Skip to content

Release v0.1.9
Compare
Choose a tag to compare
@ramedina86 ramedina86 released this 03 Jul 23:21
· 1935 commits to master since this release
v0.1.9
3dade76
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.

PyArrow support

Streamsync now uses PyArrow to serialise and deserialise dataframes. This increases performance and allows the DataFrame component to accept not only Pandas dataframes, but also PyArrow tables. This allows interoperability with other libraries, such as Polars (using .to_arrow()). Note that hardcoding dataframes in the frontend, using JSON, is no longer supported.

Improved DataFrame component
The DataFrame component is now based on a high-performance grid and supports ordering, column resizing, search and download as CSV. It's based on Apache Arrow and Arquero. It remains performant with 1M+ rows.

Bug fixes

  • Image paths weren't updated correctly when linked to state. This has been fixed.

Minor updates

  • Custom visibility, which was previously restricted to state, can now be also set using context. This allows components inside a Repeater to be hidden based on a value in a dictionary.
  • Datetime.datetime is now serialisable
Assets 2
Loading