[v4.11.1] - 🛠️ 修复 QQ 官方机器人昵称与提及清洗 (#207)

Fix QQ official nickname and mention handling * fix: restore QQ official nicknames and sanitize mentions * fix: protect QQ official identities and whitespace * docs: update QQ official nickname guidance * chore: bump version to v4.11.1 * docs: remove obsolete PDF guidance * chore: finalize v4.11.1 release prep * fix(qqofficial): 修复 _sanitize_qq_official_mentions .strip() 吞换行，补充 allow_alphanumeric_user_ids 注释 - .strip() → .strip(' \t')：仅清除首尾空白/制表符，保留用户创作的 \n 换行 - allow_alphanumeric_user_ids：在参数定义处补充注释，说明该标记还控制 ID 正则化和回退显示名 - 响应 Sourcery 代码审查 #207 指出的问题 * chore: 删除不必要的描述 --------- Co-authored-by: SXP-Simon <sxp20061207@163.com>
2026-09-22 13:38:43 +00:00 · 2026-07-21 14:53:31 +08:00
parent 20019baa2c
commit f2bcd6801e
13 changed files with 442 additions and 41 deletions
@@ -1,6 +1,7 @@
 import asyncio
 import inspect
 import json
+from pathlib import Path
 from types import SimpleNamespace

 from src.domain.models.data_models import (
@@ -465,6 +466,48 @@ def test_non_qq_avatar_mentions_ignore_alphanumeric_bracket_text():
    assert "user-capsule" not in rendered_text


+def test_qq_official_avatar_mentions_render_alphanumeric_id_with_nickname():
+    generator = build_generator_without_io()
+
+    async def fake_avatar(*args, **kwargs):
+        return "data:image/png;base64,AAAA"
+
+    generator._get_user_avatar = fake_avatar
+    rendered = asyncio.run(
+        generator._render_mentions(
+            "成员 [A_OPENID] 发言",
+            avatar_url_getter=None,
+            user_analysis={"A_OPENID": {"nickname": "测试用户"}},
+            allow_alphanumeric_user_ids=True,
+        )
+    )
+
+    rendered_text = str(rendered)
+    assert "测试用户" in rendered_text
+    assert "[A_OPENID]" not in rendered_text
+
+
+def test_qq_official_avatar_mentions_hide_placeholder_openid():
+    generator = build_generator_without_io()
+
+    async def fake_avatar(*args, **kwargs):
+        return "data:image/png;base64,AAAA"
+
+    generator._get_user_avatar = fake_avatar
+    rendered = asyncio.run(
+        generator._render_mentions(
+            "成员 A_OPENID 发言",
+            avatar_url_getter=None,
+            user_analysis={"A_OPENID": {"nickname": "A_OPENID"}},
+            allow_alphanumeric_user_ids=True,
+        )
+    )
+
+    rendered_text = str(rendered)
+    assert "A_OPENID" not in rendered_text
+    assert "群友" in rendered_text
+
+
 def test_mentions_support_alphanumeric_openid_and_hide_text():
    generator = build_generator_without_io()
    openid = "A1B2C3D4_OPENID"
@@ -546,3 +589,47 @@ def test_html_sidecar_export_removes_nested_identity_values():
    assert (
        sanitized["statistics"]["activity_visualization"]["user_activity_ranking"] == []
    )
+
+
+def test_qq_official_html_sidecar_uses_identity_sanitizer(tmp_path):
+    class HtmlConfig(FakeConfig):
+        def get_html_output_dir(self):
+            return str(tmp_path)
+
+        def get_html_filename_format(self):
+            return "report.html"
+
+    generator = build_generator_without_io()
+    generator.config_manager = HtmlConfig()
+    generator.html_templates = SimpleNamespace(
+        render_template=lambda *args, **kwargs: "<html>safe</html>"
+    )
+    generator._reuse_avatars_in_final_html = lambda html_content, *args: html_content
+
+    async def fake_prepare_render_data(*args, **kwargs):
+        return {}
+
+    generator._prepare_render_data = fake_prepare_render_data
+    openid = "A1B2C3D4_OPENID"
+    analysis_result = {
+        "statistics": {
+            "golden_quotes": [],
+            "activity_visualization": {"user_activity_ranking": [{"user_id": openid}]},
+        },
+        "topics": [],
+        "user_titles": [],
+        "user_analysis": {openid: {"nickname": "测试群友"}},
+        "summary": f"{openid} 最活跃",
+    }
+
+    _, json_path = asyncio.run(
+        generator.generate_html_report(
+            analysis_result,
+            "GROUP_OPENID",
+            allow_alphanumeric_user_ids=True,
+        )
+    )
+
+    assert json_path is not None
+    exported = Path(json_path).read_text(encoding="utf-8")
+    assert openid not in exported
@@ -18,6 +18,14 @@ class FakeHistoryManager:
            raise RuntimeError("temporary database failure")


+class RecordingHistoryManager:
+    def __init__(self):
+        self.calls = []
+
+    async def insert(self, **kwargs):
+        self.calls.append(kwargs)
+
+
 class FakeGroupRegistry:
    def __init__(self):
        self.upsert_calls = 0
@@ -27,14 +35,28 @@ class FakeGroupRegistry:


 class FakeOfficialEvent:
-    def __init__(self):
+    def __init__(
+        self,
+        text="hello",
+        mentions=None,
+        platform_name="qq_official",
+        at_target=None,
+    ):
+        self.platform_name = platform_name
+        message = []
+        if at_target:
+            message.append(SimpleNamespace(type="At", qq=at_target, name=""))
+        message.append(SimpleNamespace(type="Plain", text=text))
        self.message_obj = SimpleNamespace(
            message_id="OFFICIAL-MSG-1",
-            raw_message=SimpleNamespace(timestamp=1710000000),
+            raw_message=SimpleNamespace(
+                timestamp=1710000000,
+                mentions=list(mentions or []),
+            ),
            sender=SimpleNamespace(nickname=""),
-            message=[SimpleNamespace(type="Plain", text="hello")],
+            message=message,
        )
-        self.message_str = "hello"
+        self.message_str = text

    def get_group_id(self):
        return "GROUP_OPENID"
@@ -49,7 +71,7 @@ class FakeOfficialEvent:
        return "official-main"

    def get_platform_name(self):
-        return "qq_official"
+        return self.platform_name


 def test_failed_history_insert_releases_official_message_id():
@@ -68,3 +90,109 @@ def test_failed_history_insert_releases_official_message_id():

    assert history_manager.insert_calls == 2
    assert registry.upsert_calls == 1
+
+
+def test_new_qq_official_message_replaces_mentions_before_storage():
+    history_manager = RecordingHistoryManager()
+    registry = FakeGroupRegistry()
+    service = MessageProcessingService(
+        SimpleNamespace(message_history_manager=history_manager), registry
+    )
+    event = FakeOfficialEvent(
+        text="请问 <@KNOWN_OPENID> 和 <@!UNKNOWN_OPENID> 怎么看",
+        mentions=[
+            SimpleNamespace(
+                id="KNOWN_OPENID",
+                username="随风潜入夜",
+                is_you=False,
+            )
+        ],
+    )
+
+    asyncio.run(service.process_message(event))
+
+    stored_parts = history_manager.calls[0]["content"]["message"]
+    assert stored_parts == [
+        {"type": "plain", "text": "请问 @随风潜入夜 和 @群友 怎么看"}
+    ]
+    assert "KNOWN_OPENID" not in stored_parts[0]["text"]
+    assert "UNKNOWN_OPENID" not in stored_parts[0]["text"]
+
+
+def test_qq_official_bot_mention_is_removed_before_storage():
+    history_manager = RecordingHistoryManager()
+    service = MessageProcessingService(
+        SimpleNamespace(message_history_manager=history_manager), FakeGroupRegistry()
+    )
+    event = FakeOfficialEvent(
+        text="<@BOT_OPENID> 帮我问问 <@MEMBER_OPENID>",
+        mentions=[
+            SimpleNamespace(id="BOT_OPENID", username="机器人", is_you=True),
+            SimpleNamespace(
+                id="MEMBER_OPENID",
+                username="群友甲",
+                is_you=False,
+            ),
+        ],
+    )
+
+    asyncio.run(service.process_message(event))
+
+    stored_parts = history_manager.calls[0]["content"]["message"]
+    assert stored_parts == [{"type": "plain", "text": "帮我问问 @群友甲"}]
+
+
+def test_qq_official_message_preserves_internal_line_breaks():
+    history_manager = RecordingHistoryManager()
+    service = MessageProcessingService(
+        SimpleNamespace(message_history_manager=history_manager), FakeGroupRegistry()
+    )
+    event = FakeOfficialEvent(text="第一行\n\n第二行\t\t结尾")
+
+    asyncio.run(service.process_message(event))
+
+    stored_parts = history_manager.calls[0]["content"]["message"]
+    assert stored_parts == [{"type": "plain", "text": "第一行\n\n第二行 结尾"}]
+
+
+def test_qq_official_at_component_preserves_internal_line_breaks():
+    history_manager = RecordingHistoryManager()
+    service = MessageProcessingService(
+        SimpleNamespace(message_history_manager=history_manager), FakeGroupRegistry()
+    )
+    event = FakeOfficialEvent(
+        text="第一行\n\n第二行",
+        mentions=[SimpleNamespace(id="BOT_OPENID", username="机器人", is_you=True)],
+        at_target="BOT_OPENID",
+    )
+
+    asyncio.run(service.process_message(event))
+
+    stored_parts = history_manager.calls[0]["content"]["message"]
+    assert stored_parts == [
+        {"type": "at", "target_id": "BOT_OPENID", "name": ""},
+        {"type": "plain", "text": "第一行\n\n第二行"},
+    ]
+
+
+def test_non_qq_message_keeps_platform_mention_syntax_unchanged():
+    history_manager = RecordingHistoryManager()
+    service = MessageProcessingService(
+        SimpleNamespace(message_history_manager=history_manager), FakeGroupRegistry()
+    )
+    event = FakeOfficialEvent(
+        text="请问 <@DISCORD_USER_ID> 怎么看",
+        mentions=[
+            SimpleNamespace(
+                id="DISCORD_USER_ID",
+                username="Discord 用户",
+                is_you=False,
+            )
+        ],
+        platform_name="discord",
+    )
+
+    asyncio.run(service.process_message(event))
+
+    stored_parts = history_manager.calls[0]["content"]["message"]
+    assert stored_parts == [{"type": "plain", "text": "请问 <@DISCORD_USER_ID> 怎么看"}]
@@ -18,11 +18,11 @@ class FakeHistoryManager:
        return self.pages.get(page, [])


-def make_record(record_id, message_id, sender_id, timestamp, text):
+def make_record(record_id, message_id, sender_id, timestamp, text, sender_name=None):
    return SimpleNamespace(
        id=record_id,
        sender_id=sender_id,
-        sender_name=sender_id,
+        sender_name=sender_id if sender_name is None else sender_name,
        created_at=datetime.fromtimestamp(timestamp, timezone.utc),
        content={
            "type": "user",
@@ -61,8 +61,8 @@ def test_local_history_is_deduplicated_filtered_and_sorted():
            message_history_manager=FakeHistoryManager(
                {
                    1: [
-                        make_record(1, "MSG-2", "B_OPENID", 200, "second"),
-                        make_record(2, "MSG-1", "A_OPENID", 100, "first"),
+                        make_record(1, "MSG-2", "B_OPENID", 200, "second", "用户乙"),
+                        make_record(2, "MSG-1", "A_OPENID", 100, "first", "用户甲"),
                        make_record(3, "MSG-2", "B_OPENID", 200, "duplicate"),
                        make_record(4, "MSG-BOT", "BOT_OPENID", 300, "bot"),
                    ]
@@ -77,8 +77,53 @@ def test_local_history_is_deduplicated_filtered_and_sorted():

    assert [message.message_id for message in messages] == ["MSG-1", "MSG-2"]
    assert [message.sender_id for message in messages] == ["A_OPENID", "B_OPENID"]
+    assert [message.sender_name for message in messages] == ["用户甲", "用户乙"]
    assert [message.text_content for message in messages] == ["first", "second"]

+    raw_messages = adapter.convert_to_raw_format(messages)
+    assert [message["sender"]["nickname"] for message in raw_messages] == [
+        "用户甲",
+        "用户乙",
+    ]
+
+
+def test_local_history_sender_name_uses_group_scoped_alias_for_placeholder():
+    adapter = make_adapter()
+    empty_name_record = make_record(
+        1,
+        "MSG-1",
+        "A_OPENID",
+        100,
+        "first",
+        sender_name="",
+    )
+    openid_name_record = make_record(
+        2,
+        "MSG-2",
+        "A_OPENID",
+        101,
+        "second",
+        sender_name="A_OPENID",
+    )
+
+    empty_name_message = adapter._convert_history_record(
+        empty_name_record, "GROUP_OPENID"
+    )
+    openid_name_message = adapter._convert_history_record(
+        openid_name_record, "GROUP_OPENID"
+    )
+    other_group_message = adapter._convert_history_record(
+        openid_name_record, "OTHER_GROUP_OPENID"
+    )
+
+    assert empty_name_message is not None
+    assert openid_name_message is not None
+    assert other_group_message is not None
+    assert empty_name_message.sender_name.startswith("群友-")
+    assert "A_OPENID" not in empty_name_message.sender_name
+    assert openid_name_message.sender_name == empty_name_message.sender_name
+    assert other_group_message.sender_name != empty_name_message.sender_name
+

 def test_factory_registers_both_official_platform_types():
    assert PlatformAdapterFactory.is_supported("qq_official")