Full class name stored for process_type in db_dbnode instead of entry point. Manual fix possible?

[dev-zero]

It seems that I made a mistake at some point wrt an entry point specification in my setup.json in the sense that I did not specify the full path but the imported path (details for that below). Now I have in my db_dbnode as the process_type the full class name aiida_cp2k.workchains.my.MyWorkchain rather than the entry point name aiida.workflows:cp2k.my AiiDA uses when the entry point is correctly specified:

aiida=# select process_type from db_dbnode where id = 58676;
                process_type
--------------------------------------------
 aiida_cp2k.workchains.my.MyWorkchain

When I fix the entry point the queries the QueryBuilder generates when doing:

qb = QueryBuilder()
qb.append(WorkflowFactory("cp2k.my", project=["*"])
qb.all()

now include the entry point name rather than the full class name:

SELECT db_dbnode_1.id AS db_dbnode_1_id, db_dbnode_1.uuid AS db_dbnode_1_uuid, db_dbnode_1.node_type AS db_dbnode_1_node_type, db_dbnode_1.process_type AS db_dbnode_1_process_type, db_dbnode_1.label AS db_dbnode_1_label, db_dbnode_1.description AS db_dbnode_1_description, db_dbnode_1.ctime AS db_dbnode_1_ctime, db_dbnode_1.mtime AS db_dbnode_1_mtime, db_dbnode_1.attributes AS db_dbnode_1_attributes, db_dbnode_1.extras AS db_dbnode_1_extras, db_dbnode_1.repository_metadata AS db_dbnode_1_repository_metadata, db_dbnode_1.user_id AS db_dbnode_1_user_id, db_dbnode_1.dbcomputer_id AS db_dbnode_1_dbcomputer_id
    FROM db_dbnode AS db_dbnode_1
    WHERE CAST(db_dbnode_1.node_type AS VARCHAR) LIKE 'process.workflow.workchain.%'
    AND
    (db_dbnode_1.process_type = 'aiida.workflows:cp2k.my' OR CAST(db_dbnode_1.process_type AS VARCHAR) LIKE 'aiida.workflows:cp2k.my.')

Which of course breaks my queries. Can I simply do an SQL update statement to fix the process_type?

Details about the mistake

The mistake in my entry point specification was that I have the following project setup:

aiida_cp2k/workchains/__init__.py
aiida_cp2k/workchains/my.py

and in aiida_cp2k/workchains/__init__.py I have:

from .my import MyWorkchain

then I used as my entry point spec:

    "entry_points": {
        "aiida.workflows": [
            "cp2k.my = aiida_cp2k.workchains:MyWorkChain",
        ]
    }

which at some version of AiiDA started to give me the warning:

/scratch/tiziano/work/aiida/aiida-core/aiida/orm/querybuilder.py:1328: AiidaEntryPointWarning: Process type
'aiida_cp2k.workchains.my.MyWorkchain' does not correspond to a registered entry.
This risks queries to fail once the location of the process class changes.
Add an entry point for 'aiida_cp2k.workchains.my.MyWorkchain' to remove this warning.

Thanks in advance!

[zhubonan]

I think you can just do a SQL update - seems to be the only way out, I have used it to update computer hostnames etc..
Otherwise, AiiDA also have a low-level interface which may allow you to update some of the attributes (Node._backend_entity)? I don't have any experience with it though.

[dev-zero]

Thank you very much!