fix(provider/gemini): Prevent empty parts in assistant messages #194
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR addresses an issue where cancelling a chat request and promptly starting a new one could lead to an
Error 400, Message: * GenerateContentRequest.contents[38].parts: contents.parts must not be empty.
when using the Gemini provider.The root cause was that assistant messages (model responses) were being included in the conversation history sent to the Gemini API even if they were empty (i.e., contained no text content or tool calls). This typically occurred during rapid interactions like cancelling and restarting a chat.
The fix modifies the
convertMessages
function ininternal/llm/provider/gemini.go
to ensure that an assistant message is only added to the history ifmsg.Content().String() != ""
orlen(msg.ToolCalls()) > 0
. This prevents emptyparts
from being sent to the Gemini API.Changes:
internal/llm/provider/gemini.go
:To Test:
opencode
application (go build -o opencode
).