fix: remove exclusive mode

[jrmi]

This MR fixes two minor comments from previous MRs:

• fix: broken mixed tool formats execution while using tool format #407 (comment)
• An assistant message with 'tool_calls' must be followed by tool messages responding to each 'tool_call_id' #381 (comment)

---

Important

Remove exclusive_mode and simplify tool format handling in gptme/tools/base.py, updating related logic in llm_anthropic.py and llm_openai.py.

• Behavior:
  • Remove exclusive_mode variable and related logic in gptme/tools/base.py.
  • Simplify tool format handling by removing exclusive_mode checks in ToolUse.iter_from_content().
• Functions:
  • Update _handle_tools() in llm_anthropic.py and llm_openai.py to use before_tool.strip() instead of before_tool.replace("\n", "").
• Tests:
  • Remove print() statement from test_message_conversion_with_tool_and_non_tool() in tests/test_llm_anthropic.py.

^{This description was created by for f725110. It will automatically update as commits are pushed.}

[ellipsis-dev]

👍 Looks good to me! Reviewed everything up to f725110 in 28 seconds

More details

• Looked at 79 lines of code in 4 files
• Skipped 0 files when reviewing.
• Skipped posting 4 drafted comments based on config settings.

1. gptme/llm/llm_anthropic.py:254

• Draft comment:
  The change from if before_tool: to if before_tool.strip(): ensures that only non-whitespace text is considered, which is a more accurate check.
• Reason this comment was not posted:
  Confidence changes required: 0%
  The change from if before_tool: to if before_tool.strip(): is a good improvement. It ensures that only non-whitespace text is considered, which is more accurate for the intended logic.

2. gptme/llm/llm_openai.py:305

• Draft comment:
  The change from if before_tool.replace("\n", ""): to if before_tool.strip(): ensures that only non-whitespace text is considered, which is a more accurate check.
• Reason this comment was not posted:
  Confidence changes required: 0%
  The change from if before_tool.replace("\n", ""): to if before_tool.strip(): is a good improvement. It ensures that only non-whitespace text is considered, which is more accurate for the intended logic.

3. gptme/tools/base.py:34

• Draft comment:
  The removal of the exclusive_mode variable is appropriate since it is no longer used in the code. This helps in cleaning up the codebase.
• Reason this comment was not posted:
  Confidence changes required: 0%
  The removal of the exclusive_mode variable is appropriate since it is no longer used in the code. This helps in cleaning up the codebase.

4. gptme/tools/base.py:344

• Draft comment:
  The removal of the exclusive_mode condition in the iter_from_content method is appropriate since exclusive_mode is no longer used. This simplifies the logic.
• Reason this comment was not posted:
  Confidence changes required: 0%
  The removal of the exclusive_mode condition in the iter_from_content method is appropriate since exclusive_mode is no longer used. This simplifies the logic.

Workflow ID: wflow_jM2bAISBPJzzekK6

---

You can customize Ellipsis with 👍 / 👎 feedback, review rules, user-specific overrides, quiet mode, and more.