Asset Framework Navigation¶

An asset framework provides as structure to your collection of tags that make them easier to understand and navigate. Furthermore an asset framework allows cross-asset functionalities, as illustrated in the Cross-Asset Rollout example.

Setup¶

We start by authenticating a client and retrieving the tags we need for our example. Keep in mind the authenticated user will need access to the asset framework.

In [1]:

import os

import pandas as pd
from dotenv import find_dotenv, load_dotenv

from trendminer_interface import client_from_credentials

# Credentials and server details are securely stored in .env.docs for our example code.
# find_dotenv walks up from the working directory to locate it, so this works
# regardless of the directory the notebook is run from.
load_dotenv(find_dotenv(".env.docs", usecwd=True))

client = client_from_credentials(
    url=os.environ["TM_URL"],
    client_id=os.environ["TM_CLIENT_ID"],
    client_secret=os.environ["TM_CLIENT_SECRET"],
    username=os.environ["TM_USERNAME"],
    password=os.environ["TM_PASSWORD"],
    tz="Europe/Brussels",
)

import os import pandas as pd from dotenv import find_dotenv, load_dotenv from trendminer_interface import client_from_credentials # Credentials and server details are securely stored in .env.docs for our example code. # find_dotenv walks up from the working directory to locate it, so this works # regardless of the directory the notebook is run from. load_dotenv(find_dotenv(".env.docs", usecwd=True)) client = client_from_credentials( url=os.environ["TM_URL"], client_id=os.environ["TM_CLIENT_ID"], client_secret=os.environ["TM_CLIENT_SECRET"], username=os.environ["TM_USERNAME"], password=os.environ["TM_PASSWORD"], tz="Europe/Brussels", )

Retrieving specific assets¶

The deterministic way of retrieving a specific path is by its path of asset names, starting with the root asset. The root asset has the same name as the asset framework.

In [2]:

framework_name = "DO NOT EDIT"

asset = client.asset_framework.asset.from_path(
    names=[framework_name, "Houston", "Time"]
)
asset

framework_name = "DO NOT EDIT" asset = client.asset_framework.asset.from_path( names=[framework_name, "Houston", "Time"] ) asset

Out[2]:

Asset(name='Time')

We can always get the full hierarchy from an Asset instance:

In [3]:

asset.get_hierarchy()

asset.get_hierarchy()

Out[3]:

[Asset(name='DO NOT EDIT'), Asset(name='Houston'), Asset(name='Time')]

Navigating the tree¶

From an asset, we can always retrieve its children (which can be Asset or Attribute instances), as well as its parent Asset.

In [4]:

asset.search_children().collect()

asset.search_children().collect()

Out[4]:

[Attribute(name='Day'),
 Attribute(name='Hour'),
 Attribute(name='Month'),
 Attribute(name='Year')]

In [5]:

asset.parent

asset.parent

Out[5]:

Asset(name='Houston')

Sometimes we will want to grab a specific child asset or attribute.

In [6]:

asset.get_child_attribute(name="Hour")

asset.get_child_attribute(name="Hour")

Out[6]:

Attribute(name='Hour')

Searching for assets¶

We often want to search for a specific set of assets, so we can apply the same operation to them. For example, we can look for all of the assets with a certain name in a specific asset framework:

In [7]:

framework = client.asset_framework.from_name(framework_name)

assets = client.asset_framework.asset.search(
    name="Time",
    frameworks=[framework],
).collect()

assets

framework = client.asset_framework.from_name(framework_name) assets = client.asset_framework.asset.search( name="Time", frameworks=[framework], ).collect() assets

Out[7]:

[Asset(name='Time'), Asset(name='Time')]

A more robust approach is to explicitly define asset templates in our framework, and use these to search.

In [8]:

assets[0].template

assets[0].template

Out[8]:

'time tags'

In [9]:

client.asset_framework.asset.search(
    template="time tags",
    frameworks=[framework],
).collect()

client.asset_framework.asset.search( template="time tags", frameworks=[framework], ).collect()

Out[9]:

[Asset(name='Time'), Asset(name='Time')]

Or we might want to navigate to a specific parent asset first, and then search in its direct children. Note that the * is a wildcard character in the search.

In [10]:

from trendminer_interface import Asset

parent_asset = client.asset_framework.asset.from_path(
    names=["cip"],
)

assets = parent_asset.search_children(
    name="CIP Unit *",
    child_type=Asset,
).collect()

assets

from trendminer_interface import Asset parent_asset = client.asset_framework.asset.from_path( names=["cip"], ) assets = parent_asset.search_children( name="CIP Unit *", child_type=Asset, ).collect() assets

Out[10]:

[Asset(name='CIP unit 1'), Asset(name='CIP unit 2'), Asset(name='CIP unit 3')]

We can also search in all decendants of an asset (direct or indirect):

In [13]:

client.asset_framework.asset.search(
    name="CIP Unit *",
    ancestor=parent_asset,
).collect()

client.asset_framework.asset.search( name="CIP Unit *", ancestor=parent_asset, ).collect()

Out[13]:

[Asset(name='CIP unit 1'), Asset(name='CIP unit 2'), Asset(name='CIP unit 3')]