Upgrade Notes: v1.0.0b6¶

What changed since b3, what breaks, and how to upgrade. If you already upgraded to b4, these breaking changes were already handled — you can upgrade to b6 with just pip install --upgrade --pre getdango and dango start.

Breaking Changes (from b3)¶

These require action if upgrading directly from b3:

1. GA4 Column Names Changed¶

Column names no longer include type suffixes:

Before (b3)	After (b4+)
`sessions_integer`	`sessions`
`bounce_rate_float`	`bounce_rate`
`total_users_integer`	`total_users`
`new_users_integer`	`new_users`
`screen_page_views_integer`	`screen_page_views`

Action required: Update any custom dbt models that reference old column names.

2. Google Ads / GA4 Data Types Fixed¶

Column	Before (b3)	After (b4+)
`date` (Google Ads)	VARCHAR	DATE
`date` (GA4)	TIMESTAMPTZ	DATE
`clicks`, `impressions` (Google Ads)	VARCHAR	INTEGER

Action required: Run a full refresh for each affected source:

dango sync <your_google_ads_source> --full-refresh
dango sync <your_ga4_source> --full-refresh

3. GA4 Default Queries Updated¶

The events and conversions queries now include the landingPage dimension.

Action (existing projects only): If you want the new dimension, edit .dango/sources.yml for your GA4 source to add landingPage to the dimensions list, then run --full-refresh. New projects get this automatically.

Upgrade Steps¶

# 1. Upgrade the package
pip install --upgrade --pre getdango

# 2. Start Dango (auto-runs database migrations)
dango start

# 3. Full refresh affected sources (if upgrading from b3)
dango sync <google_ads_source> --full-refresh
dango sync <ga4_source> --full-refresh

# 4. Update custom dbt models referencing old GA4 column names

# 5. Verify dbt models compile
dango run

What Works Automatically After Upgrade¶

No action needed for these — they just work after upgrading:

Scheduler loads jobs from schedules.yml on startup (was broken in b3)
Staging YML files (sources_*.yml) are no longer overwritten on sync — user edits preserved
Cloud file sync now includes custom_sources/ and seeds/ directories
Backups now include dbt models, custom sources, and .env
Metabase dashboard export captures all cards (was exporting 0 in some cases)
Sync failures properly logged as errors with status in sync history
OAuth refresh argument changed from credential name to source type (e.g., dango oauth refresh google_sheets)

Cloud Deployments¶

After upgrading locally:

# 1. Push updated project files to the server
dango remote push

# 2. SSH in and upgrade the Dango package
dango remote ssh
# On server:
cd /srv/dango/project
source venv/bin/activate
pip install --upgrade --pre getdango
dango start

# 3. Full refresh affected sources (if upgrading from b3)
dango sync <google_ads_source> --full-refresh
dango sync <ga4_source> --full-refresh