organic-express-api-doc v0.3.0

Organelle for generating documentation by runtime reflection of express mounted route handlers and their source code.

dna

{
  "source": "node_modules/organic-express-api-doc",
  "reactOn": ["ExpressServer"],
  "mountOn": "/api-docs",
  "routes": {
    "organic-api-routes": "@processes.index.plasma.organic-api-routes"
  },
  "docsMetadata": {
    "source": "docs/api",
    "autopopulate": false,
    "renderAutopopulatedDocs": false,
    "populateFilename": "api.md"
  },
  "log": false,
  "liveTemplateReload": false,
  "templatePath": String // defaults to path.join(organic-express-api-doc, "../template/index.html")
}

routes property

Its value is used to build organelles responsible for express routes mounting.

All organelles are expected to reactOn and to emitReady. Additionally log: false is applied to their corresponding dna before triggering their build.

docsMetadata property

Indicates additional metadata information by default stored in markdown files where api routes are matched as follows

... any markdown content ...
## METHOD route
... markdown content for api action
...

So for example given the markdown bellow:

# markdown file title

## GET /api/resource

this text will be merged as description for `GET /api/resource` action,
other supported methods are "POST", "PUT", "DELETE" and "ALL"

## PUT /api/resource/:url_param ? :query_params

example markdown metadata description for action with url and query params.

### url params

* :url_param

### query params

* :query_params

Will be extracted metadata information for the following routes:

• GET /api/resource
• PUT /api/resource/:url_param

Notice that methods and urls should exactly match to your mounted express route handlers (case sensitive ;)

autopopulate property

When enabled will intercept all incoming requests and their responses and upon kill chemical will dump at dna.docsMetadata {.source} \ {.populateFilename}.json json data documentation with samples 'schemified'.

renderAutopopulatedDocs property

When enabled will load and render dna.docsMetadata {.source} \ {.populateFilename}.json json data documentation into dna.docsMetadata {.source} \ {.populateFilename} markdown file suitable for parsing by docsMetadata pipeline.

Noteice that this happens when the organelle is constructed, therefore it won't output anything if the json file is missing.

under-the-hood

the concept

sampling routes

0. load sampled routes metadata if exists
1. intercept all incoming http requests and their respective responses
2. convert their properties (headers, body, ...) to a schema-like structures
3. consolidate requests/responses into samples per method : route
4. store sampled routes metadata

structure routes map

1. intercept Express app's mount methods (all, get, post, ...)
2. analyze the handlers passed
3. stored sampled routes metadata is loaded and trasnformed to markdown file
4. load markdown files and transform them to routes metadata
5. generate in-memory documentation structure of analyzed routes forming their respective apis with their routes metadata

documentation rendering

1. mount route handler on Express app
2. respond with in-memory documentation structure transformed as html+css+js to incoming requests