Search

In this tutorial we will find a dataset in a Node by performing a search over the entries’ metadata.

To follow along, start the Tiled server with example data from a Terminal.

tiled serve pyobject --public tiled.examples.generated:tree

Search Using the Python Client

Now, in a Python interpreter, connect with the Python client.

from tiled.client import from_uri

client = from_uri("http://localhost:8000")

Tiled has an extensible collection of queries. The client just has to construct the query, and server sorts out how to execute it as efficiently as possible given however the metadata and data are stored.

This example collection of data has several entries with metadata.

>>> client['short_table'].metadata
DictView({'animal': 'dog', 'color': 'red'})

>>> client['long_table'].metadata
DictView({'animal': 'dog', 'color': 'green'})

>>> client['structured_data'].metadata
DictView({'animal': 'cat', 'color': 'green'})

# etc.

We’ll search among them for entries where the term "dog" appears anywhere in the metadata.

>>> from tiled.queries import FullText

>>> client.search(FullText("dog"))
<Node {'short_table', 'long_table'}>

The result is another client, with a subset of the entries or the original. We might next stash it in a variable and drill further down.

>>> results = client.search(FullText("dog"))
>>> results['short_table']
<DataFrameClient>

Searches may be chained:

>>> client.search(FullText("dog")).search(FullText("red"))
<Node {'short_table'}>

If there no matches, the result is an empty Node:

>>> client.search(FullText("something that will not be found"))
<Node {}>

Roadmap

Currently, FullText is the only outwardly-useful query supported. More will be added, as well as documentation on how to register user-defined ones.