Feature Request: Support multiple domains in istio hosts

Feature Request

Why

It would be nice if I could support multiple hosts in the istio: hosts declaration. This would be most easily accomplished by doing the following I think...

packages:
  podinfo:
    enabled: true
    git:
      repo: https://github.com/stefanprodan/podinfo.git
      tag: 6.3.4
      path: charts/podinfo
    wrapper:
      enabled: true
    sso:
      enabled: true
    istio:
      injection: "enabled"
      peerAuthentications:
        - name: "podinfo"
          spec:
            selector:
              matchLabels:
                app: "podinfo"
            mtls:
              mode: STRICT
            portLevelMtls:
              9898:
                mode: STRICT
      hosts:
        - names:
            - podinfo
          domains:
            - test.local
            - bigbang.local
          gateways:
            - public
          destination:
            port: 9898

Ideally we would also have backwards compatibility for domain:

Proposed Solution

I have the following solution which I tested and appears working

# chart/templates/istio/virtualservice.yaml

{{- /* Sets up virtual service for istio ingress */ -}}
{{- if and (dig "istio" "hosts" false .Values.package) .Values.bigbang.istio.enabled -}}
{{- range $host := .Values.package.istio.hosts -}}
apiVersion: networking.istio.io/v1beta1
kind: VirtualService
metadata:
  name: {{ include "resourceName" (printf "%s-%s" (index $host.names 0) (default $.Values.bigbang.domain $host.domain)) }}
  labels:
    {{- include "commonLabels" $ | nindent 4 }}
spec:
  gateways:
  {{- range $host.gateways }}
  - {{ ternary . (printf "istio-system/%s" .) (contains "/" .) }}
  {{- end }}
  hosts:
  {{- range $hostNames := $host.names }}
  {{- range $hostDomains := (default (list $host.domain) ($host.domains | default dict)) }}
  - {{ printf "%s.%s" $hostNames (default $.Values.bigbang.domain $hostDomains) | quote }}
  {{- end -}}
  {{- end -}}

  {{- /* Sets up default or simple route */ -}}
  {{- if or $host.destination (not (or (dig "route" "http" false $host) (dig "route" "tcp" false $host) (dig "route" "tls" false $host))) -}}
  {{- $dest := dict "host" (default $.Values.package.name (dig "destination" "service" nil $host)) -}}
  {{- if dig "destination" "port" false $host -}}
  {{- $dest = merge (dict "port" (dict "number" $host.destination.port)) $dest -}}
  {{- end -}}
  {{- $host = merge (dict "route" (dict "http" (append (dig "route" "http" list $host) (dict "route" (list (dict "destination" $dest)))))) $host -}}
  {{- end -}}

  {{- toYaml $host.route | nindent 2 }}
  exportTo:
  - istio-system
---
{{ end -}}
{{- end -}}

Here, we loop over host.names like normal, but then also loop over host.domains inside. If host.domains exists then we loop over that, if not, we turn host.domain into a single item list and loop over there.

The end result is the following virtual service

apiVersion: networking.istio.io/v1beta1
kind: VirtualService
metadata:
  annotations:
    meta.helm.sh/release-name: podinfo-wrapper
    meta.helm.sh/release-namespace: podinfo
  creationTimestamp: "2023-10-28T03:18:34Z"
  generation: 3
  labels:
    app.kubernetes.io/instance: podinfo-wrapper
    app.kubernetes.io/managed-by: Helm
    app.kubernetes.io/name: podinfo
    app.kubernetes.io/version: 0.4.1_33292cc4fec1
    helm.sh/chart: wrapper-0.4.1_33292cc4fec1
    helm.toolkit.fluxcd.io/name: podinfo-wrapper
    helm.toolkit.fluxcd.io/namespace: podinfo
  name: podinfo-test-local
  namespace: podinfo
  resourceVersion: "25343955"
  uid: fd8d1724-43c6-4442-ac5e-801f68d3672f
spec:
  exportTo:
  - istio-system
  gateways:
  - istio-system/public
  hosts:
  - podinfo.test.local
  - podinfo.bigbang.local
  http:
  - route:
    - destination:
        host: podinfo
        port:
          number: 9898

Another route would be to create two virtualservices, however that would cause conflicts with things like the Gateway NetworkPolicy which I honestly didn't feel like untangling. If that is the better solution I would be more than happy to look into that though.