Skip to content

Feature: semantic-diff for commit messages #36

Description

@qdrddr

Description

Currently, git diff captures pure text changes, which may include mostly irrelevant.
This consumes LLM tokens and creates commit message comments that bloat the message, potentially hiding important changes in toll pile of garage messages.

Suggested Solution

I’d like to have a octocode commit --salemantic-diff parameter when enabled, it would use diffsitter that is based on AST code changes and only would capture semantically meaningful and important changes.
https://github.com/afnanenayet/diffsitter

This idea could be extended:

By still leveraging the remaining less important changes that are captured as usual git diff, but then clearly identifying and separating them as less important, that we could ask LLM to pay less attention to and produce less detailed summary.

So I could get full picture of the changes, but have LLm focus with on summarizing with more details important chances. While keeping semantically not important formatting and spend less output LLM tokens on those (potentially even using another cheaper model).

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

No type
No fields configured for issues without a type.

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions