search-and-filter-assets.md

Quickstart: Search Assets Flow

This quickstart describes how assets can be found by their tags from Aquarius.

1. Setup

Ensure that you've already (a) installed Ocean, and (b) set up locally or remotely.

2. Alice publishes datasets

Now, you're Alice.

#data info
url = "https://raw.githubusercontent.com/trentmc/branin/main/branin.arff"

# Created a list of tags for the following assets
tags = [
    ["Branin dataset 1", "test", "ganache", "best asset"],
    ["Branin dataset 2", "test", "ocean"],
    ["Branin dataset 3", "AI", "dataset", "testing"],
]
# Publish few assets for testing
for tag in tags:
    name = tag[0]
    tx_dict = {"from": alice}
    metadata = ocean.assets.__class__.default_metadata(name, tx_dict)
    metadata.update({"tags": tag[1:]})
    (data_NFT, datatoken, ddo) = ocean.assets.create_url_asset(name, url, tx_dict, metadata=metadata)
    print(f"Just published asset, with did={ddo.did}")

3. Alice filters assets by their tags

Alice can filter the assets by a certain tag and after can retrieve the necessary information afterwards.

# Get a list of assets filtered by a given tag.
# All assets that contain the specified tag name
tag = "test"
all_ddos = ocean.assets.search(tag)

# Filter just by the `tags` key
filtered_ddos = list(
    filter(
        lambda a: tag in a.metadata["tags"],
        list(filter(lambda a: "tags" in a.metadata.keys(), all_ddos)),
    )
)

# Make sure that the provided tag is valid.
assert len(filtered_ddos) > 0, "Assets not found with this tag."

# Retrieve the wanted information from assets.
for ddo in filtered_ddos:
    print(f"ddo.did :{ddo.did}")
    print(f"ddo.metadata :{ddo.metadata}")
    print(f"ddo.nft :{ddo.nft}")
    print(f"ddo.datatokens :{ddo.datatokens}")

Running custom queries

You can run any custom ES query using OceanAssets. For example:

results = ocean.assets.query(
    {
        "query": {
            "query_string": {
                "query": "Branin dataset 1",
                "fields": ["metadata.name"],
            }
        }
    }
)
assert results[0].metadata["name"] == "Branin dataset 1"