0.6.6 (May 20, 2022)#

Added#

How to: Customize the app.
Support for calling atoti.scope.cumulative() with a time period window and some partitioning on multilevel date hierarchies.
Support for calling date_shift() on multilevel date hierarchies.
default_values parameter to create_table(), read_csv(), read_numpy(), read_pandas(), read_parquet(), read_spark(), and read_sql() (discussion #585).
Support for numeric arrays in create_parameter_simulation() (discussion #526).

Logout support for secured sessions (issue #386).
Drillthrough Widget filters (issue #359).

The filters on which a drillthrough is done can be changed without recreating the widget.#
Axis titles switch (issue #8).

The switch toggles the display of the axis titles.#
Dedicated Connection failed message when a widget in JupyterLab cannot connect to its session.

Context value keys queriesResultLimit.intermediateSize and queriesResultLimit.transientSize in atoti.Cube.shared_context, atoti.Cube.query() and atoti.Session.query_mdx(). Use queriesResultLimit.intermediateLimit and queriesResultLimit.transientLimit respectively instead.
Session’s certificate_authority parameter. Use atoti.HttpsConfig’s certificate_authority parameter instead.

atoti.scope.cumulative() on multilevel hierarchies (issue #422).
read_pandas()’s support for dataframes with some columns containing only null values (issue #547).
read_pandas()’s support for timezoned dates (issue #465).
read_pandas() and load_pandas()’ error when loading dataframes with some columns with dtype="category" (issue #467).
Inability to call atoti.Session.start_transaction() after a previous transaction failed.
Misleading “missing Graphviz” warning (issue #567).
No entry in atoti.Table.loading_report for charset encoding issue (issue #552).
Support special characters in atoti.Cube.schema and atoti.tables.Tables.schema (issue #567).
Cube creation from a Table containing empty strings (issue #324).