Commit Graph

19 Commits

Author SHA1 Message Date
1996300c40 Hopeful fix to race condition in tui's streamed response handling 2024-12-11 07:17:53 +00:00
ec21a02ec0 Tweaks/cleanups to conversation management in tui
- Pass around message/conversation values instead of pointers where it
makes more sense, and store values instead of pointers in the globally
(within the TUI) shared `App` (pointers provide no utility here).

- Split conversation persistence into separate conversation/message
  saving stages
2024-10-25 16:57:15 +00:00
0384c7cb66 Large refactor - it compiles!
This refactor splits out all conversation concerns into a new
`conversation` package. There is now a split between `conversation` and
`api`s representation of `Message`, the latter storing the minimum
information required for interaction with LLM providers. There is
necessary conversation between the two when making LLM calls.
2024-10-22 17:53:13 +00:00
3ec2675632 tui: Error handling tweak
Moved errors to bottom of screen, fix infinite loop by typing errors
properly
2024-09-23 04:43:26 +00:00
676aa7b004 Refactor TUI rendering handling and general cleanup
Improves render handling by moving the responsibility of laying out the
whole UI from each view and into the main `tui` model. Our `ViewModel`
interface has now diverged from bubbletea's `Model` and introduces
individual `Header`, `Content`, and `Footer` methods for rendering those
UI elements.

Also moved away from using value receivers on our Update and View
functions (as is common across Bubbletea) to pointer receivers, which
cleaned up some of the weirder aspects of the code (e.g. before we
essentially had no choice but to do our rendering in `Update` in order
to calculate and update the final height of the main content's
`viewport`).
2024-09-23 02:49:08 +00:00
a488ec4fd8 Fixed message loading
Root messages weren't being loaded since the refactor, and there was
dead code
2024-09-21 02:32:54 +00:00
463ca9ef40 TUI view management and input handling cleanup 2024-09-16 16:18:18 +00:00
24b5cdbbf6 More monior TUI refactor/cleanup
`tui/tui.go` is no longer responsible for passing window resize updates
to all views, instead we request a new window size message to be sent at
the same time we enter the view, allowing the view to catch and handle
it.

Add `Initialized` to `tui/shared/View` model, now we only call
`Init` on a view before entering it for the first time, rather than
calling `Init` on all views when the application starts.

Renames file, small cleanups
2024-09-16 14:04:08 +00:00
443c8096d3 TUI refactor
- Clean up, improved startup logic, initial conversation load
- Moved converation/message business logic (mostly) into `model/tui`
2024-09-16 00:48:45 +00:00
c30e652103 Cleaned up assistant cursor handling 2024-06-21 05:52:59 +00:00
3fde58b77d Package restructure and API changes, several fixes
- More emphasis on `api` package. It now holds database model structs
  from `lmcli/models` (which is now gone) as well as the tool spec,
  call, and result types. `tools.Tool` is now `api.ToolSpec`.
  `api.ChatCompletionClient` was renamed to
  `api.ChatCompletionProvider`.

- Change ChatCompletion interface and implementations to no longer do
  automatic tool call recursion - they simply return a ToolCall message
  which the caller can decide what to do with (e.g. prompt for user
  confirmation before executing)

- `api.ChatCompletionProvider` functions have had their ReplyCallback
  parameter removed, as now they only return a single reply.

- Added a top-level `agent` package, moved the current built-in tools
  implementations under `agent/toolbox`. `tools.ExecuteToolCalls` is now
  `agent.ExecuteToolCalls`.

- Fixed request context handling in openai, google, ollama (use
  `NewRequestWithContext`), cleaned up request cancellation in TUI

- Fix tool call tui persistence bug (we were skipping message with empty
  content)

- Now handle tool calling from TUI layer

TODO:
- Prompt users before executing tool calls
- Automatically send tool results to the model (or make this toggleable)
2024-06-21 05:24:02 +00:00
dfe43179c0 Include token count in api.Chunk
And calculate the tokens/chunk for gemini responses, fixing the tok/s
meter for gemini models.

Further, only consider the first candidate of streamed gemini responses.
2024-06-09 20:49:18 +00:00
42c3297e54 Make Conversation a pointer refernece on Message
Instead of a value, which lead some odd handling of conversation
references.

Also fixed some formatting and removed an unnecessary (and probably
broken) setting of ConversationID in a call to
`cmdutil.HandleConversationReply`
2024-06-09 18:51:44 +00:00
a22119f738 Better handling of newly saved conversations
When a new conversation is created in the chat view's
`persistConversation`, we now set `rootMessages` appropriately.
2024-06-09 18:51:44 +00:00
d2d946b776 Wrap chunk content in a Chunk type
Preparing to include additional information with each chunk (e.g. token
count)
2024-06-09 18:31:43 +00:00
c1ead83939 Rename shared.State to shared.Shared 2024-06-09 16:19:19 +00:00
c9e92e186e Chat view cleanup
Replace `waitingForReply` and the `status` string with the `state`
variable.
2024-06-09 16:19:17 +00:00
45df957a06 Fixes to message/conversation handling in tui chat view
This set of changes fixes root/child message cycling and ensures all
database operations happen within a `tea.Cmd`
2024-06-08 21:28:29 +00:00
136c463924 Split chat view into files 2024-06-02 22:40:46 +00:00