tests.
* tests/unit/auth/test_resources.py: Use global user objects from conftest in
the tests. Set a User object (rather than UUID) in the global `g` variable.
|
|
* gn3/auth/authentication/exceptions.py: define `AuthenticationError`
* gn3/auth/authorisation/exceptions.py: define `AuthorisationError`
|
|
* gn3/auth/authorisation/resources.py: Define Resource and ResourceCategory
classes. Create the `create_resource` stub.
* tests/unit/auth/test_resources.py: test for `create_resource` function
|
|
* gn3/api/general.py: (dataset_metadata) Delete.
* gn3/api/metadata.py: Import Blueprint, jsonify, current_app,
SPARQLWrapper and get_dataset_metadata.
(metadata): New Blueprint
(jsonify_dataset_metadata): New function/end-point.
* gn3/app.py: Import metadata
(create_app): Register metadata blueprint.
|
|
* gn3/auth/authorisation/groups.py: Assign the group leader at group creation
time.
* tests/unit/auth/test_groups.py: Ensure the group leader is only ever a
member of a single group.
|
|
* gn3/auth/authorisation/groups.py: Add `GroupRole` type. Fix typing
annotations. Fix bugs.
* tests/unit/auth/conftest.py: Fix bugs.
* tests/unit/auth/test_groups.py: Fix test to run.
|
|
* gn3/auth/authentication.py -> gn3/auth/authentication/__init__.py: Convert
module to package
* gn3/auth/authentication/users.py: Define the `User` type
|
|
* gn3/auth/authorisation/privileges.py: Set id to UUID type
* gn3/auth/authorisation/roles.py: fix parameters to types that sqlite3
supports
* gn3/auth/db.py: add logging for errors and re-raise the exception
* tests/unit/auth/test_roles.py: fix test
|
|
* gn3/db/rdf.py: Delete gn3.setting.SPARQL_ENDPOINT import.
(sparql_query): Inject SPARQLWrapper.
(get_dataset_metadata): Ditto.
|
|
|
|
* gn3/auth/authorisation/checks.py: Return results of calling the function
rather than a dict of values that include the results.
* gn3/auth/authorisation/groups.py: Use the newer form of `authorised_p`
decorator.
* tests/unit/auth/test_groups.py: Update tests
|
|
Use specified types for privileges, roles and types rather than using strings
to help with limiting bugs.
* gn3/auth/authorisation/groups.py: Specify and use the `Group` type
* gn3/auth/authorisation/privileges.py: Specify and use the `Privilege` type
* gn3/auth/authorisation/roles.py: Specify the `Role` type. Add the
`create_role` function.
|
|
|
|
* gn3/auth/authorisation/__init__.py: delete function
* gn3/auth/authorisation/checks.py: move function to `checks` module
|
|
|
|
|
|
* gn3/auth/authorisation/__init__.py: Add `authorised_p` decorator to be used
for all function requiring authorisation.
* gn3/auth/authorisation/groups.py: Add `create_group` function stub
* tests/unit/auth/conftest.py: Add fixture for test users
* tests/unit/auth/test_groups.py: Add tests for `create_group`
|
|
* gn3/migrations.py: Minor change
* migrations/auth/20221110_01_WtZ1I-create-resources-table.py: new migration
* tests/unit/auth/test_create_table_migrations.py: test new migration
|
|
selected
This is because base_samples was set to all_samples_ordered, which only
includes primary samples + parents/f1s. Setting this to an empty list
fixed the issue and caused it to use all samples again.
|
|
* gn3/settings.py: Omit trailing slash
* tests/unit/auth/test_create_table_migrations.py: Generalise testing
migrations that create tables.
* tests/unit/auth/test_create_user_credentials_table.py: delete
* tests/unit/auth/test_migration_create_users_table.py: delete
|
|
|
|
* gn3/auth/authentication.py: Fix issues caught by tests
* tests/unit/auth/test_credentials.py: Add fixtures and tests for credentials
checking
|
|
* gn3/auth/authentication.py: new function `credentials_in_database`
* gn3/auth/authentication/__init__.py: replace package with module
* gn3/settings.py: new `AUTH_MIGRATIONS` configuration variable
* migrations/auth/20221103_02_sGrIs-create-user-credentials-table.py: new
migration
* tests/unit/auth/test_credentials.py: test the `credentials_in_database`
function
* tests/unit/conftest.py: more test fixtures
|
|
Initialise the authentication/authorisation system packages and set up the
initial database migrations to set up the system.
* README.md: Add documentation on migrations
* gn3/auth/__init__.py: init package
* gn3/auth/authentication/__init__.py: init package
* gn3/auth/authorisation/__init__.py: init package
* gn3/migrations.py: provide migration utilities
* migrations/auth/20221103_01_js9ub-initialise-the-auth-entic-oris-ation-database.py:
new migration
* tests/unit/auth/test_init_database.py: test new migration applies and rolls
back as expected
* tests/unit/conftest.py: fixtures for unit tests
* yoyo.auth.ini: basic configuration for yoyo-migration for auth system
migrations
|
|
* gn3/api/general.py: Replace gn3.db.datasets import with gn3.db.rdf.
(dataset_metadata) <jsonify>: Replace datasets.dataset_metadata with
rdf.get_dataset_metadata.
* gn3/db/datasets.py: Remove unused imports.
(sparql_query, dataset_metadata): Delete.
* gn3/db/rdf.py: (sparql_query, get_dataset_metadata): New functions.
|
|
* gn3/db/sample_data.py (get_trait_csv_sample_data): Pass __query
directly to cursor.execute. Rename value to _value in for loop to
make mypy pass.
* gn3/db_utils.py (Connection): Add class docstring.
(Connection.cursor): Add docstring.
* mypy.ini: Add an entry for xapian.
|
|
App settings should be accessed from current_app. It should not be hard-coded
to a variable in a module.
* gn3/db_utils.py: Do not import XAPIAN_DB_PATH from gn3.settings.
(xapian_database): Accept path argument.
* gn3/api/search.py: Import current_app from flask.
(search_results): Pass Xapian index path to xapian_database.
|
|
* gn3/db_utils.py (Connection): Disable missing-class-docstring,
missing-function-docstring and too-few-public-methods pylint checks.
|
|
* gn3/api/search.py: New file.
* gn3/app.py: Register the search blueprint.
|
|
* gn3/settings.py (XAPIAN_DB_PATH): New variable.
* gn3/db_utils.py: Import xapian and XAPIAN_DB_PATH from gn3.settings.
(xapian_database): New function.
|
|
* gn3/genodb.py (matrix): Reshape column major genotype database storage as
ncols x nrows instead of as nrows x ncols.
|
|
|
|
* gn3/monads.py: Import SSDictCursor from MySQLdb.cursors.
(query_sql): Add server_side argument.
|
|
* gn3/db_utils.py: Import contextlib. Import Any, Iterator, Protocol and Tuple
from typing.
(database_connector): Deprecate function.
(Connection): New class.
(database_connection): New function.
|
|
* gn3/monads.py: New file.
Signed-off-by: Arun Isaac <arunisaac@systemreboot.net>
|
|
|
|
|
|
* gn3/computations/rust_correlation.py: (parse_tissue_corr_data): The function
was parsing data from a file, and then converting the parsed values into a
comma-separated string for output, despite the fact that the data parsed is
not output in this specific function. This is a really bad idea, and leads
to errors showing up down the line, far removed from the cause.
|
|
|
|
Use the built-in csv module for file output in order to avoid subtle errors in
the formats.
|