omero-ruletagger 1.0.0

Define rules that create tags in an omero server.

OMERO.RuleTagger (Previously Known as OMERO.Autotagger)

A Python package for automated tagging of OMERO objects based on customizable rules and regex.

It works by walking the omero model as described by the omero-py wrappers and the underlying ObjectI classes. After the walk it compares a given reference value against the value it walked to, then tags if the condition passes.

Known Limitations

• Currently impossible to add arguments to the model walk
  • For example a you can't filter a condition to only check annotations in a certain namespace
  • This example could be fixed by comparing the 'ns' property against the desired value, but it still won't filter
• Cannot specify objects to operate on within the rules file.
  • Planned addition in schemav2 but wanted to get this out and get some feedback for schemav2
• Doesn't handle Annotations very well.
  • This is because the omero-py package doesn't have getters for specific annotation types
    • This causes AttributeErrors
  • Map and Table Annotations especially need extra thought to be able to meaningfully rule against
• Also doesn't handle properties
  • Due to the compilation process, we need getters to access properties
  • Could be possible to add with some lambda
    • Decided to leave as is for now because most properties have a getter (thanks java!)
• Pretty slow and inefficient
  • Makes more requests than it needs to
  • Causes it to take longer than it needs to
  • Decided to leave as is for readability sake
    • We run it once a night, could probably do it once a week
    • Both walking and recursion can get complex, tried my best to stay legible (unlike my first implementation)

Documentation

For a proper in-depth explanation of the tool use the documentation.

Installation

Install using pip:

pip install omero-ruletagger

Dependencies

• Python ≥3.8
• PyYAML
• inflect
• omero-py

Usage

Command Line Interface

The package provides three main commands:

1. Validate rules:

omero-ruletagger validate rules.yml

2. Apply tags:

omero-ruletagger run rules.yml -O "Image:123" -O "Dataset:456"

3. Dry run:

omero-ruletagger dry-run rules.yml -O "Image:789" -o results.csv

Connection Options

• -s, --server: OMERO server hostname
• -p, --port: OMERO server port
• -u, --user: OMERO username
• -w, --password: OMERO password
• -k, --key: Existing session key
• -S, --secure: Use secure connection
• --sudo: Connect as administrator
• -v, --verbose: Enable verbose logging

Rule Configuration

Create a YAML file defining your tagging rules. Example:

# test_rules.yml
- capture: "([^-.]+)"
  blacklist:
    - 1

- name: "FALSE"
  rules:
    - attribute_path: ["image", "roicount"]
      operation: lt
      value: 1

- name: "TRUE"
  rules:
    - attribute_path: ["image", "roi", "count"]
      operation: gt
      value: 0

- name: "Subtractive"
  absolute: false
  type: subtractive
  rules:
    - attribute_path: ["image", "roi","count"]
      operation: gt
      value: 0

- name: "UselessAND_TRUE"
  rules:
    - attribute_path: ["image", "roi", "shape", "strokecolor"]
      operation: eq
      value: 255
    - attribute_path: ["image", "roi", "shape", "count"]
      operation: eq
      value: 1

- name: "InherentlyFalseAND"
  rules:
    - attribute_path: ["image", "roi", "shape", "strokecolor"]
      operation: eq
      value: 255
    - attribute_path: ["image", "roi", "shape", "strokecolor"]
      operation: eq
      value: 254

- name: "DescriptionTest"
  rules:
    - attribute_path: ["image", "descriptions"]
      operation: match
      value: ".*test.*"

- name: "PhysicalSizeTest"
  rules:
    - attribute_path: ["image", "primarypixel", "physicalsizex"]
      operation: eq
      value: 1000
    - attribute_path: ["image", "primarypixel", "physicalsizey"]
      operation: eq
      value: 1000
    - attribute_path: ["image", "primarypixel", "physicalsizex", "unit", "name"]
      operation: eq
      value: "MICROMETER"

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE.txt file for details.

Links

• Documentation
• Source Code
• Issue Tracker