invenio-rdm-migrator 4.4.5

InvenioRDM module for data migration.

DataCite-based data model for Invenio.

Development

Install

Make sure that you have libpq-dev installed in your system. See psycopg installation instructions for more information.

Choose a version of search and database, then run:

pip install -e .

Tests

./run-tests.sh

How to run it

To run the migration you need:

• A running InvenioRDM instance.

• If your data contains references to other records (e.g. vocabularies), then it is also required to run the setup step.

invenio-cli services setup --force --no-demo-data

• Install Invenio-RDM-Migrator, any other dependencies must be handled in the Pipfile of your instance.

$ pip install invenio-rdm-migrator

• Create/edit the configuration file on your instance, for example streams.yaml:

data_dir: /path/to/data
tmp_dir: /path/to/tmp
state_dir: /path/to/state
log_dir: /path/to/logs
db_uri: postgresql+psycopg2://inveniordm:inveniordm@localhost:5432/inveniordm
old_secret_key: CHANGE_ME
new_secret_key: CHANGE_ME
records:
    extract:
        filename: /path/to/records.json

• You will need to create a small python script putting together the different blocks of the ETL. You can find an eample at my-site/site/my_site/migrator/__main__.py.

from invenio_rdm_migrator.streams import StreamDefinition
from invenio_rdm_migrator.streams.records import RDMRecordCopyLoad

if __name__ == "__main__":
    RecordStreamDefinition = StreamDefinition(
        name="records",
        extract_cls=JSONLExtract,
        transform_cls=ZenodoToRDMRecordTransform,
        load_cls=RDMRecordCopyLoad,
    )

    runner = Runner(
        stream_definitions=[
            RecordStreamDefinition,
        ],
        config_filepath="path/to/your/streams.yaml",
    )

    runner.run()

• Finally, you can execute the above code. Since it is in the __main__ file of the python package, you can run it as a module:

$ python -m my_site.migrator

• Once the migration has completed, in your instance you can reindex the data. Following the records example above, it would look like:

$ invenio-cli pyshell

In [1]: from invenio_access.permissions import system_identity
In [2]: from invenio_rdm_records.proxies import current_rdm_records_service
In [3]: current_rdm_records_service.rebuild_index(identity=system_identity)

ETL {Extract/Transform/Load} architecture

There are four packages in this module extract, transform, load, and streams. The first three correspond to the three steps of an ETL process. The streams package contains the logic to run the process and different stream-specific implementations of ETL classes (e.g. records).

Extract

The extract is the first part of the data processing stream. It’s functionality is quite simple: return an iterator (e.g. of records), where each yielded value is a dictionary. Note that the data in this step is _transformed_ in format (e.g. JSON, XML), not in content. For example, the implementation of XMLExtract would look as follows:

class XMLExtract(Extract):
...

    def run(self):
        with open("file.xml") as file:
            for entry in file:
                yield xml.loads(entry)

Transform

The transformer is in charge of modifying the content to suit, in this case, the InvenioRDM data model (e.g. for records) so it can be imported in the DB. It will loop through the entries (i.e. the iterator returned by the extract class), transform and yield (e.g. the record). Diving more in the example of a record:

To transform something to an RDM record, you need to implement streams/records/transform.py:RDMRecordTransform. For each record it will yield what is considered a semantically “full” record: the record itself, its parent, its draft in case it exists and the files related them.

{
    "record": self._record(entry),
    "draft": self._draft(entry),
    "parent": self._parent(entry),
}

This means that you will need to implement the functions for each key. Note that, only _record and _parent should return content, the others can return None.

Some of these functions can themselves use a transform/base:Entry transformer. An _entry_ transformer is an extra layer of abstraction, to provide an interface with the methods needed to generate valid data for part of the Transform class. In the record example, you can implement transform.base:RDMRecordEntry, which can be used in the RDMRecordTransform._record function mentioned in the code snippet above. Note that implementing this interface will produce valid _data_ for a record. However, there is no abc for _metadata_. It is an open question how much we should define these interfaces and avoid duplicating the already existing Marshmallow schemas of InvenioRDM.

At this point you might be wondering “Why not Marshmallow then?”. The answer is “separation of responsibilities, performance and simplicity”. The later lays with the fact that most of the data transformation is custom, so we would end up with a schema full of Method fields, which does not differ much from what we have but would have an impact on performance (Marshmallow is slow…). Regarding the responsibilities part, validating (mostly referential, like vocabularies) can only be done on (or after) _load_ where RDM instance knowledge/appctx is available.

Note that no validation, not even structural, is done in this step.

Load

The final step to have the records available in the RDM instance is to load them. There are two types of loading _bulk_ or _transactions_.

Bulk

Bulk loading will insert data in the database table by table using COPY. Since the order of the tables is not guaranteed it is necessary to drop foreign keys before loading. They can be restored afterwards. In addition, dropping indices would increase performance since they will only be calculated once, when they are restored after loading.

Bulk loading is done using the load.postgresql.bulk:PostgreSQLCopyLoad class, which will carry out 2 steps:

1. Prepare the data, writing one DB row per line in a csv file:

$ /path/to/data/tables1668697280.943311
    |
    | - pidstore_pid.csv
    | - rdm_parents_metadata.csv
    | - rdm_records_metadata.csv
    | - rdm_versions_state.csv

2. Perform the actual loading, using COPY. Inserting all rows at once is more efficient than performing one INSERT per row.

Internally what is happening is that the prepare function makes use of TableGenerator implementations and then yields the list of csv files. So the load only iterates through the filenames, not the actual entries.

A TableGenerator will, for each value in the iterator, yield one or more rows (lines to be written to the a csv file). For example for a record it will yield: recid, DOI and OAI (PersistentIdentifiers), record and parent metadata, etc. which will be written to the respective CSV file.

Transactions

Another option is to migrate transactions. For example, once you have done the initial part of it in bulk, you can migrate the changes that were persisted while the bulk migration happened. That can be achieved by migrating transactions. A transaction is a group of operations, which can be understod as SQL statement and thus have two values: the operation type (created, update, delete) and its data represented as a database model.

Transaction loading is done using the load.postgresql.transactions:PostgreSQLExecuteLoad class, which will carry out 2 similar steps to the one above:

1. Prepare the data, storing in memory a series of Operations.

2. Perform the actual loading by adding or removing from the session, or updating the corresponding object. Each operation is flushed to the database to avoid foreing key violations. However, each transaction is atomic, meaning that an error in one of the operations will cause the full transaction to fail as a group.

Internally, the load will use an instance of load.postgresql.transactions.generators.group:TxGenerator to prepare the operations. This class contains a mapping between table names and load.postgresql.transactions.generators.row:RowGenerators, which will return a list of operations with the data as database model in the obj attribute.

Note that the TxGenerator is tightly coupled to the transform.transactions.Tx since it expects the dictionaries to have a specific structure:

{
    "tx_id": the actual transaction id, useful for debug and error handling
    "action": this information refers to the semantic meaning of the group
                   for example: record metadata update or file upload
    "operations": [
        {
            "op": c (create), u (update), d (delete)
            "table": the name of the table in the source system (e.g. Zenodo)
            "data": the transformed data, this can use any `Transform` implementation
        }
    ]
}

State

During a migration run, there is a need to share information across different streams or different generators on the same stream. For example, the records stream needs to access the UUID to slug map that was populated on the communities stream; or the drafts generator needs to know which parent records have been created on the records generator to keep the version state consistent.

All this information is persisted to a SQLite database. This state database is kept in memory during each stream processing, and it is persisted to disk if the stream finishes without errors. The state will be saved with the name of the stream (e.g. records.db) to avoid overwriting a previous state. Therefore, a migration can be restarted from any stream.

There are two ways to add more information to the state:

• Full entities, for example record or users, require their own DB table. Those must be defined at state.py:State._initialize_db. In addition, to abstract the access to that table, a state entity is required. It needs to be initialized in the Runner.py:Runner constructor and added the the state_entities dictionary.

• Independent value, for example the maximum value of generated primary keys. Those can be stored in the global_state. This state has two columns: key and value; adding information to it would look like {key: name_of_the_value, value: actual_value}.

Notes

Using python generators

Using generators instead of lists, allows us to iterate through the data only once and perform the E-T-L steps on them. Instead of loop for E, loop for T, loop for L. In addition, this allows us to have the csv files open during the writing and closing them at the end (open/close is an expensive op when done 3M times).

Changes

Version 4.4.1

• Fix default value for nullable model fields.

Version 4.4.0

• Add GitHub stream.

• Add verified_at and blocked_at for usesr models.

• Handle parent DOIs for records.

• Add media files to records and drafts.

• Add deletion_status to record models.

• Switch to orjson for JSON dumping/loading.

• Add multi-processing for transform.

• Refactor state to also use Python dict for caching.

Version 4.3.0

• Add community basic CRUD actions.

• Add DB session fixtures.

Version 4.2.0

• Rename FileUploadAction to DraftFileUploadAction.

Version 4.1.0

• Add file upload action.

• Add draft edit action.

Version 4.0.0

• Namespace actions by load and transform.

Version 3.1.0

• Add DatetimeMixin to transform timestamps into iso formatted date strings.

• Add JSONLoadMixin to load dictionaries from strings.

Version 3.0.0

• Operation instances have split the model and the data into two attributes.

• Add user actions.

• PostgreSQLTx resolve_references function has now a default behaviour (pass).

• Add nullable configuration to draft and user related models.

• Minor bug fixes.

Version 2.0.0

• Make state globally available.

• Refactor transactions into actions. Create transaction and load data classes.

• Removed empty kafka extract module.

• Improved error handling and created specialized classes.

• Move dict_set to utils.

• Remove Python 3.8 from test matrix.

Version 1.0.0

• Initial public release.