ga4-export-fixer 0.9.0 → 0.9.1

package/README.md

@@ -16,6 +16,7 @@ The goal of the package is to **speed up development** when building data models
 
 ## Table of Contents
 <!-- TOC -->
+- [Blog Posts](#blog-posts)
 - [Main Features](#main-features)
 - [Planned Features](#planned-features)
 - [Installation](#installation)
@@ -30,6 +31,10 @@ The goal of the package is to **speed up development** when building data models
 - [License](#license)
 <!-- /TOC -->
 
+## Blog Posts
+
+- [ga4-export-fixer: Getting started in Dataform](https://tanelytics.com/ga4-export-fixer-getting-started-in-dataform/)
+
 ## Main Features
 
 <table>
@@ -150,6 +155,7 @@ The goal of the package is to **speed up development** when building data models
 Features under consideration for future releases:
 
 - Aggregated tables (ga4_session, ga4_ecommerce...)
+- Roll-ups for creating a single ga4_events_enhanced table from multiple source exports
 - Web and app specific default configurations
 - Custom channel grouping
 - Custom traffic source attribution
@@ -172,7 +178,7 @@ Include the package in the package.json file in your Dataform repository.
 {
   "dependencies": {
     "@dataform/core": "3.0.42",
-    "ga4-export-fixer": "0.9.0"
+    "ga4-export-fixer": "0.9.1"
   }
 }
 ```

package/helpers/ga4Transforms.js

@@ -97,7 +97,7 @@ const isFinalData = (detectionMethod, dayThreshold) => {
 /**
  * The standard GA4 BigQuery export top-level column names, based on the official schema.
  *
- * list updated 2026-02-18
+ * list updated 2026-07-30
  */
 const ga4ExportColumns = [
   "event_date",
@@ -108,6 +108,7 @@ const ga4ExportColumns = [
   "event_value_in_usd",
   "event_bundle_sequence_id",
   "event_server_timestamp_offset",
+  "event_original_occurrence_timestamp",
   "user_id",
   "user_pseudo_id",
   "privacy_info",

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "ga4-export-fixer",
-  "version": "0.9.0",
+  "version": "0.9.1",
   "description": "",
   "main": "index.js",
   "files": [

package/tables/ga4EventsEnhanced/columns/columnDescriptions.json

@@ -290,6 +290,7 @@
     "event_value_in_usd": "Currency-converted value (in USD) of the event's 'value' parameter",
     "event_bundle_sequence_id": "Sequential ID of the bundle in which the event was uploaded",
     "event_server_timestamp_offset": "Timestamp offset between collection time and upload time in microseconds",
+    "event_original_occurrence_timestamp": "Timestamp in microseconds (UTC) of when the event originally occurred, for events that arrived at GA4 after the standard 72-hour collection window. Such events are processed in the current day's export (previously they were dropped); this field preserves the original event time. Null for events that arrived within 72 hours.",
     "privacy_info": {
         "description": "Consent status information for the user when consent mode is enabled",
         "columns": {

package/tables/ga4EventsEnhanced/columns/columnLineage.json

@@ -92,6 +92,9 @@
     "event_server_timestamp_offset": {
         "source": "ga4_export"
     },
+    "event_original_occurrence_timestamp": {
+        "source": "ga4_export"
+    },
     "privacy_info": {
         "source": "ga4_export"
     },

package/tables/ga4EventsEnhanced/columns/columnTypicalUse.json

@@ -2,6 +2,7 @@
     "event_date": "Partition column. Always include in WHERE clauses to limit scanned data and reduce query cost",
     "event_datetime": "Human-readable timestamp in the configured timezone. Use for time-of-day analysis and reporting",
     "event_timestamp": "Microsecond-precision UTC timestamp. Use for precise event ordering and time difference calculations",
+    "event_original_occurrence_timestamp": "Use to identify events that arrived late (after 72 hours) and to recover their true occurrence time. On such rows event_timestamp reflects when GA4 processed the event, not when it actually happened. Filter WHERE event_original_occurrence_timestamp IS NOT NULL to isolate delayed events",
     "event_name": "Primary event filter. Use in WHERE clauses to select specific event types (e.g. WHERE event_name = 'purchase')",
     "session_id": "Use for counting unique sessions (COUNT(DISTINCT session_id)) and as a GROUP BY key for session-level aggregations",
     "user_pseudo_id": "Device-level user identifier. Use for counting unique devices or as a fallback when merged_user_id is not needed",