echo command ignores env variables and resources

[peczenyj]

I was expecting that, running the echo command, it can return the full configuration including the resources and values expanded from env vars.

Example

If I have this .env fime

BUFFER_MEMORY_LIMIT=2048

this config.yaml

input:
  resource: the_input

pipeline:
  processors:
    - mapping: root = content().uppercase()

buffer:
  memory:
    limit: ${BUFFER_MEMORY_LIMIT:1024}

output:
  resource: the_output

and this resources.yaml file

---
input_resources:
  - label: the_input
    stdin: {}

output_resources:
  - label: the_output
    stdout: {}

By running the latest version of redpandadata/connect I got

$ docker run --env-file ./foo/.env --net host -v ./foo/:/c ${IMAGE} echo -r "/c/resources.yaml" /c/config.yaml

http:
  enabled: true
  address: 0.0.0.0:4195
  root_path: /benthos
  debug_endpoints: false
  cert_file: ""
  key_file: ""
  cors:
    enabled: false
    allowed_origins: []
  basic_auth:
    enabled: false
    realm: restricted
    username: ""
    password_hash: ""
    algorithm: sha256
    salt: ""
input:
  resource: the_input
buffer:
  memory:
    limit: 1024           # should be 2048 no?
pipeline:
  threads: -1
  processors:
    - mapping: root = content().uppercase()
output:
  resource: the_output
input_resources: []       # no signal of the_input
processor_resources: []
output_resources: []      # no signal of the_output 
cache_resources: []
rate_limit_resources: []
logger:
  level: INFO
  format: logfmt
  add_timestamp: false
  level_name: level
  timestamp_name: time
  message_name: msg
  static_fields:
    '@service': redpanda-connect
  file:
    path: ""
    rotate: false
    rotate_max_age_days: 0
metrics:
  prometheus: {}
  mapping: ""
tracer:
  none: {}
shutdown_delay: 0s
shutdown_timeout: 20s
tests: []
redpanda:
  seed_brokers: []
  client_id: benthos
  tls:
    enabled: false
    skip_cert_verify: false
    enable_renegotiation: false
    root_cas: ""
    root_cas_file: ""
    client_certs: []
  sasl: []
  metadata_max_age: 5m
  pipeline_id: ""
  logs_topic: ""
  logs_level: info
  status_topic: ""
  idempotent_write: true
  timeout: 10s
  max_message_bytes: 1MB
  broker_write_max_bytes: 100MB

but if I call the endpoint /debug/config/yaml I can retrieve the full configuration.

based on the help message

   This simple command is useful for sanity checking a config if it isn't
   behaving as expected, as it shows you a normalised version after environment
   variables have been resolved:

I should see, at least, a different value in the buffer memory limit, no ?