Compare commits

..

17 Commits

Author SHA1 Message Date
lebaudantoine e3b954ff97 🚸(frontend) show two initials in the Avatar when possible
Display two initials in the Avatar whenever the participant's name
allows it, instead of a single letter.

A single initial makes it too hard to distinguish participants when
their cameras are off, especially in larger
2026-07-28 23:36:57 +02:00
lebaudantoine 0ec8ac806f 💄(frontend) improve participant name rendering in the list
Rework how the participant name is displayed in the participant
list to show as much of the name as possible before truncating.

When the name has to be truncated, add a tooltip so users can hover
to see the full name.

Requested by users.
2026-07-28 23:36:55 +02:00
lebaudantoine 02d4191854 (frontend) introduce an "unauthenticated" participant badge
Add a visual badge on participants who are not authenticated, so it
is immediately clear who could be an anonymous participant. This is
a small but explicit security signal in the participant list.

Beyond that, the badge also plays a functional role: since only
authenticated participants can be promoted or demoted, the badge
helps users see at a glance who is eligible for a role change.
2026-07-28 23:36:14 +02:00
lebaudantoine 2623005c7e (frontend) notify user when their meeting role changes
Show a notification to the user whenever their role in the meeting
changes, so they immediately see when they have been promoted or
demoted.
2026-07-28 23:36:14 +02:00
lebaudantoine a70376c5ec 🐛(frontend) fall back to user.full_name on request-entry
Since the username refactoring, the username in the store could be
undefined when the join input was pre-filled from user.full_name,
because no keystroke was needed to populate the store.

This led to a 400 error on the request-entry endpoint whenever the
user joined without editing the pre-filled name.

Fall back to user.full_name when the store username is missing, so
the endpoint always receives a value.

Acknowledged as a somewhat wobbly fix, but ships as-is until the
underlying flow is reworked.
2026-07-28 23:36:14 +02:00
lebaudantoine 1306c13f28 (frontend) close admin side panel when the user is demoted
Listen to role changes in the admin panel and close the side panel
if the current user is demoted while it is open. Without this,
unprivileged users could still see the admin side panel until they
closed it manually.

I checked the other features that could be affected by hot role
changes; this was the only one still exposing admin-only UI after a
demotion. Everything else already handles live permission updates
correctly.
2026-07-28 23:36:14 +02:00
lebaudantoine 14b23d9e7b (frontend) allow promoting authenticated participants
Introduce a new feature that lets a user promote one of the
authenticated participants of the meeting to a role with additional
privileges.

Known limitations:

* Only authenticated participants can be promoted, but there is no
  visual indicator yet distinguishing authenticated from anonymous
  participants. This will be added in a follow-up commit.
* The resource_access data fetched in the initial API call becomes
  stale after a promotion. It is not currently used in the product,
  so this is not visible, but it should either be refreshed later
  or removed from the initial fetch.
* Demoting a promoted user turns them into a member, which is still
  a privileged role. This is a deliberate choice until we introduce
  finer-grained tuning of participant roles.
2026-07-28 23:36:13 +02:00
lebaudantoine c8fc153f19 ♻️(frontend) extract closeSidePanel action at the store level
Extract the logic that closes the side panel into a utility function
declared at the store module level, as recommended by Valtio.

This avoids re-creating the function on every render and prevents
extra re-renders in components that use it.
2026-07-28 23:36:13 +02:00
lebaudantoine 12eba9dcb8 🐛(backend) allow any string as sub in the API serializer
The API serializer was too restrictive on the `sub` field, expecting
a UUID. This worked in our development and production setups because
our Keycloak is configured to emit UUID subs, but it broke for other
providers.

Per the OIDC spec and the DB model, `sub` can be any string. Align
the serializer with this and accept arbitrary string values.

Fixes #1525.
2026-07-28 23:36:11 +02:00
lebaudantoine 200bcc2655 💄(frontend) render Avatar initials in uppercase
Uppercase the initials rendered in the Avatar so their vertical
centering stays consistent.

With lowercase letters, the initials were slightly shifted toward
the bottom of the Avatar, which broke the alignment.
2026-07-28 23:35:32 +02:00
lebaudantoine b1a4c298fa 🔥(frontend) remove leftover console.count call
Drop a stray console.count call that was accidentally committed in a
previous PR and had been left in the codebase.
2026-07-28 23:34:20 +02:00
lebaudantoine e19bf601c1 ♻️(frontend) derive is_administrable from participant metadata
The is_administrable flag was previously read from the room API
response through the room serializer, giving the frontend static
information about the user's rights.

Refactor the frontend so it derives this flag from the participant
role carried in the participant metadata instead.

Two benefits:

* The flag now updates live along with the participant
  attributes/metadata, so role changes are reflected immediately.
* It removes the duplication between the API response and the
  metadata, which both used to determine the user's capabilities.
2026-07-28 23:34:19 +02:00
lebaudantoine d73be18378 (frontend) add client for the update participant role endpoint
Add the frontend client that calls the update participant role
endpoint. Straightforward API call, no special handling.
2026-07-28 23:34:19 +02:00
lebaudantoine bbe26adf69 (backend) expose is_authenticated in the LiveKit token
Include the is_authenticated flag on the user in the LiveKit token
and participant metadata.

The frontend needs this information (used in the next commit) to
know whether it can offer to promote a user with access to the room
admin.
2026-07-28 23:34:19 +02:00
lebaudantoine 5b23c1f5b2 ♻️(backend) pass the participant role in the LiveKit token
The backend previously passed an abstract is_admin_or_owner boolean
flag in the LiveKit token. That kept the frontend minimalistic and
saved it from having to handle role comparisons.

As we introduce more features that need to distinguish between the
room owner and admins, refactor the token to carry the role
directly. The frontend can then derive the relevant flags from a
richer piece of information.
2026-07-28 23:34:18 +02:00
lebaudantoine 13cee8d276 (backend) add endpoint to update a participant role during a meeting
Add an endpoint that allows updating a user's role while in a
meeting. The goal is to let users promote other connected
participants to admin or moderator, so the burden of administrating
a meeting can be shared.
2026-07-28 23:34:18 +02:00
lebaudantoine 725aaa423b ️(backend) add permission class checking the user is in the call
Introduce a new permission class that verifies the caller making a
request is both authenticated and actually present in the call.

It will be used to gate actions that require the user to be live in
the room, for example:

* allowing someone in from the waiting room
* promoting another participant to a different role

More generally, this covers every action where, for security
reasons, we need to make sure the user is truly present in the call
and that someone is not reusing their cookie as an API key.
2026-07-28 23:34:18 +02:00
4 changed files with 0 additions and 215 deletions
-6
View File
@@ -389,12 +389,6 @@ build-k8s-cluster: \
./bin/start-kind.sh
.PHONY: build-k8s-cluster
build-k8s-cluster-orbstack: ## setup the kubernetes environment on OrbStack's built-in cluster (macOS)
build-k8s-cluster-orbstack: \
env.d/development/kube-secret
./bin/start-orbstack.sh
.PHONY: build-k8s-cluster-orbstack
start-tilt-keycloak: ## start the kubernetes cluster using kind, without Pro Connect for authentication, use keycloak
DEV_ENV=dev-keycloak tilt up --namespace=meet -f ./bin/Tiltfile
.PHONY: build-k8s-cluster
-6
View File
@@ -1,11 +1,5 @@
load('ext://uibutton', 'cmd_button', 'bool_input', 'location')
load('ext://namespace', 'namespace_create', 'namespace_inject')
# OrbStack's built-in cluster (macOS) is a supported alternative to kind.
# Recent Tilt versions (>= 0.33) detect it as a local dev cluster; this is
# a no-op for kind and a safety net for older Tilt versions.
allow_k8s_contexts('orbstack')
namespace_create('meet')
DEV_ENV = os.getenv('DEV_ENV', 'dev-keycloak')
-182
View File
@@ -1,182 +0,0 @@
#!/usr/bin/env bash
#
# Bootstrap the local dev environment on OrbStack's built-in Kubernetes
# cluster (macOS) instead of kind.
#
# This replicates what bin/start-kind.sh (numerique-gouv/tools
# kind/create_cluster.sh) provides, minus what OrbStack makes unnecessary:
# - no kind cluster: OrbStack ships a lightweight single-node cluster
# - no local registry (kind-registry): OrbStack's cluster shares the
# Docker image store, so images built by Tilt are directly visible
# to pods. Tilt detects the "orbstack" context as a local cluster
# and skips pushing images entirely.
#
# Requirements: OrbStack (with Kubernetes enabled), kubectl, mkcert, curl.
set -o errexit
APPLICATION=${1:-meet}
CONTEXT="orbstack"
echo "0. Check OrbStack Kubernetes is available"
if ! command -v mkcert >/dev/null 2>&1; then
echo "❌ mkcert is not installed. Install it first: brew install mkcert"
exit 1
fi
if ! kubectl config get-contexts -o name | grep -qx "${CONTEXT}"; then
echo "Context '${CONTEXT}' not found. Trying to start OrbStack Kubernetes..."
if command -v orb >/dev/null 2>&1; then
orb start k8s
else
echo "❌ Enable Kubernetes in OrbStack (Settings > Kubernetes) and retry."
exit 1
fi
fi
kubectl config use-context "${CONTEXT}"
echo "0b. Check ports 80/443 are free on localhost"
# OrbStack forwards LoadBalancer service ports to 127.0.0.1. If the kind
# cluster is still running, its docker proxy already holds 80/443.
# Skip the check if ingress-nginx is already installed here: in that case
# the listener on 80/443 is our own LoadBalancer.
if ! kubectl -n ingress-nginx get deployment ingress-nginx-controller >/dev/null 2>&1; then
for port in 80 443; do
if lsof -nP -iTCP:"${port}" -sTCP:LISTEN >/dev/null 2>&1; then
echo "❌ Port ${port} is already in use on the host."
echo " If the kind cluster is running, delete it first:"
echo " kind delete cluster --name suite"
exit 1
fi
done
fi
echo "1. Create ca"
CURRENT_DIR=$(pwd)
mkcert -install
cd /tmp
mkcert "127.0.0.1.nip.io" "*.127.0.0.1.nip.io"
cd "${CURRENT_DIR}"
echo "2. Install ingress-nginx (cloud provider: LoadBalancer service)"
# OrbStack exposes LoadBalancer services on 127.0.0.1, so the cloud
# manifest replaces kind's hostPort-based deploy. Every sub-step below is
# guarded individually so the script is safe to re-run after a partial
# failure (unlike the upstream kind script, which guards the whole block
# on namespace existence).
# Make sure no stale registry configmap tells Tilt to push to localhost:5001
# (there is no registry on OrbStack).
kubectl -n kube-public delete configmap local-registry-hosting --ignore-not-found
if ! kubectl -n ingress-nginx get deployment ingress-nginx-controller >/dev/null 2>&1; then
kubectl apply -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/main/deploy/static/provider/cloud/deploy.yaml
fi
if ! kubectl -n ingress-nginx get deployment nginx-errors >/dev/null 2>&1; then
kubectl apply -n ingress-nginx -f https://raw.githubusercontent.com/kubernetes/ingress-nginx/refs/heads/main/docs/examples/customization/custom-errors/custom-default-backend.yaml
fi
kubectl -n ingress-nginx create secret tls mkcert --key /tmp/127.0.0.1.nip.io+1-key.pem --cert /tmp/127.0.0.1.nip.io+1.pem || echo ok
# The meet charts render Ingresses without ingressClassName. The kind
# provider manifest handles this via --watch-ingress-without-class=true;
# the cloud manifest does not, so add it here (otherwise: 404 everywhere).
if ! kubectl -n ingress-nginx get deployment ingress-nginx-controller -o jsonpath='{.spec.template.spec.containers[0].args}' | grep -q 'watch-ingress-without-class'; then
kubectl -n ingress-nginx patch deployments.apps ingress-nginx-controller --type 'json' -p '[{"op": "add", "path": "/spec/template/spec/containers/0/args/-", "value":"--watch-ingress-without-class=true"},{"op": "add", "path": "/spec/template/spec/containers/0/args/-", "value":"--default-ssl-certificate=ingress-nginx/mkcert"},{"op": "add", "path": "/spec/template/spec/containers/0/args/-", "value":"--default-backend-service=ingress-nginx/nginx-errors"}
]'
fi
if ! kubectl -n ingress-nginx get deployment nginx-errors -o jsonpath='{.spec.template.spec.containers[0].image}' | grep -q 'error-pages'; then
kubectl -n ingress-nginx patch deployment nginx-errors --type=json -p='[
{"op": "replace", "path": "/spec/template/spec/containers/0/image", "value": "ghcr.io/tarampampam/error-pages:3.3.0"},
{"op": "add", "path": "/spec/template/spec/containers/0/env", "value": [{"name": "TEMPLATE_NAME", "value": "ghost"}, {"name": "SHOW_DETAILS", "value": "false"}, {"name": "SEND_SAME_HTTP_CODE", "value": "true"}]}
]'
fi
cat <<EOF | kubectl apply -n ingress-nginx -f -
apiVersion: v1
data:
allow-snippet-annotations: "true"
annotations-risk-level: Critical
custom-http-errors: 500,501,502,503,504
kind: ConfigMap
metadata:
name: ingress-nginx-controller
namespace: ingress-nginx
EOF
echo "2b. Wait for the ingress controller to be ready"
kubectl -n ingress-nginx rollout status deployment/ingress-nginx-controller --timeout=180s
echo "3. Patch CoreDNS so in-cluster pods resolve *.127.0.0.1.nip.io to the ingress"
# nip.io resolves to 127.0.0.1, which inside a pod is the pod itself.
# Rewrite these names to the ingress-nginx service, like the kind setup does.
# Unlike kind, we amend OrbStack's existing Corefile instead of replacing it.
if ! kubectl -n kube-system get configmap coredns -o jsonpath='{.data.Corefile}' | grep -q '127\.0\.0\.1\.nip\.io'; then
kubectl -n kube-system get configmap coredns -o jsonpath='{.data.Corefile}' \
| awk '/forward \./ && !done { print " rewrite stop {"; print " name regex (.*).127.0.0.1.nip.io ingress-nginx-controller.ingress-nginx.svc.cluster.local answer auto"; print " }"; done=1 } { print }' \
>/tmp/Corefile.orbstack
kubectl -n kube-system create configmap coredns --from-file=Corefile=/tmp/Corefile.orbstack --dry-run=client -o yaml | kubectl apply -f -
kubectl -n kube-system rollout restart deployments/coredns
fi
if ! kubectl get ns "${APPLICATION}" >/dev/null 2>&1; then
echo "4. Setup namespace"
kubectl create ns "${APPLICATION}"
fi
kubectl config set-context --current --namespace="${APPLICATION}"
kubectl -n "${APPLICATION}" create secret generic mkcert --from-file=rootCA.pem="$(mkcert -CAROOT)/rootCA.pem" || echo ok
if ! kubectl get configmap certifi -n "${APPLICATION}" >/dev/null 2>&1; then
echo "5. Inject our custom CA in a configmap for certifi"
curl https://raw.githubusercontent.com/certifi/python-certifi/refs/heads/master/certifi/cacert.pem -o /tmp/cacert.pem
cat "$(mkcert -CAROOT)/rootCA.pem" >>/tmp/cacert.pem
kubectl -n "${APPLICATION}" create configmap certifi --from-file=cacert.pem=/tmp/cacert.pem
kubectl -n "${APPLICATION}" create secret generic certifi --from-file=/tmp/cacert.pem || echo ok
fi
echo "5b. Smoke test: the ingress chain answers on https://127.0.0.1"
# Before Tilt deploys the app this returns the styled 404 from the default
# backend — that still proves LB -> controller works. 000 means the
# LoadBalancer is not bound to localhost.
HTTP_CODE=$(curl -sk -o /dev/null -w '%{http_code}' --max-time 10 https://127.0.0.1/ || true)
if [ "${HTTP_CODE}" = "000" ]; then
echo "⚠️ Nothing answered on https://127.0.0.1 — check the LoadBalancer:"
echo " kubectl -n ingress-nginx get svc ingress-nginx-controller"
else
echo "✅ Ingress reachable (HTTP ${HTTP_CODE})"
fi
echo "6. Check pod readiness across all namespaces..."
sleep_interval=10
echo "Initial wait time: $((sleep_interval * 2)) seconds…"
sleep $((sleep_interval * 2))
check_pods_ready() {
local max_attempts=60 # Maximum number of attempts (10 minutes with 10s intervals)
local attempt=1
while [ $attempt -le $max_attempts ]; do
echo "Attempt $attempt/$max_attempts - Checking pod status..."
not_ready_count=$( kubectl get po -A --no-headers | grep -v -E "Running|Completed"| wc -l | tr -d ' ')
if [ "$not_ready_count" -eq 0 ]; then
echo "✅ All pods are ready!"
return 0
else
echo "$not_ready_count pod(s) still not ready. Waiting $sleep_interval seconds…"
sleep $sleep_interval
((attempt++))
fi
done
echo "❌ Timeout: Some pods are still not ready after 10 minutes"
echo "Final pod status:"
kubectl get po -A
return 1
}
if check_pods_ready; then
echo "🎉 Cluster is fully ready!"
else
echo "⚠️ Some pods may need manual intervention"
exit 1
fi
-21
View File
@@ -143,24 +143,3 @@ $ make start-tilt-keycloak
```
Monitor Tilts progress at [http://localhost:10350/](http://localhost:10350/). After Tilt actions finish, you can access the app at [https://meet.127.0.0.1.nip.io/](https://meet.127.0.0.1.nip.io/).
### Alternative: OrbStack's built-in Kubernetes (macOS)
If you use [OrbStack](https://orbstack.dev/) on macOS, you can run the stack on its built-in Kubernetes cluster instead of kind. It uses noticeably less RAM (no nested kubeadm node container) and no local registry is needed: OrbStack's cluster shares the Docker image store, so Tilt uses images directly without pushing.
Enable Kubernetes in OrbStack (Settings > Kubernetes), then:
```shellscript
$ make build-k8s-cluster-orbstack
```
This installs ingress-nginx (exposed by OrbStack on `127.0.0.1:80/443`), the mkcert TLS certificates, and the CoreDNS rewrite for `*.127.0.0.1.nip.io`, then you start Tilt as usual:
```shellscript
$ make start-tilt-keycloak
```
Notes:
- Ports 80/443 must be free: delete the kind cluster first if you used it (`kind delete cluster --name suite`).
- If you "Reset Kubernetes" in OrbStack, re-run `make build-k8s-cluster-orbstack`.
- kind remains the reference setup (matches CI and lets you pin the Kubernetes version).