Renderers (Export Layer)¶

Renderers generate output in diverse formats based on the stored models. They register themselves via @RendererRegistry.register().

Class Hierarchy¶

classDiagram
    class Renderer {
        <<abstract>>
        +render(args, schema)*
        +get_included_columns(args)
    }

    class ModelRenderer {
        +getModels(args, schema)
    }

    class ClassRenderer {
        +getClass(args, schema)
    }

    class JSONRenderer
    class CSVRenderer
    class I18nRenderer
    class XLSXRenderer
    class Jinja2Renderer
    class GGM_MDRenderer
    class JSON_SchemaRenderer
    class TTLRenderer
    class RDFRenderer
    class JSONLDRenderer
    class SQLARenderer
    class EARepoUpdater
    class SchemaDiffMarkdownRenderer

    Renderer <|-- ModelRenderer
    Renderer <|-- JSONRenderer
    Renderer <|-- CSVRenderer
    Renderer <|-- I18nRenderer
    Renderer <|-- XLSXRenderer
    ModelRenderer <|-- Jinja2Renderer
    ModelRenderer <|-- GGM_MDRenderer
    ModelRenderer <|-- JSON_SchemaRenderer
    Renderer <|-- TTLRenderer
    Renderer <|-- RDFRenderer
    Renderer <|-- JSONLDRenderer
    ModelRenderer <|-- SQLARenderer
    Renderer <|-- EARepoUpdater
    Renderer <|-- SchemaDiffMarkdownRenderer

Overview¶

Renderer	Type	File	Output Format
JSONRenderer	`json`	`pandasrenderer.py`	JSON (array of records / indexed)
CSVRenderer	`csv`	`pandasrenderer.py`	CSV per table
I18nRenderer	`i18n`	`pandasrenderer.py`	Translation JSON
XLSXRenderer	`xlsx`	`xlsxrenderer.py`	Excel (.xlsx)
Jinja2Renderer	`jinja2`	`jinja2renderer.py`	Custom template output
GGM_MDRenderer	`ggm_md`	`jinja2renderer.py`	Markdown (GGM format)
JSON_SchemaRenderer	`json_schema`	`jinja2renderer.py`	JSON Schema
TTLRenderer	`ttl`	`lodrenderer.py`	Turtle (RDF)
RDFRenderer	`rdf`	`lodrenderer.py`	RDF/XML
JSONLDRenderer	`jsonld`	`lodrenderer.py`	JSON-LD
SQLARenderer	`sqla`	`sqlarenderer.py`	Python SQLAlchemy code
EARepoUpdater	`ea_repo`	`earepoupdater.py`	Direct EA database update
SchemaDiffMD	`schema_diff_md`	`jinja2renderer.py`	Schema comparison markdown

Tabular Renderers¶

JSON, CSV, XLSX — Pandas-based export with support for:

Column filtering via --output_columns
Key renaming via --mapper
Multiple record types: RECORD_TYPE_RECORD (array) or RECORD_TYPE_INDEXED (object with ID as key)

Template Renderers¶

Jinja2, GGM Markdown, JSON Schema — Based on Jinja2 templates in crunch_uml/templates/:

Template	Application
`ggm_markdown.j2`	Dutch government documentation
`json_schema.j2`	JSON Schema for validation
`ddas_markdown.j2`	DDAS-specific documentation
`ggm_sqlalchemy.j2`	SQLAlchemy model code

HTML-to-Markdown conversion via BeautifulSoup + markdownify.

Linked Data Renderers¶

TTL, RDF, JSON-LD — Via rdflib with namespace support (--linked_data_namespace). Generates RDF/OWL ontologies based on the stored model.

EA Repo Updater¶

Destructive Operations

The EA Repo Updater has direct ODBC access to Enterprise Architect databases. Contains flags for dangerous operations:

--ea_allow_insert — Allow new records
--ea_allow_delete — Allow deletions

Tag strategies: update | upsert | replace

Schema Diff Renderer¶

Compares two schemas via --compare_schema_name and generates a markdown diff report.

CLI Arguments (Export)¶

Argument	Description
`-f / --outputfile`	Path to output file
`-t / --outputtype`	Renderer type
`-pi / --output_package_ids`	Filter on specific packages
`-jt / --output_jinja2_template`	Custom Jinja2 template
`-jtd`	Template directory
`--linked_data_namespace`	Namespace for LOD renderers
`--compare_schema_name`	Schema for diff comparison

Planned Extensions¶

GraphQL Schema Renderer

Generate GraphQL schemas based on the stored model.

OpenAPI Renderer

Generate OpenAPI/Swagger specifications for REST APIs.

Adding a New Renderer¶

from crunch_uml.renderers.renderer import Renderer, RendererRegistry

@RendererRegistry.register("my_format", descr="Custom output")
class MyRenderer(Renderer):
    def render(self, args, schema):
        models = schema.get_all_classes()
        # Generate output
        ...