Merge branch 'main' into fix-wda-context-was-getting-cancelled

Author: gmegidish

CHANGELOG.md

@@ -1,12 +1,16 @@
-## [0.0.52](https://github.com/mobile-next/mobilecli/releases/tag/0.0.52) (2026-01-26)
+## [0.0.53](https://github.com/mobile-next/mobilecli/releases/tag/0.0.53) (2026-01-28)
 
-* General: Support for jsonrpc over websocket through /ws endpoint ([#154](https://github.com/mobile-next/mobilecli/pull/154))
 * iOS: Fixed a bug where tcp forwarders were created on every request for mjpeg ([#158](https://github.com/mobile-next/mobilecli/pull/158))
 * iOS: Lazily creating mjpeg tcp forwarder only when needed ([#158](https://github.com/mobile-next/mobilecli/pull/158))
 * iOS: Multiple real devices would overwrite each other's wda tcp forwader ([#158](https://github.com/mobile-next/mobilecli/pull/158))
 * iOS: Keep running WebdriverAgent xcuitest as long as "mobilecli server" is running ([#158](https://github.com/mobile-next/mobilecli/pull/158))
 * iOS: Graceful shutdown and improved tunnel closing ([#158](https://github.com/mobile-next/mobilecli/pull/158))
 
+## [0.0.52](https://github.com/mobile-next/mobilecli/releases/tag/0.0.52) (2026-01-27)
+* General: Websocket support for jsonrpc at /ws, single connection multiple requests ([#154](https://github.com/mobile-next/mobilecli/pull/154))
+* iOS: Detect devicekit-ios process already running to Start Broadcasting a 2nd time ([#156](https://github.com/mobile-next/mobilecli/pull/156))
+* iOS: Return more elements in "dump ui", eg Button with no label ([#155](https://github.com/mobile-next/mobilecli/pull/155))
+
 ## [0.0.51](https://github.com/mobile-next/mobilecli/releases/tag/0.0.51) (2026-01-25)
 
 * General: Plenty of code cleanups and refactoring for smaller and more maintained code ([#144](https://github.com/mobile-next/mobilecli/pull/144))

README.md

@@ -26,17 +26,16 @@ A universal command-line tool for managing iOS and Android devices, simulators,
   </a>
 </p>
 
-
 ## Features 🚀
 
 - **Device Management**: List, manage, interactive with connected mobile devices
 - **Cross-Platform Support**: Works with iOS physical devices, iOS simulators, Android devices, and Android emulators
 - **Emulator/Simulator Control**: Boot and shutdown emulators and simulators programmatically
 - **Screenshot Capture**: Take screenshots from any connected device with format options
 - **Multiple Output Formats**: Save screenshots as PNG or JPEG with quality control
-- **Screencapture video streaming**: Stream mjpeg video directly from device
+- **Screencapture video streaming**: Stream mjpeg/h264 video directly from device
 - **Device Control**: Reboot devices, tap screen coordinates, press hardware buttons
-- **App management**: Launch app, terminate apps. Install and uninstall coming next ⏭️
+- **App Management**: Launch app, terminate apps, install and uninstall
 
 ## Installation 📦
 
@@ -178,6 +177,36 @@ mobilecli io text --device <device-id> 'hello world'
 - `VOLUME_UP`, `VOLUME_DOWN` - Volume up and down
 - `DPAD_UP`, `DPAD_DOWN`, `DPAD_LEFT`, `DPAD_RIGHT`, `DPAD_CENTER` - D-pad controls (Android only)
 
+## HTTP API 🔌
+
+***mobilecli*** provides an http interface for all the functionality that is available through command line. As a matter of fact, it is preferable to
+use mobilecli as a webserver, so it can cache and keep tunnels alive, speeding up your interactions with the mobile device or simulator/emulator.
+
+```bash
+# Start the server (default port 12000)
+mobile server start
+
+curl http://localhost:12000/rpc -XPOST -d '{"jsonrpc":"2.0", "id": 1, "method": "devices", "params": {}}'
+curl http://localhost:12000/rpc -XPOST -d '{"jsonrpc":"2.0", "id": 1, "method": "screenshot", "params": {"deviceId": "your-device-id"}}'
+
+## WebSocket Support 🔌
+
+***mobilecli*** includes a WebSocket server that allows multiple requests over a single connection using the same JSON-RPC 2.0 format as the HTTP API.
+
+```bash
+# Start the server (default port 12000)
+mobilecli server start
+
+# Connect and send requests using wscat
+wscat -c ws://localhost:12000/ws
+> {"jsonrpc":"2.0","id":1,"method":"devices","params":{}}
+< {"jsonrpc":"2.0","id":1,"result":[...]}
+> {"jsonrpc":"2.0","id":2,"method":"screenshot","params":{"deviceId":"your-device-id"}}
+< {"jsonrpc":"2.0","id":2,"result":{...}}
+```
+
+**Note**: `screencapture` is not supported over WebSocket - use the HTTP `/rpc` endpoint for video streaming.
+
 ## Platform-Specific Notes
 
 ### iOS Real Devices

devices/wda/active-app.go

@@ -1,38 +1,51 @@
 package wda
 
-import "fmt"
+import (
+	"encoding/json"
+	"fmt"
+)
+
+// activeAppValue represents the value field from the WDA active app response
+type activeAppValue struct {
+	BundleID string  `json:"bundleId"`
+	Name     string  `json:"name"`
+	PID      float64 `json:"pid"`
+}
+
+// activeAppResponse represents the WDA API response for active app info
+type activeAppResponse struct {
+	Value activeAppValue `json:"value"`
+}
 
 // ActiveAppInfo represents information about the currently active application
 type ActiveAppInfo struct {
-	BundleID string `json:"bundleId"`
-	Name     string `json:"name"`
-	ProcessID int   `json:"pid"`
+	BundleID  string `json:"bundleId"`
+	Name      string `json:"name"`
+	ProcessID int    `json:"pid"`
 }
 
 // GetActiveAppInfo returns information about the currently active/foreground application
 // This uses the /wda/activeAppInfo endpoint which doesn't require a session
 func (c *WdaClient) GetActiveAppInfo() (*ActiveAppInfo, error) {
-	response, err := c.GetEndpoint("wda/activeAppInfo")
+	responseMap, err := c.GetEndpoint("wda/activeAppInfo")
 	if err != nil {
 		return nil, fmt.Errorf("failed to get active app info: %w", err)
 	}
 
-	// extract value from response
-	value, ok := response["value"].(map[string]interface{})
-	if !ok {
-		return nil, fmt.Errorf("unexpected response format: missing or invalid 'value' field")
+	// marshal the response back to json and unmarshal into typed struct
+	jsonData, err := json.Marshal(responseMap)
+	if err != nil {
+		return nil, fmt.Errorf("failed to marshal response: %w", err)
 	}
 
-	bundleID, _ := value["bundleId"].(string)
-	name, _ := value["name"].(string)
-	pid := 0
-	if pidFloat, ok := value["pid"].(float64); ok {
-		pid = int(pidFloat)
+	var response activeAppResponse
+	if err := json.Unmarshal(jsonData, &response); err != nil {
+		return nil, fmt.Errorf("failed to unmarshal response: %w", err)
 	}
 
 	return &ActiveAppInfo{
-		BundleID:  bundleID,
-		Name:      name,
-		ProcessID: pid,
+		BundleID:  response.Value.BundleID,
+		Name:      response.Value.Name,
+		ProcessID: int(response.Value.PID),
 	}, nil
 }

go.mod

@@ -4,6 +4,7 @@ go 1.25.0
 
 require (
 	github.com/danielpaulus/go-ios v1.0.182
+	github.com/gorilla/websocket v1.5.3
 	github.com/sirupsen/logrus v1.9.3
 	github.com/spf13/cobra v1.9.1
 	github.com/stretchr/testify v1.10.0

go.sum

@@ -28,6 +28,8 @@ github.com/google/pprof v0.0.0-20210407192527-94a9f03dee38 h1:yAJXTCF9TqKcTiHJAE
 github.com/google/pprof v0.0.0-20210407192527-94a9f03dee38/go.mod h1:kpwsk12EmLew5upagYY7GY0pfYCcupk39gWOCRROcvE=
 github.com/google/uuid v1.1.2 h1:EVhdT+1Kseyi1/pUmXKaFxYsDNy9RQYkMWRH68J/W7Y=
 github.com/google/uuid v1.1.2/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
+github.com/gorilla/websocket v1.5.3 h1:saDtZ6Pbx/0u+bgYQ3q96pZgCzfhKXGPqt7kZ72aNNg=
+github.com/gorilla/websocket v1.5.3/go.mod h1:YR8l580nyteQvAITg2hZ9XVh4b55+EU/adAjf1fMHhE=
 github.com/grandcat/zeroconf v1.0.0 h1:uHhahLBKqwWBV6WZUDAT71044vwOTL+McW0mBJvo6kE=
 github.com/grandcat/zeroconf v1.0.0/go.mod h1:lTKmG1zh86XyCoUeIHSA4FJMBwCJiQmGfcP2PdzytEs=
 github.com/ianlancetaylor/demangle v0.0.0-20200824232613-28f6c0f3b639/go.mod h1:aSSvb/t6k1mPoxDqO4vJh6VOCGPwU4O0C2/Eqndh1Sc=

server/server.go

@@ -103,6 +103,7 @@ func StartServer(addr string, enableCORS bool) error {
 
 	mux.HandleFunc("/", sendBanner)
 	mux.HandleFunc("/rpc", handleJSONRPC)
+	mux.HandleFunc("/ws", NewWebSocketHandler(enableCORS))
 
 	// if host is missing, default to localhost
 	if !strings.Contains(addr, ":") {