hafsah2018 / schema_salad

Semantic Annotations for Linked Avro Data

Schema Salad

Salad is a schema language for describing JSON or YAML structured linked data documents. Salad schema describes rules for preprocessing, structural validation, and hyperlink checking for documents described by a Salad schema. Salad supports rich data modeling with inheritance, template specialization, object identifiers, object references, documentation generation, code generation, and transformation to RDF. Salad provides a bridge between document and record oriented data modeling and the Semantic Web.

Usage

$ pip install schema_salad

To install from source:

git clone https://github.com/common-workflow-language/schema_salad
cd schema_salad
python setup.py install

Commands

Schema salad can be used as a command line tool or imported as a Python module:

$ schema-salad-tool
usage: schema-salad-tool [-h] [--rdf-serializer RDF_SERIALIZER]
                      [--print-jsonld-context | --print-rdfs | --print-avro
                      | --print-rdf | --print-pre | --print-index
                      | --print-metadata | --print-inheritance-dot
                      | --print-fieldrefs-dot | --codegen language
                      | --print-oneline]
                      [--strict | --non-strict] [--verbose | --quiet
                      | --debug]
                      [--version]
                      [schema] [document]

$ python
>>> import schema_salad

Validate a schema:

$ schema-salad-tool myschema.yml

Validate a document using a schema:

$ schema-salad-tool myschema.yml mydocument.yml

Generate HTML documentation:

$ schema-salad-tool myschema.yml > myschema.html

Get JSON-LD context:

$ schema-salad-tool --print-jsonld-context myschema.yml mydocument.yml

Convert a document to JSON-LD:

$ schema-salad-tool --print-pre myschema.yml mydocument.yml > mydocument.jsonld

Generate Python classes for loading/generating documents described by the schema:

$ schema-salad-tool --codegen=python myschema.yml > myschema.py

Display inheritance relationship between classes as a graphviz 'dot' file and render as SVG:

$ schema-salad-tool --print-inheritance-dot myschema.yml | dot -Tsvg > myschema.svg

Quick Start

Let's say you have a 'basket' record that can contain items measured either by weight or by count. Here's an example:

basket:
  - product: bananas
    price: 0.39
    per: pound
    weight: 1
  - product: cucumbers
    price: 0.79
    per: item
    count: 3

We want to validate that all the expected fields are present, the measurement is known, and that "count" cannot be a fractional value. Here is an example schema to do that:

- name: Product
  doc: |
    The base type for a product.  This is an abstract type, so it
    can't be used directly, but can be used to define other types.
  type: record
  abstract: true
  fields:
    product: string
    price: float

- name: ByWeight
  doc: |
    A product, sold by weight.  Products may be sold by pound or by
    kilogram.  Weights may be fractional.
  type: record
  extends: Product
  fields:
    per:
      type:
        type: enum
        symbols:
          - pound
          - kilogram
      jsonldPredicate: '#per'
    weight: float

- name: ByCount
  doc: |
    A product, sold by count.  The count must be a integer value.
  type: record
  extends: Product
  fields:
    per:
      type:
        type: enum
        symbols:
          - item
      jsonldPredicate: '#per'
    count: int

- name: Basket
  doc: |
    A basket of products.  The 'documentRoot' field indicates it is a
    valid starting point for a document.  The 'basket' field will
    validate subtypes of 'Product' (ByWeight and ByCount).
  type: record
  documentRoot: true
  fields:
    basket:
      type:
        type: array
        items: Product

You can check the schema and document in schema_salad/tests/basket_schema.yml and schema_salad/tests/basket.yml:

$ schema-salad-tool basket_schema.yml basket.yml
Document `basket.yml` is valid

Documentation

See the specification and the metaschema (salad schema for itself). For an example application of Schema Salad see the Common Workflow Language.

Rationale

The JSON data model is an popular way to represent structured data. It is attractive because of it's relative simplicity and is a natural fit with the standard types of many programming languages. However, this simplicity comes at the cost that basic JSON lacks expressive features useful for working with complex data structures and document formats, such as schemas, object references, and namespaces.

JSON-LD is a W3C standard providing a way to describe how to interpret a JSON document as Linked Data by means of a "context". JSON-LD provides a powerful solution for representing object references and namespaces in JSON based on standard web URIs, but is not itself a schema language. Without a schema providing a well defined structure, it is difficult to process an arbitrary JSON-LD document as idiomatic JSON because there are many ways to express the same data that are logically equivalent but structurally distinct.

Several schema languages exist for describing and validating JSON data, such as JSON Schema and Apache Avro data serialization system, however none understand linked data. As a result, to fully take advantage of JSON-LD to build the next generation of linked data applications, one must maintain separate JSON schema, JSON-LD context, RDF schema, and human documentation, despite significant overlap of content and obvious need for these documents to stay synchronized.

Schema Salad is designed to address this gap. It provides a schema language and processing rules for describing structured JSON content permitting URI resolution and strict document validation. The schema language supports linked data through annotations that describe the linked data interpretation of the content, enables generation of JSON-LD context and RDF schema, and production of RDF triples by applying the JSON-LD context. The schema language also provides for robust support of inline documentation.