All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
- New
QueryChat.app()method enables quicker/easier chatting with a dataset. (#xx) - Enabled bookmarking by default in both
.app()and.server()methods. In latter case, you'll need to also specify thebookmark_store(either inshiny.App()orshiny.express.app_opts()) for it to take effect. (#xx)
- The entire functional API (i.e.,
init(),sidebar(),server(), etc) has been deprecated in favor of a new class/OOP API. Namely, the newQueryChat()class is now the recommended way to start (instead ofinit()), which has methods to replace old functions (e.g.,.sidebar(),.server(), etc). (#101) - The previously deprecated
create_chat_callbackparameter ofinit()was removed. (#101) querychat.querychat.QueryChatwas renamed toquerychat.querychat.QueriedValues. (#101)
-
The
.sqlquery and.titlereturned fromquerychat.server()are now reactive values, meaning you can now.set()their value, and.df()will update accordingly. (#98) -
Added
querychat.greeting()to help you create a greeting message for your querychat bot. (#87) -
Added
querychat_reset_dashboard()tool for easily resetting the dashboard filters when asked by the user. (#81)
-
Added rich tool UI support using shinychat development version and chatlas >= 0.11.1. (#67)
-
querychat's system prompt and tool descriptions were rewritten for clarity and future extensibility. (#90)
- Fixed another issue with data sources that aren't already narwhals DataFrames (#83)
- Fixed an issue with the query tool when used with SQLAlchemy data sources. (@npelikan #79)
-
querychat.init()now accepts aclientargument, replacing the previouscreate_chat_callbackargument. (#60)The
clientcan be:- a
chatlas.Chatobject, - a function that returns a
chatlas.Chatobject, - or a provider-model string, e.g.
"openai/gpt-4.1", to be passed tochatlas.ChatAuto().
If
clientis not provided, querychat will use theQUERYCHAT_CLIENTenvironment variable, which should be a provider-model string. If the envvar is not set, querychat uses OpenAI with the default model fromchatlas.ChatOpenAI(). - a
-
querychat.ui()now adds a.querychatclass to the chat container andquerychat.sidebar()adds a.querychat-sidebarclass to the sidebar, allowing for easier customization via CSS. (#68)
This first release of the querychat package.