fix: 文档与代码清理 (Closes #164, #163)

[Qixuan112]

概述

修复 Issue #164 和 #163 的文档错误与代码冗余问题。

---

Issue #164 - 文档与代码清理（6个问题）

修复内容

1. README 仓库链接错误 ✅

   • README.md: Tale_ai → Tale-AI
   • docs/README.zh.md: Tale_ai → Tale-AI

2. webui/app.py 重复 logger 初始化 ✅

   • 删除第 38-39 行重复的 logger 初始化

3. Flask 3.x 配置失效 ✅

   • app.config['JSON_AS_ASCII'] = False → app.json.ensure_ascii = False
   • 兼容 Flask 3.x JSON provider API

4. docs/config-guide.md 文档不一致 ✅

   • type: chat → type: llm
   • 与实际配置 schema 一致

5. core/function_caller.py 冗余死代码 ✅

   • 删除 Priority 2 硬编码 handlers（81行）
   • 统一使用 handler registry

6. core/parse_xml.py 降级路径结构不一致 ✅

   • _fallback_extract 补充 "session_sends": []
   • 与 _parse_root 结构对齐

---

Issue #163 - 代码冗余清理

修复内容

• 移除 __import__ 动态导入 ✅
  • core/llm/toolllm.py: 使用已导入的 get_registry() 替代 __import__(...)
  • 提升代码可读性和维护性

新增测试

• tests/test_available_tools.py: 4个单元测试验证 AVAILABLE_TOOLS 功能

---

代码统计

• 删除：92行
• 新增：95行
• 净变化：+3行

测试状态

✅ 所有测试通过

• Issue [代码冗余] core/llm/toolllm.py: AVAILABLE_TOOLS 使用 __import__ 动态导入，已有静态导入可用 #163 新增测试：4/4 通过
• 现有测试：无回归

---

Closes #164
Closes #163

Summary by CodeRabbit

• New Features

  • Tool availability is now managed consistently through the unified tool registry, improving support for configured plugins and integrations.
  • Fallback message parsing now includes session send information for more consistent results.

• Documentation

  • Updated repository links in the English and Chinese setup guides.
  • Corrected the service type in the configuration example.

• Bug Fixes

  • Improved web interface handling of non-ASCII characters in JSON responses.
  • Added validation coverage for available tools and their configuration.

[coderabbitai]

📝 Walkthrough

Walkthrough

The PR removes duplicate hardcoded tool handlers, uses the imported tool registry directly, aligns fallback parsing and Flask JSON settings, changes send-failure logging, and updates repository and configuration documentation.

Changes

Cleanup and consistency alignment

Layer / File(s)	Summary
Tool registry dispatch alignment core/function_caller.py, core/llm/toolllm.py, tests/test_available_tools.py	Tool execution now falls through to plugin dispatch after registered handlers. AVAILABLE_TOOLS uses get_registry(). Tests validate tool structure and registry consistency.
Runtime output and logging consistency core/parse_xml.py, core/adapter/src/qq/adapter.py, core/adapter/src/websocket/adapter.py, webui/app.py	Fallback parsing adds session_sends. Failed sends log at info level. Flask uses app.json.ensure_ascii = False.
Repository and configuration documentation README.md, docs/README.zh.md, docs/config-guide.md	Repository URLs use Tale-AI. The service example uses type: llm.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Possibly related issues

Possibly related PRs

• Qixuan112/TaleAI#118: Both changes update send-failure logging in the QQ and WebSocket adapters.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Out of Scope Changes check	⚠️ Warning	The send failure log-level changes in the QQ and WebSocket adapters are not covered by the linked issue objectives.	Revert the unrelated log-level changes or link an issue that explicitly requires them.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly identifies the documentation and code cleanup covered by issues #164 and #163.
Linked Issues check	✅ Passed	The changes address the documented repository links, Flask configuration, cleanup, parser consistency, registry import, and related tests.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch fix/issues-164-163

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

Actionable comments posted: 1

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

webui/app.py (1)

37-38: 🩺 Stability & Availability | 🔴 Critical | ⚡ Quick win

Restore the module-level logger binding.

The change removes get_logger and logger = get_logger(__name__), but later code still references logger. Those paths will raise NameError and can fail request handling. Keep one logger initialization and remove only the duplicate initialization.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@webui/app.py` around lines 37 - 38, Restore the module-level logger binding
in webui/app.py by importing and calling get_logger once to initialize logger,
ensuring all existing logger references remain valid while removing any
duplicate initialization.

🧹 Nitpick comments (1)

core/function_caller.py (1)

218-227: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick win

Remove the unreachable Priority 2 _plugin_dispatch branch.

_plugin_dispatch is only populated by register_plugin_handler, which also adds the handler to _handler_registry. Since _handler_registry is checked first, the _plugin_dispatch fallback cannot be reached. Remove this branch and its duplicate comment/error handling to complete the migration cleanup.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@core/function_caller.py` around lines 218 - 227, Remove the unreachable
Priority 2 _plugin_dispatch branch from the function dispatch logic, including
its comment and exception handling. Preserve the existing _handler_registry path
and the final unknown-function fallback in the surrounding function.

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@core/parse_xml.py`:
- Line 145: Update parse_xml_msg so every return path, including empty-input
handling and XXE rejection, includes the session_sends field with the same
default value used by _fallback_extract. Preserve the existing result contents
and behavior for all other paths.

---

Outside diff comments:
In `@webui/app.py`:
- Around line 37-38: Restore the module-level logger binding in webui/app.py by
importing and calling get_logger once to initialize logger, ensuring all
existing logger references remain valid while removing any duplicate
initialization.

---

Nitpick comments:
In `@core/function_caller.py`:
- Around line 218-227: Remove the unreachable Priority 2 _plugin_dispatch branch
from the function dispatch logic, including its comment and exception handling.
Preserve the existing _handler_registry path and the final unknown-function
fallback in the surrounding function.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 96383962-6d12-470e-bb61-44531f2ab3f8

📥 Commits

Reviewing files that changed from the base of the PR and between e916d3b and 5945695.

📒 Files selected for processing (10)

• README.md
• core/adapter/src/qq/adapter.py
• core/adapter/src/websocket/adapter.py
• core/function_caller.py
• core/llm/toolllm.py
• core/parse_xml.py
• docs/README.zh.md
• docs/config-guide.md
• tests/test_available_tools.py
• webui/app.py

[coderabbitai]

🗄️ Data Integrity & Integration | 🟡 Minor | ⚡ Quick win

Add session_sends to every parser return path.

Line 145 fixes _fallback_extract, but parse_xml_msg still omits session_sends for empty input and XXE rejection returns. This leaves inconsistent result shapes. Add the field to those early returns or centralize result construction.

Proposed fix

-        return {"messages": [], "action": None, "actions": [], "plan": None}
+        return {
+            "messages": [],
+            "action": None,
+            "actions": [],
+            "plan": None,
+            "session_sends": [],
+        }

Apply the same field to the XXE rejection result.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@core/parse_xml.py` at line 145, Update parse_xml_msg so every return path,
including empty-input handling and XXE rejection, includes the session_sends
field with the same default value used by _fallback_extract. Preserve the
existing result contents and behavior for all other paths.