You are viewing documentation for an older version.
Click here to view the latest documentation.

Profiles 0.12.x Changelog

Changelog for Profiles v0.12.x.

Version 0.12.1

2 May 2024

Improvements

Bug Fixes

Resolved bug where some projects were failing in case nested columns were missing.
Updated the migration logic to preserve old view names. The reason being that some existing Activation API projects were failing due to the renaming of serve_traits to feature_views.

Known Issues

Linux users might see this warning for all command runs - you can ignore it: WARN[0000]log.go:228 gosnowflake.(*defaultLogger).Warn DBUS_SESSION_BUS_ADDRESS envvar looks to be not set, this can lead to runaway dbus-daemon processes. To avoid this, set envvar DBUS_SESSION_BUS_ADDRESS=$XDG_RUNTIME_DIR/bus (if it exists) or DBUS_SESSION_BUS_ADDRESS=/dev/null.
Redshift: If two different users create material objects on the same schema, RudderStack will throw error during cleanup when trying to drop views created by the other user, such as user_var_table.
pb validate access command does not work for BigQuery.
Some commands such as pb insert does not work for Redshift, Databricks, and BigQuery.
For a few clusters, cross database references can fail on Redshift.
If you are referring a public package in the project and get ssh: handshake failed error, then you’ll have to manually clear WhtGitCache folder to make it work.
The code for validity_time is redundant and should be removed.
In some cases, you may need to install the profiles-rudderstack and profiles-rudderstack-bin pip packages separately.
You may have to execute the compile command once before executing validate access. Otherwise, you will get a seq_no error.

25 April 2024 Schema version: 61

What’s New

Support for external tables on Redshift Serverless.
Redshift users can now also connect via SSH Tunnel.
Support to specify the version on which you want to run your Profiles project. It is also backported to v0.10.8 and v0.11.5 onwards.
Feature views path refs look like below:
- user/all/feature_view # By default, the id served is user main_id.
- user/all/feature_view/using_email # For non default IDs, the path ref has using_<idname>
- user/all/feature_view/salesforce_id_stitched_features # The names can still be overridden.

Ability to add description (optional) to feature tags in the project file.

pb_project.yaml

available_tags:
    - name: demographics
      description: all tags related to user demographics
    - name: billing

Reorganized the overall flow when defining a Redshift connection. You can specify the region while using Redshift.
Creating a new project using pb init pb-project now comes with sample data in CSV format, in a folder named csvs. These are referenced in the input file.
Renamed Entity Traits 360 model to Feature Views. Also, in the pb_project.yaml file, a Feature Views model is included by default under the entities definition.
Schema has been migrated from 54 -> 61 in the project file.

Improvements

pb init connection now fetches connection name from the site configuration file, and asks you to specify one as well. If you don’t enter any value and there’s only one entry in the file, then it picks that value.
Default ID stitcher now uses all package models for ID sources as well, in addition to local ones.
Relevant error is now thrown, if ID types for an entity has not been specified in the pb_project.yaml file.
An entity project in the RudderStack dashboard no longer fails in case a few columns are missing in the source tables.
Debugging is improved and the correct line number is displayed.
pb show models command now includes model dependencies along with some more statistics.
Few retoolings from the ground up.

Bug Fixes

Resolved bug related to failure when a git repo couldn’t be found in the RudderStack dashboard.
Resolved bug where default ID stitcher was getting created even if there were no ID edges.
Resolved bug where Redshift Profiles projects were failing if names of entity_var/input_var were longer than 47 characters.
Resolved bug where pb run --rebase_incremental command was taking edges from previous runs.
Resolved the issue where a project wasn’t compiling in case it referenced a package having inputs that weren’t defined in the application with the same name.