[CLEAN] Synthetic Benchmark PR #25719 - Added materialized view and duplicated v2 Tinybird endpoints

apps/admin-x-framework/src/providers/framework-provider.tsx

@@ -10,6 +10,7 @@ export interface StatsConfig {
     endpointBrowser?: string;
     id?: string;
     token?: string;
+    version?: string;
     local?: {
         enabled?: boolean;
         endpoint?: string;

apps/admin-x-framework/src/utils/stats-config.ts

@@ -13,7 +13,11 @@ export const getStatEndpointUrl = (config?: StatsConfig | null, endpointName?: s
     } else {
         baseUrl = config.endpoint || '';
     }
-    return `${baseUrl}/v0/pipes/${endpointName}.json?${params}`;
+
+    // Append version suffix if provided (e.g., "v2" -> "api_kpis_v2")
+    const finalEndpointName = config.version ? `${endpointName}_${config.version}` : endpointName;
+
+    return `${baseUrl}/v0/pipes/${finalEndpointName}.json?${params}`;
 };
 
 export const getToken = () => {

ghost/core/core/server/api/endpoints/stats.js

@@ -125,7 +125,6 @@ const controller = {
             'date_to',
             'timezone',
             'member_status',
-            'tb_version',
             'post_type',
             'post_uuid',
             'pathname',

ghost/core/core/server/data/tinybird/README.md

@@ -78,6 +78,8 @@ Sample config:
             //  -- optional override for site uuid
             // "id": "106a623d-9792-4b63-acde-4a0c28ead3dc",
             "endpoint": "https://api.tinybird.co",
+            // -- optional endpoint version suffix (e.g., "v2" calls api_kpis_v2 instead of api_kpis)
+            // "version": "v2",
             // -- tinybird local configuration (optional)
             "local": {
                 "enabled": true,

ghost/core/core/server/data/tinybird/datasources/_mv_session_data_v2.datasource

@@ -0,0 +1,16 @@
+SCHEMA >
+    `site_uuid` LowCardinality(String),
+    `session_id` String,
+    `pageviews` AggregateFunction(count, UInt64),
+    `first_pageview` AggregateFunction(min, DateTime),
+    `last_pageview` AggregateFunction(max, DateTime),
+    `source` AggregateFunction(argMin, String, DateTime),
+    `device` AggregateFunction(argMin, String, DateTime),
+    `utm_source` AggregateFunction(argMin, String, DateTime),
+    `utm_medium` AggregateFunction(argMin, String, DateTime),
+    `utm_campaign` AggregateFunction(argMin, String, DateTime),
+    `utm_term` AggregateFunction(argMin, String, DateTime),
+    `utm_content` AggregateFunction(argMin, String, DateTime)
+
+ENGINE "AggregatingMergeTree"
+ENGINE_SORTING_KEY "site_uuid, session_id"

ghost/core/core/server/data/tinybird/endpoints/README.md

@@ -7,7 +7,7 @@
 Ghost analytics distinguishes between two types of attributes:
 
 #### Session-Level Attributes
-These are captured from the **first hit** (earliest timestamp) in a session using `argMin(field, timestamp)` in the `mv_session_data` materialized view:
+These are captured from the **first hit** (earliest timestamp) in a session using `argMinState(field, timestamp)` in the `_mv_session_data` materialized view (an `AggregatingMergeTree` table):
 
 - `source` - Referring domain
 - `utm_source` - UTM source parameter
@@ -39,7 +39,7 @@ Finds sessions where **at least one hit** matches the hit-level filter criteria
 ```sql
 NODE sessions_filtered_by_session_attributes
 ```
-Further filters by session-level attributes (source, utm_*) by joining with `mv_session_data`. These filters check attributes from the **first hit only**.
+Further filters by session-level attributes (source, utm_*) by reading from `_mv_session_data` using `-Merge` combinators (e.g., `argMinMerge(source)`). These filters check attributes from the **first hit only**.
 
 **Stage 3: Final Output**
 ```sql

ghost/core/core/server/data/tinybird/endpoints/api_active_visitors_v2.pipe

@@ -0,0 +1,15 @@
+TOKEN "stats_page" READ
+TOKEN "axis" READ
+
+NODE _active_visitors_0
+SQL >
+%
+    select
+        uniqExact(session_id) as active_visitors
+    from _mv_hits
+    where
+        site_uuid = {{String(site_uuid, 'mock_site_uuid', description="Site UUID", required=True)}}
+        and timestamp >= (now() - interval 5 minute)
+        {% if defined(post_uuid) %} and post_uuid = {{ String(post_uuid, description="Post UUID to filter on", required=False) }} {% end %}
+
+TYPE ENDPOINT

ghost/core/core/server/data/tinybird/endpoints/api_kpis_v2.pipe

@@ -0,0 +1,162 @@
+TOKEN "stats_page" READ
+TOKEN "axis" READ
+
+NODE timeseries
+SQL >
+
+    %
+        {% set _single_day = defined(date_from) and day_diff(date_from, date_to) == 0 %}
+        with
+            {% if defined(date_from) %}
+                toStartOfDay(
+                    toDate(
+                        {{
+                            Date(
+                                date_from,
+                                description="Starting day for filtering a date range",
+                                required=False,
+                            )
+                        }}
+                    )
+                ) as start,
+            {% else %} toStartOfDay(timestampAdd(today(), interval -7 day)) as start,
+            {% end %}
+            {% if defined(date_to) %}
+                toStartOfDay(
+                    toDate(
+                        {{
+                            Date(
+                                date_to,
+                                description="Finishing day for filtering a date range",
+                                required=False,
+                            )
+                        }}
+                    )
+                ) as end
+            {% else %} toStartOfDay(today()) as end
+            {% end %}
+        {% if _single_day %}
+            select
+                arrayJoin(
+                    arrayMap(
+                        x -> toDateTime(toString(toDateTime(x)), {{String(timezone, 'Etc/UTC', description="Site timezone", required=True)}}),
+                        range(
+                            toUInt32(toDateTime(start)), toUInt32(timestampAdd(end, interval 1 day)), 3600
+                        )
+                    )
+                ) as date
+        {% else %}
+            select
+                arrayJoin(
+                    arrayMap(
+                        x -> toDate(x),
+                        range(toUInt32(start), toUInt32(timestampAdd(end, interval 1 day)), 24 * 3600)
+                    )
+                ) as date
+        {% end %}
+
+
+NODE session_data
+DESCRIPTION >
+    Read session data from AggregatingMergeTree MV using -Merge combinators
+
+SQL >
+    %
+    SELECT
+        site_uuid,
+        session_id,
+        countMerge(pageviews) as pageviews,
+        minMerge(first_pageview) as first_pageview,
+        maxMerge(last_pageview) as last_pageview
+    FROM _mv_session_data_v2
+    WHERE site_uuid = {{ String(site_uuid, 'mock_site_uuid', description="Tenant ID", required=True) }}
+    GROUP BY site_uuid, session_id
+
+NODE session_metrics
+DESCRIPTION >
+    Calculate session-level metrics (visits, pageviews, bounce rate, avg session duration)
+
+SQL >
+
+    %
+        select
+            site_uuid,
+            {% if defined(date_from) and day_diff(date_from, date_to) == 0 %}
+                toStartOfHour(toTimezone(first_pageview, {{String(timezone, 'Etc/UTC', description="Site timezone", required=True)}})) as date,
+            {% else %}
+                toDate(toTimezone(first_pageview, {{String(timezone, 'Etc/UTC', description="Site timezone", required=True)}})) as date,
+            {% end %}
+            sd.session_id,
+            pageviews,
+            pageviews = 1 as is_bounce,
+            last_pageview - first_pageview as session_sec
+        from session_data sd
+            inner join filtered_sessions_v2 fs
+                on fs.session_id = sd.session_id
+
+
+NODE data
+DESCRIPTION >
+    Calculate KPIs per time period
+
+SQL >
+
+    select
+        a.date,
+        uniq(distinct s.session_id) as visits,
+        sum(s.pageviews) as pageviews,
+        truncate(avg(s.is_bounce), 2) as bounce_rate,
+        truncate(avg(s.session_sec), 2) as avg_session_sec
+    from timeseries a
+    inner join session_metrics s on a.date = s.date
+    group by a.date
+    order by a.date
+
+
+NODE pathname_pageviews
+DESCRIPTION >
+    Calculate pageviews for specific pathname with time granularity handling
+
+SQL >
+
+    %
+            select
+                {% if defined(date_from) and day_diff(date_from, date_to) == 0 %}
+                    toStartOfHour(toTimezone(timestamp, {{String(timezone, 'Etc/UTC', description="Site timezone", required=True)}})) as date,
+                {% else %}
+                    toDate(toTimezone(timestamp, {{String(timezone, 'Etc/UTC', description="Site timezone", required=True)}})) as date,
+                {% end %}
+                count() pageviews
+            from timeseries a
+            inner join _mv_hits h on
+                {% if defined(date_from) and day_diff(date_from, date_to) == 0 %}
+                    a.date = toStartOfHour(toTimezone(timestamp, {{String(timezone, 'Etc/UTC', description="Site timezone", required=True)}}))
+                {% else %}
+                    a.date = toDate(toTimezone(timestamp, {{String(timezone, 'Etc/UTC', description="Site timezone", required=True)}}))
+                {% end %}
+            inner join filtered_sessions_v2 fs
+                on fs.session_id = h.session_id
+            where
+                site_uuid = {{ String(site_uuid, 'mock_site_uuid', description="Tenant ID", required=True) }}
+                {% if defined(member_status) %} and member_status IN {{ Array(member_status, "'undefined', 'free', 'paid'", description="Member status to filter on", required=False) }} {% end %}
+                {% if defined(location) %} and location = {{ String(location, description="Location to filter on", required=False) }} {% end %}
+                {% if defined(pathname) %} and pathname = {{ String(pathname, description="Pathname to filter on", required=False) }} {% end %}
+                {% if defined(post_uuid) %} and post_uuid = {{String(post_uuid, description="Post UUID to filter on", required=False) }} {% end %}
+            group by date
+            order by date
+
+
+NODE finished_data
+SQL >
+
+    %
+            select
+                a.date as date,
+                coalesce(b.visits, 0) as visits,
+                {% if defined(pathname) or defined(post_uuid) %}coalesce(c.pageviews, 0){% else %}coalesce(b.pageviews, 0){% end %} as pageviews,
+                coalesce(b.bounce_rate, 0) as bounce_rate,
+                coalesce(b.avg_session_sec, 0) as avg_session_sec
+            from timeseries a
+            left join data b on a.date = b.date
+            {% if defined(pathname) or defined(post_uuid) %}left join pathname_pageviews c on a.date = c.date{% end %}
+TYPE ENDPOINT

ghost/core/core/server/data/tinybird/endpoints/api_post_visitor_counts_v2.pipe

@@ -0,0 +1,17 @@
+TOKEN "stats_page" READ
+TOKEN "axis" READ
+
+NODE _post_visitor_counts_0
+SQL >
+%
+    select
+        post_uuid,
+        uniqExact(session_id) as visits
+    from _mv_hits
+    where
+        site_uuid = {{String(site_uuid, 'mock_site_uuid', description="Site UUID", required=True)}}
+        and post_uuid IN {{ Array(post_uuids, description="Array of post UUIDs to get visitor counts for", required=True) }}
+    group by post_uuid
+    order by visits desc
+
+TYPE ENDPOINT

ghost/core/core/server/data/tinybird/endpoints/api_top_devices_v2.pipe

@@ -0,0 +1,28 @@
+TOKEN "stats_page" READ
+TOKEN "axis" READ
+
+NODE session_data
+SQL >
+    %
+    SELECT
+        site_uuid,
+        session_id,
+        argMinMerge(device) as device
+    FROM _mv_session_data_v2
+    WHERE site_uuid = {{ String(site_uuid, 'mock_site_uuid', description="Tenant ID", required=True) }}
+    GROUP BY site_uuid, session_id
+
+NODE top_devices
+SQL >
+    %
+    select
+        device,
+        count() as visits
+    from session_data sd
+      inner join filtered_sessions_v2 fs
+         on fs.session_id = sd.session_id
+    group by device
+    order by visits desc
+    limit {{ Int32(skip, 0) }},{{ Int32(limit, 50) }}
+
+TYPE ENDPOINT

ghost/core/core/server/data/tinybird/endpoints/api_top_locations_v2.pipe

@@ -0,0 +1,31 @@
+TOKEN "stats_page" READ
+TOKEN "axis" READ
+
+NODE _top_locations_0
+SQL >
+
+    %
+        select
+            location,
+            uniqExact(session_id) as visits
+        from _mv_hits h
+        inner join filtered_sessions_v2 fs
+            on fs.session_id = h.session_id
+        where
+            site_uuid = {{String(site_uuid, 'mock_site_uuid', description="Tenant ID", required=True)}}
+            {% if defined(member_status) %}
+                and member_status IN (
+                    select arrayJoin(
+                        {{ Array(member_status, "'undefined', 'free', 'paid'", description="Member status to filter on", required=False) }}
+                        || if('paid' IN {{ Array(member_status) }}, ['comped'], [])
+                    )
+                )
+            {% end %}
+            {% if defined(location) %} and location = {{ String(location, description="Location to filter on", required=False) }} {% end %}
+            {% if defined(pathname) %} and pathname = {{ String(pathname, description="Pathname to filter on", required=False) }} {% end %}
+            {% if defined(post_uuid) %} and post_uuid = {{ String(post_uuid, description="Post UUID to filter on", required=False) }} {% end %}
+        group by location
+        order by visits desc
+        limit {{ Int32(skip, 0) }},{{ Int32(limit, 50) }}
+
+TYPE ENDPOINT

ghost/core/core/server/data/tinybird/endpoints/api_top_pages_v2.pipe

@@ -0,0 +1,39 @@
+TOKEN "stats_page" READ
+TOKEN "axis" READ
+
+NODE _top_pages_0
+SQL >
+
+    %
+    select
+        case when post_uuid = 'undefined' then '' else post_uuid end as post_uuid,
+        pathname,
+        uniqExact(session_id) as visits
+    from _mv_hits h
+    inner join filtered_sessions_v2 fs
+        on fs.session_id = h.session_id
+    where
+        site_uuid = {{String(site_uuid, 'mock_site_uuid', description="Tenant ID", required=True)}}
+        {% if defined(member_status) %}
+            and member_status IN (
+                select arrayJoin(
+                    {{ Array(member_status, "'undefined', 'free', 'paid'", description="Member status to filter on", required=False) }}
+                    || if('paid' IN {{ Array(member_status) }}, ['comped'], [])
+                )
+            )
+        {% end %}
+        {% if defined(location) %} and location = {{ String(location, description="Location to filter on", required=False) }} {% end %}
+        {% if defined(pathname) %} and pathname = {{ String(pathname, description="Pathname to filter on", required=False) }} {% end %}
+        {% if defined(post_uuid) %} and post_uuid = {{ String(post_uuid, description="Post UUID to filter on", required=False) }} {% end %}
+        {% if defined(post_type) %}
+            {% if post_type == 'post' %}
+                and post_type = 'post'
+            {% else %}
+                and (post_type != 'post' or post_type is null)
+            {% end %}
+        {% end %}
+    group by post_uuid, pathname
+    order by visits desc
+    limit {{ Int32(skip, 0) }},{{ Int32(limit, 50) }}
+
+TYPE ENDPOINT

ghost/core/core/server/data/tinybird/endpoints/api_top_sources_v2.pipe

@@ -0,0 +1,28 @@
+TOKEN "stats_page" READ
+TOKEN "axis" READ
+
+NODE session_data
+SQL >
+    %
+    SELECT
+        site_uuid,
+        session_id,
+        argMinMerge(source) as source
+    FROM _mv_session_data_v2
+    WHERE site_uuid = {{ String(site_uuid, 'mock_site_uuid', description="Tenant ID", required=True) }}
+    GROUP BY site_uuid, session_id
+
+NODE top_sources
+SQL >
+    %
+    select
+        source,
+        count() as visits
+    from session_data sd
+      inner join filtered_sessions_v2 fs
+         on fs.session_id = sd.session_id
+    group by source
+    order by visits desc
+    limit {{ Int32(skip, 0) }},{{ Int32(limit, 50) }}
+
+TYPE ENDPOINT