v0.4.1 Release

[0.4.1] - 2022-05-20

Update

• Avoid serializing PyArrow Table to ipc format in pre_transform_spec (#111)
• Switch to mimalloc global allocator in the vegafusion-python-embed packages (#111)
• Update to DataFusion 8.0.0 (#110)

v0.4.0 Release

[0.4.0] - 2022-05-12

Added

• Add support for timeFormat and utcFormat Vega expression functions (#105)
• Add support for variance, variancep, stdev, and stdevp aggregate operations (#108)

Update

• Support consistent view of time-based visualizations across viewer timezones (#107)

v0.3.0 Release

[0.3.0] - 2022-04-22

Fixed

• Fixed dropped messages under load in VegaFusionWidget (#101)

Changed

• Lowered min pyarrow dependency version from 6 to 5 in vegafusion Python package (#102)

v0.2.0 Release

[0.2.0] - 2022-04-16

Fixed

• Fixed range selection on numeric string columns (#95)

Added

• Add support for pre-transforming datasets in a Vega specification (#97)

v0.1.0 Release

[0.1.0] - 2022-03-10

Added

• Extracted dropdown menu logic into new vegafusion-embed JavaScript package (#88). This makes it possible to use the menu when embedding VegaFusion charts in non-jupyter contexts. See example at https://github.com/vegafusion/vegafusion/tree/main/javascript/vegafusion-chart-editor.
• conda-forge packages are now available (#73)

Fixed

• Fixed "Ambiguous reference to field" error in certain Vega specifications (#85)
• Determine the "local" timezone on the client rather than the server (#79, #82). This fixes local timezone handling when server and client are in different time zones. It also improves the performance of UTC to Local timezone conversions.

Updated

• Update DataFusion to version 7 (#78)
• Improved performance of date parsing for many common date formats (#80)

v0.0.4

[0.0.4] - 2022-02-25

Added

• New VegaFusion server that makes the VegaFusion Runtime available over gRPC(-Web).
• New optional memory limit for cache. This can be used instead of, or in addition to, the existing capacity limit.

Fixed

• Improved reliability of the runtime cache in the presence of task interruptions

v0.0.2 trial license release

These packages have identical functionality as the AGPL 0.0.2 version, but are released under the Polyform Trial License with an expiration date of 2022-04-15 (license attached). With these packages, the download source link is not displayed in the chart's dropdown menu.

To install, first download the attached vegafusion_python-0.0.2-*.whl package that matches your Python version and operating system. Then also download the vegafusion_jupyter-0.0.2-py3-none-any.whl package.

Then use pip to install both packages

pip install vegafusion_python-0.0.2-*.whl
pip install vegafusion_jupyter-0.0.2-py3-none-any.whl

v0.0.2

Initial public release under the AGPLv3 license

0.0.1a0

First wheels from CI