In this release we're releasing Google Adwords integration, many Salesforce improvements, updates to our Google Analytics integration, and more!
- Google Adwords Beta - We're on a roll with Google services! Hot on the heels of our Google Analytics beta release, we're proud to now also support Google Adwords.
- You can start pulling your Google Adwords data on a daily basis - just specify your Customer ID, select the report type, use our easy-peasy auto-complete to select your fields, and go!
- Since Google Adwords data only makes sense at a delay, we default to 1 day but you can configure it to as much time as you want. On any given day we'll read the data from that number of days in the past.
- Because the integration is still in beta, you can sign in only as the last step at the bottom of the dialog. Enjoy, and if you have any questions or comments please give us a shout out at firstname.lastname@example.org.
- Salesforce Improvements - We love Salesforce (and know you do, too), and now our Salesforce input is better than ever!
- No more connected app necessary! You can now connect to Salesforce with just your username, password, and token.
- We know API calls are precious, so we added the ability to cap the number of API and Bulk API calls we'll make per day. Don't worry, we'll pick up where we left off the following day, so heavy days don't exhaust your quota.
- Salesforce records now get consolidated! We didn't want MySQL, MongoDB, and Postgres to have all the fun, so we added consolidation to our Salesforce input as well. Expect to see your records consolidated into a coherent Salesforce table per object.
- We've added delete support, such that deleted records are now honored and marked as such in Redshift.
- We've added the ability to indicate whether your account is a sandbox.
- The efficiency of the input is much improved with week over week data pulls during backfill to better handle data-heavy objects.
- The Google Analytics beta input is much improved (though, we still consider it beta)! We've added:
- The ability to add sort parameters.
- Auto-complete of the metrics and dimensions based on API name.
- Notifications for quota exceeded situations.
- Notifications and metadata to indicate sampling level for a particular input. The metadata includes the total results, sample space, and sample size for the given day on each event, so you can decide what to do with those events in the Code Engine as they stream into the system.
- We added the ability for the Code Engine to support unicode characters in field names as well as in the transform itself.
- We extended our PostgreSQL input to allow defining any monotonically increasing field as the replication index, where once we were limiting it to Timestamp fields.
- A few releases ago we made metadata always flexible, so we wouldn't break your mappings by adding new metadata. You told us it's irritating that all your event type dots turned yellow in the mapping screen, so now the dots are back to green if only metadata is unmapped.
- We've retired our old Java reporter input, now supporting only our fancy log4j interface version.
- By popular demand, we now allow connections to MongoDB to be created without defining a username and password. You do you, MongoDB!
- We had an issue where we were keeping connections to Redshift open unnecessarily, so those have been closed.
- Some of you were seeing numeric conversion errors on field values like 0E-9 - these values no longer give an error.
- Empty string is no longer being treated as NULL when being mapped into a varchar type, but as an empty string (as it should be!).
- Null fields coming from MySQL and being mapped to varchar will now show up as NULL in Redshift instead of as an empty string.
- Field names with "$$" in them were being mapped to a single "$", but they're now being mapped as is.