api

API

🕘 5 minutes

Syre's API is available for

• Python
• R

If your favorite language isn't yet supported let us know either on our Discord or by email at [email protected].

Below is an overview of Syre's API. For more details and examples, visit your language's docs.

Sandwich model

Syre uses a sandwich model to allow you to interact with your projects, where Syre is the bread and your analysis is the meat. This allows you to easily adapt scripts you've already created to Syre, or take Syre scripts and use them outside the framework.

Initializing a database

The first thing to do in any project is initialize the database. This gives you access to your project. If you want to interact with your project manually, assign the dev_root parameter to the absolute path of the Container you want to interact with. When Syre is analyzing a project, the dev_root is ignored.

Getting resources

Syre exposes four methods to get project resources.

• find_asset / find_container: Finds a single Asset/Container matching the provided filter, returning the languages None type if no matches are found. If multiple resources are found, a random one is returned.
• find_assets / find_containers: Finds multiple Assets/Containers, returning them as the language's list type.

Each of these functions accepts filter parameters that correspond to the properties you can assign them in the details bar of the desktop app

• name (string | None)
• type (string | None)
• tags (list)
• metadata (dictionary)

A resource must match all of the provided filter parameters.

Hierarchical database

Syre projects act as a hierarchical database. This means that each subtree has an isolated environment to operate in. When an analysis script is running over a certain container, it has access to everything in that Container and its subtree, but nothing outside of that.

Adding Assets

To add a new Asset to your project, use the add_asset method. This takes the file name to use for the Asset and each of properties you want to assign to it.

Buckets

If you give the file name as a relative path, e.g. bucket/my_data.png, Syre will create subfolders matching the path to store the file in. These subfolders are used to organize Assets without creating new Containers, and are called buckets.

Dev Mode

You can perform certain tasks in a script only when you are interacting with it manually. Think of these as debug statements that are ignored when you are actually analyzing your project. This is done with the dev_mode method, which returns true when you manually running the script, and false when the script is being run by Syre.

Filtering resources

You can filter resources you've already collected using the filter method. This takes in a list of resources and filter parameters, and returns a new list containing only those that match.

Example

This example is in Python-like pseudocode. For examples in your preferred language, visit the docs for it.

import syre

# initialize database
db = syre.Database(dev_root="/absolute/path/to/my/project/data/container")

# get a container with name `Child 1`
child_one = db.find_container(name="Child 1")
if syre.dev_mode:
    print(child_one.type)

# get assets with type `raw_data`
raw_assets = db.find_assets(type="raw_data")

# read data from asset files
raw_data = []
for asset in raw_assets:
    raw_data = read_data(asset.file)
    data.append((raw_data, asset.metadata))

# ----------------
# --- ANALYSIS ---
# ----------------
analyzed_data = analyze_data(raw_data)

# create new asset for analyzed data and save to file
analyzed_data_path = db.add_asset(
    name="Analyzed data",
    type="analyzed_data",
    metadata={
        "str_metadata": "my_string_value",
        "num_metadata": 0,
        "bool_metadata": true
    }
)

analyzed_data.save_to_file(analyzed_data_path)