Merge branch 'dev' into fix/pages-url-query

src/queries/sql/pageviews/getPageviewExpandedMetrics.ts

@@ -38,20 +38,21 @@ async function relationalQuery(
   const { type, limit = 500, offset = 0 } = parameters;
   let column = FILTER_COLUMNS[type] || type;
   const { rawQuery, parseFilters, getTimestampDiffSQL } = prisma;
-  const { filterQuery, joinSessionQuery, cohortQuery, queryParams } = parseFilters(
-    {
-      ...filters,
-      websiteId,
-    },
-    { joinSession: SESSION_COLUMNS.includes(type) },
-  );
+  const { filterQuery, joinSessionQuery, cohortQuery, excludeBounceQuery, queryParams } =
+    parseFilters(
+      {
+        ...filters,
+        websiteId,
+      },
+      { joinSession: SESSION_COLUMNS.includes(type) },
+    );
 
   let entryExitQuery = '';
   let excludeDomain = '';
   const isPathType = type === 'path' || type === 'entry' || type === 'exit';
 
   if (column === 'referrer_domain') {
-    excludeDomain = `and website_event.referrer_domain != website_event.hostname
+    excludeDomain = `and website_event.referrer_domain != regexp_replace(website_event.hostname, '^www.', '')
       and website_event.referrer_domain != ''`;
     if (type === 'domain') {
       column = toPostgresGroupedReferrer(GROUPED_DOMAINS);
@@ -68,7 +69,7 @@ async function relationalQuery(
         from website_event
         where website_event.website_id = {{websiteId::uuid}}
           and website_event.created_at between {{startDate}} and {{endDate}}
-          and website_event.event_type != 2
+          and website_event.event_type NOT IN (2, 5)
         group by visit_id
       ) x
       on x.visit_id = website_event.visit_id
@@ -95,7 +96,7 @@ async function relationalQuery(
       sum(${getTimestampDiffSQL('t.min_time', 't.max_time')}) as "totaltime"
     from (
       select
-        ${selectColumn} as name,
+        ${selectColumn} as "name",
         website_event.session_id,
         website_event.visit_id,
         count(*) as "c",
@@ -103,11 +104,12 @@ async function relationalQuery(
         max(website_event.created_at) as "max_time"
       from website_event
       ${cohortQuery}
-      ${joinSessionQuery}
-      ${entryExitQuery}
+      ${excludeBounceQuery}
+      ${joinSessionQuery} 
+      ${entryExitQuery} 
       where website_event.website_id = {{websiteId::uuid}}
       and website_event.created_at between {{startDate}} and {{endDate}}
-      and website_event.event_type != 2
+      and website_event.event_type NOT IN (2, 5)
         ${excludeDomain}
         ${filterQuery}
       group by ${groupByColumn}, website_event.session_id, website_event.visit_id
@@ -131,7 +133,7 @@ async function clickhouseQuery(
   const { type, limit = 500, offset = 0 } = parameters;
   let column = FILTER_COLUMNS[type] || type;
   const { rawQuery, parseFilters } = clickhouse;
-  const { filterQuery, cohortQuery, queryParams } = parseFilters({
+  const { filterQuery, cohortQuery, excludeBounceQuery, queryParams } = parseFilters({
     ...filters,
     websiteId,
   });
@@ -159,7 +161,7 @@ async function clickhouseQuery(
       from website_event
       where website_id = {websiteId:UUID}
         and created_at between {startDate:DateTime64} and {endDate:DateTime64}
-        and event_type != 2
+        and event_type NOT IN (2, 5)
       group by visit_id) x
       ON x.visit_id = website_event.visit_id`;
 
@@ -187,10 +189,11 @@ async function clickhouseQuery(
         max(created_at) max_time
       from website_event
       ${cohortQuery}
+      ${excludeBounceQuery}
       ${entryExitQuery}
       where website_id = {websiteId:UUID}
         and created_at between {startDate:DateTime64} and {endDate:DateTime64}
-        and event_type != 2
+        and event_type NOT IN (2, 5)
         and name != ''
         ${excludeDomain}
         ${filterQuery}

src/queries/sql/pageviews/getPageviewMetrics.ts

@@ -34,20 +34,21 @@ async function relationalQuery(
   const { type, limit = 500, offset = 0 } = parameters;
   let column = FILTER_COLUMNS[type] || type;
   const { rawQuery, parseFilters } = prisma;
-  const { filterQuery, joinSessionQuery, cohortQuery, queryParams } = parseFilters(
-    {
-      ...filters,
-      websiteId,
-    },
-    { joinSession: SESSION_COLUMNS.includes(type) },
-  );
+  const { filterQuery, joinSessionQuery, cohortQuery, excludeBounceQuery, queryParams } =
+    parseFilters(
+      {
+        ...filters,
+        websiteId,
+      },
+      { joinSession: SESSION_COLUMNS.includes(type) },
+    );
 
   let entryExitQuery = '';
   let excludeDomain = '';
   const isPathType = type === 'path' || type === 'entry' || type === 'exit';
 
   if (column === 'referrer_domain') {
-    excludeDomain = `and website_event.referrer_domain != website_event.hostname
+    excludeDomain = `and website_event.referrer_domain != regexp_replace(website_event.hostname, '^www.', '')
       and website_event.referrer_domain != ''`;
   }
 
@@ -63,7 +64,7 @@ async function relationalQuery(
         from website_event
         where website_event.website_id = {{websiteId::uuid}}
           and website_event.created_at between {{startDate}} and {{endDate}}
-          and website_event.event_type != 2
+          and website_event.event_type NOT IN (2, 5)
         order by visit_id, created_at ${order}
       ) x
       on x.visit_id = website_event.visit_id
@@ -84,11 +85,13 @@ async function relationalQuery(
       count(distinct website_event.session_id) as y
     from website_event
     ${cohortQuery}
+    ${excludeBounceQuery}
     ${joinSessionQuery}
     ${entryExitQuery}
     where website_event.website_id = {{websiteId::uuid}}
       and website_event.created_at between {{startDate}} and {{endDate}}
-      and website_event.event_type != 2
+      and website_event.event_type NOT IN (2, 5)
+      and ${column} != ''
       ${excludeDomain}
       ${filterQuery}
     group by 1
@@ -109,7 +112,7 @@ async function clickhouseQuery(
   const { type, limit = 500, offset = 0 } = parameters;
   let column = FILTER_COLUMNS[type] || type;
   const { rawQuery, parseFilters } = clickhouse;
-  const { filterQuery, cohortQuery, queryParams } = parseFilters({
+  const { filterQuery, cohortQuery, excludeBounceQuery, queryParams } = parseFilters({
     ...filters,
     websiteId,
   });
@@ -136,7 +139,7 @@ async function clickhouseQuery(
       from website_event
       where website_id = {websiteId:UUID}
         and created_at between {startDate:DateTime64} and {endDate:DateTime64}
-        and event_type != 2
+        and event_type NOT IN (2, 5)
       group by visit_id) x
       ON x.visit_id = website_event.visit_id`;
 
@@ -150,10 +153,12 @@ async function clickhouseQuery(
       uniq(website_event.session_id) as y
     from website_event
     ${cohortQuery}
+    ${excludeBounceQuery}
     ${entryExitQuery}
     where website_id = {websiteId:UUID}
       and created_at between {startDate:DateTime64} and {endDate:DateTime64}
-      and event_type != 2
+      and event_type NOT IN (2, 5)
+      and ${column} != ''
       ${excludeDomain}
       ${filterQuery}
     group by x
@@ -189,9 +194,10 @@ async function clickhouseQuery(
         ${columnQuery} as t
       from website_event_stats_hourly as website_event
       ${cohortQuery}
+      ${excludeBounceQuery}
       where website_id = {websiteId:UUID}
         and created_at between {startDate:DateTime64} and {endDate:DateTime64}
-        and event_type != 2
+        and event_type NOT IN (2, 5)
         ${excludeDomain}
         ${filterQuery}
       ${groupByQuery}) as g

src/queries/sql/pageviews/getPageviewStats.ts

@@ -16,10 +16,11 @@ export async function getPageviewStats(...args: [websiteId: string, filters: Que
 async function relationalQuery(websiteId: string, filters: QueryFilters) {
   const { timezone = 'utc', unit = 'day' } = filters;
   const { getDateSQL, parseFilters, rawQuery } = prisma;
-  const { filterQuery, cohortQuery, joinSessionQuery, queryParams } = parseFilters({
-    ...filters,
-    websiteId,
-  });
+  const { filterQuery, cohortQuery, excludeBounceQuery, joinSessionQuery, queryParams } =
+    parseFilters({
+      ...filters,
+      websiteId,
+    });
 
   return rawQuery(
     `
@@ -28,10 +29,11 @@ async function relationalQuery(websiteId: string, filters: QueryFilters) {
       count(*) y
     from website_event
     ${cohortQuery}
+    ${excludeBounceQuery}
     ${joinSessionQuery}  
     where website_event.website_id = {{websiteId::uuid}}
       and website_event.created_at between {{startDate}} and {{endDate}}
-      and website_event.event_type != 2
+      and website_event.event_type NOT IN (2, 5)
       ${filterQuery}
     group by 1
     order by 1
@@ -47,7 +49,7 @@ async function clickhouseQuery(
 ): Promise<{ x: string; y: number }[]> {
   const { timezone = 'UTC', unit = 'day' } = filters;
   const { parseFilters, rawQuery, getDateSQL } = clickhouse;
-  const { filterQuery, cohortQuery, queryParams } = parseFilters({
+  const { filterQuery, cohortQuery, excludeBounceQuery, queryParams } = parseFilters({
     ...filters,
     websiteId,
   });
@@ -65,9 +67,10 @@ async function clickhouseQuery(
         count(*) as y
       from website_event
       ${cohortQuery}
+      ${excludeBounceQuery}
       where website_id = {websiteId:UUID}
         and created_at between {startDate:DateTime64} and {endDate:DateTime64}
-        and event_type != 2
+        and event_type NOT IN (2, 5)
         ${filterQuery}
       group by t
     ) as g
@@ -84,9 +87,10 @@ async function clickhouseQuery(
         sum(views) as y
       from website_event_stats_hourly as website_event
       ${cohortQuery}
+      ${excludeBounceQuery}
       where website_id = {websiteId:UUID}
         and created_at between {startDate:DateTime64} and {endDate:DateTime64}
-        and event_type != 2
+        and event_type NOT IN (2, 5)
         ${filterQuery}
       group by t
     ) as g