I want to insert profiles for a dataset in the stats part of datahub. is there a json file or another solution I can use for an example on how to do what i want. i will appreciate your help

Also we support profiling for any sql based sources ! https://datahubproject.io/docs/metadata-ingestion/source_docs/sql_profiles We will be bringing out more improvements to this in the next release (going to be demoed in townhall this Friday!)

xL thank you for your help. do you have any idea if there is a python code doing the same thing?

you mean to say, you want to manually create a dataset profile? you could look at the profiling code: https://github.com/linkedin/datahub/blob/master/metadata-ingestion/src/datahub/ingestion/source/ge_data_profiler.py and the sqlalchemy bit: https://github.com/linkedin/datahub/blob/master/metadata-ingestion/src/datahub/ingestion/source/sql/sql_common.py and see how you want to do it

xL what im trying to figure out is how can i send the json file you sent me https://github.com/linkedin/datahub/blob/master/metadata-ingestion/examples/mce_files/demo-profiles.json via a python script to a stats part of a dataset

So this json file will write several profile run results to a particular dataset as specified by the entityUrn. You could change the urn to write this dummy data to your dataset. To write this to datahub, create a yml like the following: https://github.com/linkedin/datahub/blob/master/metadata-ingestion/examples/recipes/file_to_datahub_rest.yml Then you can write it to datahub using the datahub ingest -c config-file.yml

also is there a aspect for adding profile to the dataset, like the link below:https://github.com/linkedin/datahub/blob/master/metadata-ingestion/examples/mce_files/bootstrap_mce.json

Dataset profiles are not aspects of a dataset unlike schemaMetadata or institutionalMemory.

thank you for your time. so the only way to add data profile is by manually using https://github.com/linkedin/datahub/blob/master/metadata-ingestion/examples/mce_files/demo-profiles.json

Well yes, it's either 1. Use the provided connectors or 2. Write your own json. You can call the datasetprofile class to help guide the populating of values.