From 4e340412c020f685cb402a735b983f6e332e206b Mon Sep 17 00:00:00 2001 From: reivilibre <38398653+reivilibre@users.noreply.github.com> Date: Mon, 19 Jul 2021 16:11:34 +0100 Subject: Add a new version of the R30 phone-home metric, which removes a false impression of retention given by the old R30 metric (#10332) Signed-off-by: Olivier Wilkinson (reivilibre) --- synapse/app/phone_stats_home.py | 4 + synapse/storage/databases/main/metrics.py | 129 ++++++++++++++++++++++++++++++ 2 files changed, 133 insertions(+) (limited to 'synapse') diff --git a/synapse/app/phone_stats_home.py b/synapse/app/phone_stats_home.py index 8f86cecb76..7904c246df 100644 --- a/synapse/app/phone_stats_home.py +++ b/synapse/app/phone_stats_home.py @@ -107,6 +107,10 @@ async def phone_stats_home(hs, stats, stats_process=_stats_process): for name, count in r30_results.items(): stats["r30_users_" + name] = count + r30v2_results = await hs.get_datastore().count_r30_users() + for name, count in r30v2_results.items(): + stats["r30v2_users_" + name] = count + stats["cache_factor"] = hs.config.caches.global_factor stats["event_cache_size"] = hs.config.caches.event_cache_size diff --git a/synapse/storage/databases/main/metrics.py b/synapse/storage/databases/main/metrics.py index e3a544d9b2..dc0bbc56ac 100644 --- a/synapse/storage/databases/main/metrics.py +++ b/synapse/storage/databases/main/metrics.py @@ -316,6 +316,135 @@ class ServerMetricsStore(EventPushActionsWorkerStore, SQLBaseStore): return await self.db_pool.runInteraction("count_r30_users", _count_r30_users) + async def count_r30v2_users(self) -> Dict[str, int]: + """ + Counts the number of 30 day retained users, defined as users that: + - Appear more than once in the past 60 days + - Have more than 30 days between the most and least recent appearances that + occurred in the past 60 days. + + (This is the second version of this metric, hence R30'v2') + + Returns: + A mapping from client type to the number of 30-day retained users for that client. + + The dict keys are: + - "all" (a combined number of users across any and all clients) + - "android" (Element Android) + - "ios" (Element iOS) + - "electron" (Element Desktop) + - "web" (any web application -- it's not possible to distinguish Element Web here) + """ + + def _count_r30v2_users(txn): + thirty_days_in_secs = 86400 * 30 + now = int(self._clock.time()) + sixty_days_ago_in_secs = now - 2 * thirty_days_in_secs + one_day_from_now_in_secs = now + 86400 + + # This is the 'per-platform' count. + sql = """ + SELECT + client_type, + count(client_type) + FROM + ( + SELECT + user_id, + CASE + WHEN + LOWER(user_agent) LIKE '%%riot%%' OR + LOWER(user_agent) LIKE '%%element%%' + THEN CASE + WHEN + LOWER(user_agent) LIKE '%%electron%%' + THEN 'electron' + WHEN + LOWER(user_agent) LIKE '%%android%%' + THEN 'android' + WHEN + LOWER(user_agent) LIKE '%%ios%%' + THEN 'ios' + ELSE 'unknown' + END + WHEN + LOWER(user_agent) LIKE '%%mozilla%%' OR + LOWER(user_agent) LIKE '%%gecko%%' + THEN 'web' + ELSE 'unknown' + END as client_type + FROM + user_daily_visits + WHERE + timestamp > ? + AND + timestamp < ? + GROUP BY + user_id, + client_type + HAVING + max(timestamp) - min(timestamp) > ? + ) AS temp + GROUP BY + client_type + ; + """ + + # We initialise all the client types to zero, so we get an explicit + # zero if they don't appear in the query results + results = {"ios": 0, "android": 0, "web": 0, "electron": 0} + txn.execute( + sql, + ( + sixty_days_ago_in_secs * 1000, + one_day_from_now_in_secs * 1000, + thirty_days_in_secs * 1000, + ), + ) + + for row in txn: + if row[0] == "unknown": + continue + results[row[0]] = row[1] + + # This is the 'all users' count. + sql = """ + SELECT COUNT(*) FROM ( + SELECT + 1 + FROM + user_daily_visits + WHERE + timestamp > ? + AND + timestamp < ? + GROUP BY + user_id + HAVING + max(timestamp) - min(timestamp) > ? + ) AS r30_users + """ + + txn.execute( + sql, + ( + sixty_days_ago_in_secs * 1000, + one_day_from_now_in_secs * 1000, + thirty_days_in_secs * 1000, + ), + ) + row = txn.fetchone() + if row is None: + results["all"] = 0 + else: + results["all"] = row[0] + + return results + + return await self.db_pool.runInteraction( + "count_r30v2_users", _count_r30v2_users + ) + def _get_start_of_day(self): """ Returns millisecond unixtime for start of UTC day. -- cgit 1.4.1