Improve propagation sync and improve markdown rendering

- Added methods to collect and manage propagation sync metrics in ReticulumMeshChat, improving message tracking and delivery confirmation.
- Updated frontend components to display detailed sync status notifications, including stored messages and delivery confirmations.
- Implemented URL safety checks in the markdown renderer to prevent XSS vulnerabilities by sanitizing links and image sources.
- Refactored various comments and code for clarity and maintainability across multiple files.

electron/main-legacy.js

@@ -301,8 +301,7 @@ app.whenReady().then(async () => {
         }
     }
 
-    // find path to python/cxfreeze reticulum meshchatx executable
-    // Note: setup.py creates ReticulumMeshChatX (with X), not ReticulumMeshChat
+    // find path to python/cxfreeze executable (setup.py builds ReticulumMeshChatX)
     const exeName = process.platform === "win32" ? "ReticulumMeshChatX.exe" : "ReticulumMeshChatX";
 
     // get app path (handles both development and packaged app)
@@ -362,7 +361,7 @@ app.whenReady().then(async () => {
         const requiredArguments = [
             "--headless", // reticulum meshchatx usually launches default web browser, we don't want this when using electron
             "--port",
-            "9337", // FIXME: let system pick a random unused port?
+            "9337",
             // '--test-exception-message', 'Test Exception Message', // uncomment to test the crash dialog
         ];
 

electron/main.js

@@ -458,8 +458,7 @@ app.whenReady().then(async () => {
         }
     }
 
-    // find path to python/cxfreeze reticulum meshchatx executable
-    // Note: setup.py creates ReticulumMeshChatX (with X), not ReticulumMeshChat
+    // find path to python/cxfreeze executable (setup.py builds ReticulumMeshChatX)
     const exeName = process.platform === "win32" ? "ReticulumMeshChatX.exe" : "ReticulumMeshChatX";
 
     // get app path (handles both development and packaged app)
@@ -524,7 +523,7 @@ app.whenReady().then(async () => {
         const requiredArguments = [
             "--headless", // reticulum meshchatx usually launches default web browser, we don't want this when using electron
             "--port",
-            "9337", // FIXME: let system pick a random unused port?
+            "9337",
             // '--test-exception-message', 'Test Exception Message', // uncomment to test the crash dialog
         ];
 

meshchatx/meshchat.py

@@ -279,6 +279,7 @@ class ReticulumMeshChat:
         # Multi-identity support
         self.contexts: dict[str, IdentityContext] = {}
         self.current_context: IdentityContext | None = None
+        self._propagation_sync_metrics: dict[str, dict] = {}
 
         self.setup_identity(identity)
         self.web_audio_bridge = WebAudioBridge(None, None)
@@ -819,7 +820,6 @@ class ReticulumMeshChat:
             self._identity_session_id += 1
 
             # Teardown current identity state and managers
-            # This also calls cleanup_rns_state_for_identity
             self.teardown_identity()
 
             # Give loops a moment to finish
@@ -1684,6 +1684,94 @@ class ReticulumMeshChat:
             return
         ctx.message_router.cancel_propagation_node_requests()
 
+    def _get_propagation_sync_metrics(self, context=None):
+        ctx = context or self.current_context
+        if not ctx:
+            return None
+
+        key = ctx.identity_hash
+        if key not in self._propagation_sync_metrics:
+            self._propagation_sync_metrics[key] = {
+                "started_at": None,
+                "baseline_total_messages": 0,
+                "baseline_delivered_messages": 0,
+                "messages_stored": 0,
+                "delivery_confirmations": 0,
+                "messages_hidden": 0,
+            }
+
+        return self._propagation_sync_metrics[key]
+
+    def _begin_propagation_sync_metrics(self, context=None):
+        ctx = context or self.current_context
+        if not ctx or not ctx.database:
+            return
+
+        metrics = self._get_propagation_sync_metrics(context=ctx)
+        if metrics is None:
+            return
+
+        metrics["started_at"] = datetime.now(UTC).isoformat()
+        metrics["baseline_total_messages"] = ctx.database.messages.count_lxmf_messages()
+        metrics["baseline_delivered_messages"] = (
+            ctx.database.messages.count_lxmf_messages_by_state("delivered")
+        )
+        metrics["messages_stored"] = 0
+        metrics["delivery_confirmations"] = 0
+        metrics["messages_hidden"] = 0
+
+    def _collect_propagation_sync_metrics(self, context=None):
+        ctx = context or self.current_context
+        if not ctx or not ctx.database:
+            return {
+                "messages_stored": 0,
+                "delivery_confirmations": 0,
+                "messages_hidden": 0,
+            }
+
+        metrics = self._get_propagation_sync_metrics(context=ctx)
+        if metrics is None:
+            return {
+                "messages_stored": 0,
+                "delivery_confirmations": 0,
+                "messages_hidden": 0,
+            }
+        if metrics["started_at"] is None:
+            return {
+                "messages_stored": 0,
+                "delivery_confirmations": 0,
+                "messages_hidden": 0,
+            }
+
+        messages_received = ctx.message_router.propagation_transfer_last_result or 0
+        current_total_messages = ctx.database.messages.count_lxmf_messages()
+        current_delivered_messages = ctx.database.messages.count_lxmf_messages_by_state(
+            "delivered",
+        )
+
+        messages_stored = max(
+            current_total_messages - metrics["baseline_total_messages"],
+            0,
+        )
+        delivery_confirmations = max(
+            current_delivered_messages - metrics["baseline_delivered_messages"],
+            0,
+        )
+        messages_hidden = max(
+            messages_received - messages_stored - delivery_confirmations,
+            0,
+        )
+
+        metrics["messages_stored"] = messages_stored
+        metrics["delivery_confirmations"] = delivery_confirmations
+        metrics["messages_hidden"] = messages_hidden
+
+        return {
+            "messages_stored": messages_stored,
+            "delivery_confirmations": delivery_confirmations,
+            "messages_hidden": messages_hidden,
+        }
+
     # stops and removes the active propagation node
     def remove_active_propagation_node(self, context=None):
         ctx = context or self.current_context
@@ -2125,12 +2213,7 @@ class ReticulumMeshChat:
         sys.exit(code)
 
     def get_routes(self):
-        # This is a bit of a hack to get the routes without running the full server
-        # It's mainly for testing purposes
         routes = web.RouteTableDef()
-
-        # We need to mock some things that are usually setup in run()
-        # but for just getting the route handlers, we can skip most of it
         self._define_routes(routes)
         return routes
 
@@ -2378,7 +2461,6 @@ class ReticulumMeshChat:
                         )
 
                 result = self.database.restore_database(path)
-                # Note: This might require an app relaunch to be fully effective
                 return web.json_response(
                     {"status": "success", "result": result, "requires_relaunch": True},
                 )
@@ -3263,7 +3345,6 @@ class ReticulumMeshChat:
                     InterfaceEditor.update_value(interface_details, data, "callsign")
                     InterfaceEditor.update_value(interface_details, data, "ssid")
 
-            # FIXME: move to own sections
             # RNode Airtime limits and station ID
             InterfaceEditor.update_value(interface_details, data, "callsign")
             InterfaceEditor.update_value(interface_details, data, "id_interval")
@@ -6055,6 +6136,7 @@ class ReticulumMeshChat:
 
         @routes.get("/api/v1/lxmf/propagation-node/status")
         async def propagation_node_status(request):
+            sync_metrics = self._collect_propagation_sync_metrics()
             return web.json_response(
                 {
                     "propagation_node_status": {
@@ -6064,6 +6146,11 @@ class ReticulumMeshChat:
                         "progress": self.message_router.propagation_transfer_progress
                         * 100,  # convert to percentage
                         "messages_received": self.message_router.propagation_transfer_last_result,
+                        "messages_stored": sync_metrics["messages_stored"],
+                        "delivery_confirmations": sync_metrics[
+                            "delivery_confirmations"
+                        ],
+                        "messages_hidden": sync_metrics["messages_hidden"],
                     },
                 },
             )
@@ -8893,6 +8980,8 @@ class ReticulumMeshChat:
         if not ctx:
             return
 
+        self._begin_propagation_sync_metrics(context=ctx)
+
         # update last synced at timestamp
         ctx.config.lxmf_preferred_propagation_node_last_synced_at.set(int(time.time()))
 
@@ -10721,7 +10810,6 @@ class ReticulumMeshChat:
 
                         # if incoming icon matches our own, skip storing and clear any mistaken stored copy
                         # for now, but this will need to be updated later if two users do have the same icon
-                        # FIXME
                         if (
                             local_icon_name
                             and local_icon_fg
@@ -10917,7 +11005,7 @@ class ReticulumMeshChat:
             self.send_failed_message_via_propagation_node(lxmf_message, context=context)
 
         # update state
-        self.on_lxmf_sending_state_updated(lxmf_message)
+        self.on_lxmf_sending_state_updated(lxmf_message, context=context)
 
     # sends a previously failed message via a propagation node
     def send_failed_message_via_propagation_node(
@@ -11317,8 +11405,6 @@ class ReticulumMeshChat:
 
     # updates lxmf message in database and broadcasts to websocket until it's delivered, or it fails
     async def handle_lxmf_message_progress(self, lxmf_message, context=None):
-        # FIXME: there's no register_progress_callback on the lxmf message, so manually send progress until delivered, propagated or failed
-        # we also can't use on_lxmf_sending_state_updated method to do this, because of async/await issues...
         ctx = context or self.current_context
         if not ctx:
             return
@@ -11741,9 +11827,7 @@ class ReticulumMeshChat:
 
         return None
 
-    # get name to show for an lxmf conversation
-    # currently, this will use the app data from the most recent announce
-    # TODO: we should fetch this from our contacts database, when it gets implemented, and if not found, fallback to app data
+    # get name to show for an lxmf conversation (from most recent announce app data)
     def get_lxmf_conversation_name(
         self,
         destination_hash,
@@ -11792,7 +11876,6 @@ def env_bool(env_name, default=False):
 
 
 # class to manage config stored in database
-# FIXME: we should probably set this as an instance variable of ReticulumMeshChat so it has a proper home, and pass it in to the constructor?
 def main():
     # apply asyncio 3.13 patch if needed
     AsyncUtils.apply_asyncio_313_patch()

meshchatx/src/backend/async_utils.py

@@ -16,9 +16,7 @@ class AsyncUtils:
         if sys.version_info >= (3, 13):
             import asyncio.base_events
 
-            # We need to patch the loop's sendfile to raise NotImplementedError for SSL transports.
-            # This will force aiohttp to use its own fallback which works correctly.
-
+            # Patch sendfile so aiohttp uses its SSL fallback on 3.13.
             original_sendfile = asyncio.base_events.BaseEventLoop.sendfile
 
             async def patched_sendfile(

meshchatx/src/backend/database/messages.py

@@ -86,6 +86,17 @@ class MessageDAO:
     def get_all_lxmf_messages(self):
         return self.provider.fetchall("SELECT * FROM lxmf_messages")
 
+    def count_lxmf_messages(self):
+        row = self.provider.fetchone("SELECT COUNT(*) AS count FROM lxmf_messages")
+        return row["count"] if row and row["count"] is not None else 0
+
+    def count_lxmf_messages_by_state(self, state):
+        row = self.provider.fetchone(
+            "SELECT COUNT(*) AS count FROM lxmf_messages WHERE state = ?",
+            (state,),
+        )
+        return row["count"] if row and row["count"] is not None else 0
+
     def get_conversation_messages(self, destination_hash, limit=100, offset=0):
         return self.provider.fetchall(
             "SELECT * FROM lxmf_messages WHERE peer_hash = ? ORDER BY timestamp DESC LIMIT ? OFFSET ?",

meshchatx/src/backend/database/schema.py

@@ -157,8 +157,6 @@ class DatabaseSchema:
 
         # Other essential tables that were present from version 1
         # Peewee automatically creates tables if they don't exist.
-        # Here we define the full schema for all tables as they should be now.
-
         tables = {
             "announces": """
                 CREATE TABLE IF NOT EXISTS announces (

meshchatx/src/backend/docs_manager.py

@@ -201,9 +201,7 @@ class DocsManager:
         # Also try in the public directory
         search_paths.append(os.path.join(self.public_dir, "meshchatx-docs"))
 
-        # Also try relative to this file
-        # This file is in meshchatx/src/backend/docs_manager.py
-        # Project root is 3 levels up
+        # Also try relative to this file (project root 3 levels up)
         this_dir = os.path.dirname(os.path.abspath(__file__))
         search_paths.append(
             os.path.abspath(os.path.join(this_dir, "..", "..", "..", "docs")),
@@ -407,8 +405,7 @@ class DocsManager:
                             if f"_{lang}.html" in file:
                                 target_files.append(os.path.join(root, file))
                         else:
-                            # For English, we want files that DON'T have a language suffix
-                            # This is a bit heuristic
+                            # English: no language suffix; other langs use _<lang>.html
                             has_lang_suffix = False
                             for lang_code in known_langs:
                                 if f"_{lang_code}.html" in file:
@@ -634,7 +631,6 @@ class DocsManager:
                     else:
                         shutil.copy2(s, d)
             else:
-                # Just extract everything directly to version_dir, but remove root folder if exists
                 zip_ref.extractall(temp_extract)
                 src_path = os.path.join(temp_extract, root_folder)
                 if os.path.exists(src_path) and os.path.isdir(src_path):

meshchatx/src/backend/interfaces/WebsocketClientInterface.py

@@ -8,7 +8,6 @@ from websockets.sync.connection import Connection
 
 
 class WebsocketClientInterface(Interface):
-    # TODO: required?
     DEFAULT_IFAC_SIZE = 16
 
     RECONNECT_DELAY_SECONDS = 5

meshchatx/src/backend/interfaces/WebsocketServerInterface.py

@@ -11,7 +11,6 @@ from meshchatx.src.backend.interfaces.WebsocketClientInterface import (
 
 
 class WebsocketServerInterface(Interface):
-    # TODO: required?
     DEFAULT_IFAC_SIZE = 16
 
     RESTART_DELAY_SECONDS = 5
@@ -63,7 +62,6 @@ class WebsocketServerInterface(Interface):
     def clients(self):
         return len(self.spawned_interfaces)
 
-    # TODO docs
     def received_announce(self, from_spawned=False):
         if from_spawned:
             self.ia_freq_deque.append(time.time())
@@ -105,14 +103,10 @@ class WebsocketServerInterface(Interface):
             spawned_interface.parent_interface = self
             spawned_interface.online = True
 
-            # TODO implement?
             spawned_interface.announce_rate_target = None
             spawned_interface.announce_rate_grace = None
             spawned_interface.announce_rate_penalty = None
 
-            # TODO ifac?
-            # TODO announce rates?
-
             # activate child interface
             RNS.log(
                 f"Spawned new WebsocketClientInterface: {spawned_interface}",

meshchatx/src/backend/markdown_renderer.py

@@ -1,6 +1,22 @@
 import html
 import re
 
+_SAFE_LINK_PREFIXES = ("https://", "http://", "/", "#", "mailto:")
+_UNSAFE_PROTOCOLS = ("javascript:", "data:", "vbscript:", "file:")
+
+
+def _safe_href(url):
+    if not url or not isinstance(url, str):
+        return "#"
+    u = url.strip().lower()
+    if any(u.startswith(p) for p in _UNSAFE_PROTOCOLS):
+        return "#"
+    if any(u.startswith(p) for p in _SAFE_LINK_PREFIXES):
+        return url
+    if ":" in u.split("/")[0]:
+        return "#"
+    return url
+
 
 class MarkdownRenderer:
     """A simple Markdown to HTML renderer."""
@@ -101,17 +117,29 @@ class MarkdownRenderer:
             flags=re.MULTILINE,
         )
 
-        # Links
+        # Links (href sanitized to prevent javascript:/data: XSS)
+        def link_repl(match):
+            label, url = match.group(1), match.group(2)
+            safe_url = _safe_href(url)
+            return f'<a href="{html.escape(safe_url)}" class="text-blue-600 dark:text-blue-400 hover:underline" target="_blank" rel="noopener noreferrer">{label}</a>'
+
         text = re.sub(
             r"\[([^\]]+)\]\(([^)]+)\)",
-            r'<a href="\2" class="text-blue-600 dark:text-blue-400 hover:underline" target="_blank">\1</a>',
+            link_repl,
             text,
         )
 
-        # Images
+        # Images (src sanitized)
+        def img_repl(match):
+            alt, src = match.group(1), match.group(2)
+            safe_src = _safe_href(src)
+            if safe_src == "#":
+                return html.escape(match.group(0))
+            return f'<div class="my-6"><img src="{html.escape(safe_src)}" alt="{alt}" class="max-w-full h-auto rounded-xl shadow-lg border border-gray-100 dark:border-zinc-800"></div>'
+
         text = re.sub(
             r"!\[([^\]]*)\]\(([^)]+)\)",
-            r'<div class="my-6"><img src="\2" alt="\1" class="max-w-full h-auto rounded-xl shadow-lg border border-gray-100 dark:border-zinc-800"></div>',
+            img_repl,
             text,
         )
 

meshchatx/src/backend/telephone_manager.py

@@ -79,10 +79,6 @@ class TelephoneManager:
 
     @property
     def is_recording(self):
-        # Check if voicemail manager or this manager is recording
-        # This is a bit of a hack since we don't have a direct link to voicemail_manager here
-        # but we can check if our own recording is active if we had it.
-        # For now, we'll just return False and let meshchat.py handle the combined status.
         return False
 
     def init_telephone(self):
@@ -138,8 +134,6 @@ class TelephoneManager:
 
     def on_telephone_ringing(self, caller_identity: RNS.Identity):
         if self.initiation_status:
-            # This is an outgoing call where the remote side is now ringing.
-            # We update the initiation status to "Ringing..." for the UI.
             self._update_initiation_status("Ringing...")
             return
 

meshchatx/src/frontend/components/App.vue

@@ -1094,8 +1094,14 @@ export default {
                 // show result
                 const status = this.propagationNodeStatus?.state;
                 const messagesReceived = this.propagationNodeStatus?.messages_received ?? 0;
+                const messagesStored = this.propagationNodeStatus?.messages_stored ?? 0;
+                const deliveryConfirmations =
+                    this.propagationNodeStatus?.delivery_confirmations ?? 0;
+                const messagesHidden = this.propagationNodeStatus?.messages_hidden ?? 0;
                 if (status === "complete" || status === "idle") {
-                    ToastUtils.success(this.$t("app.sync_complete", { count: messagesReceived }));
+                    const base = this.$t("app.sync_complete", { count: messagesReceived });
+                    const details = `${messagesStored} stored, ${deliveryConfirmations} confirmations, ${messagesHidden} hidden`;
+                    ToastUtils.success(`${base} (${details})`);
                 } else {
                     ToastUtils.error(this.$t("app.sync_error", { status: status }));
                 }

meshchatx/src/frontend/components/call/CallPage.vue

@@ -3223,7 +3223,6 @@ export default {
             }
         },
         async playVoicemail(voicemail) {
-            // This is now handled by AudioWaveformPlayer, but we keep it for backward compatibility or direct calls
             if (this.playingVoicemailId === voicemail.id) {
                 if (this.audioPlayer) {
                     this.audioPlayer.pause();

meshchatx/src/frontend/js/LinkUtils.js

@@ -8,8 +8,7 @@ export default class LinkUtils {
 
         // Hash is 32 hex chars. Path is optional (NomadNet only).
         const hashPattern = "[a-fA-F0-9]{32}";
-        // This matches optional prefixes (nomadnet://, nomadnet@, lxmf://, lxmf@)
-        // followed by a 32-char hex hash, and an optional path starting with :
+        // Optional prefix (nomadnet://, nomadnet@, lxmf://, lxmf@), then hash, optional path.
         const reticulumRegex = new RegExp(
             `(nomadnet://|nomadnet@|lxmf://|lxmf@)?(${hashPattern})(?::(/[\\w\\d./?%&=-]*))?`,
             "g"

meshchatx/src/frontend/js/MarkdownRenderer.js

@@ -7,11 +7,13 @@ export default class MarkdownRenderer {
      * Ported and simplified from meshchatx/src/backend/markdown_renderer.py
      */
     static render(text) {
-        if (!text) {
+        if (text == null) {
             return "";
         }
+        if (typeof text !== "string") {
+            text = String(text);
+        }
 
-        // Escape HTML entities first to prevent XSS
         text = Utils.escapeHtml(text);
 
         // Fenced code blocks - process these FIRST and replace with placeholders
@@ -81,9 +83,12 @@ export default class MarkdownRenderer {
      * Strips markdown from text for previews.
      */
     static strip(text) {
-        if (!text) {
+        if (text == null) {
             return "";
         }
+        if (typeof text !== "string") {
+            text = String(text);
+        }
 
         // Strip fenced code blocks
         text = text.replace(/```(\w+)?\n([\s\S]*?)\n```/g, "[Code Block]");

meshchatx/src/frontend/js/MicronParser.js

@@ -115,9 +115,10 @@ class MicronParser {
     }
 
     convertMicronToHtml(markup) {
+        if (markup == null) return "";
+        if (typeof markup !== "string") markup = String(markup);
         let html = "";
 
-        // parse header tags for page-level color defaults
         const headerColors = this.parseHeaderTags(markup);
 
         const plainStyle = this.SELECTED_STYLES?.plain || { fg: this.DEFAULT_FG_DARK, bg: this.DEFAULT_BG };

meshchatx/src/frontend/js/Utils.js

@@ -197,7 +197,8 @@ class Utils {
     }
 
     static escapeHtml(text) {
-        if (!text) return "";
+        if (text == null) return "";
+        text = String(text);
         const map = {
             "&": "&",
             "<": "&lt;",