Schema Validation

Overview

OCPPData.jl includes a runtime validation layer powered by JSONSchema.jl. This lets you validate raw message payloads (as Dict) against the official OCPP JSON schemas before attempting typed deserialization.

This is useful for:

Validating incoming messages from untrusted charge points
Providing clear error diagnostics (missing fields, wrong types, invalid enum values)
Catching malformed messages early in the pipeline

API

validate(spec::AbstractOCPPSpec, action::String, payload::AbstractDict, msg_type::Symbol)

Argument	Values	Example
`spec`	`V16.Spec()` or `V201.Spec()`	`V16.Spec()`
`action`	Action name	`"BootNotification"`
`payload`	Message payload as a Dict	`Dict("chargePointVendor" => "V", ...)`
`msg_type`	`:request` or `:response`	`:request`

Returns: nothing if valid, or a String describing the validation error.

Examples

using OCPPData, OCPPData.V16, OCPPData.V201

Valid Payload

result = validate(
    V16.Spec(),
    "BootNotification",
    Dict("chargePointVendor" => "V", "chargePointModel" => "M"),
    :request,
)
isnothing(result)

true

Missing Required Field

validate(
    V16.Spec(),
    "BootNotification",
    Dict("chargePointVendor" => "V"),
    :request,
)

"Validation failed:\npath:         top-level\ninstance:     Dict(\"chargePointVendor\" => \"V\")\nschema key:   required\nschema value: Any[\"chargePointVendor\", \"chargePointModel\"]\n"

Wrong Field Type

validate(
    V16.Spec(),
    "BootNotification",
    Dict("chargePointVendor" => 123, "chargePointModel" => "M"),
    :request,
)

"Validation failed:\npath:         [chargePointVendor]\ninstance:     123\nschema key:   type\nschema value: string\n"

V16 Response Validation

result = validate(
    V16.Spec(),
    "BootNotification",
    Dict(
        "status" => "Accepted",
        "currentTime" => "2025-01-01T00:00:00Z",
        "interval" => 300,
    ),
    :response,
)
isnothing(result)

true

V201 Validation

result = validate(
    V201.Spec(),
    "BootNotification",
    Dict(
        "reason" => "PowerUp",
        "chargingStation" => Dict("model" => "M", "vendorName" => "V"),
    ),
    :request,
)
isnothing(result)

true

Empty Payload (HeartbeatRequest)

result = validate(V16.Spec(), "Heartbeat", Dict{String,Any}(), :request)
isnothing(result)

true

How It Works

Schemas are lazy-loaded and cached on first use. The first call to validate for a given action reads the JSON schema file from disk and compiles it into a JSONSchema.Schema object. Subsequent calls for the same action reuse the cached schema.

Schema files are resolved based on version-specific naming conventions, encoded via multiple dispatch on the Spec type:

V16: BootNotification.json (request), BootNotificationResponse.json (response)
V201: BootNotificationRequest.json, BootNotificationResponse.json

Validation vs typed parsing

validate checks the raw Dict payload against the JSON schema. It does not convert the payload to a typed struct. Use JSON.parse(json_str, T) for typed deserialization after validation passes.