nGQL parser crashes on character 标 (U+6807) in tag/edge names

[gaogao94]

Description

The nGQL parser crashes when creating tags or edges whose names contain the Chinese character 标 (U+6807, UTF-8 bytes: e6 a0 87).

Reproduce

CREATE TAG IF NOT EXISTS 待续目标(描述 string);

Expected behavior

Tag 待续目标 is created successfully.

Actual behavior

nebula-console returns:

[ERROR (-1004)]: SyntaxError: syntax error near `标(描�`

Python client (nebula3) throws:

UnicodeDecodeError: utf-8 codec can't decode byte 0xe8 in position 39: invalid continuation byte

Notes

• Other Chinese characters in the same name work fine: 待续目 (without 标) succeeds
• 测试目标 and 续卡目标 also fail — confirming 标 is the trigger
• Single character 目 (U+76EE) works; single character 标 fails
• The response bytes appear corrupted, suggesting the parser's buffer is damaged when encountering this character
• Reproduced on both nebula-console and Python client, so the bug is server-side (graphd)

Environment

• NebulaGraph version: 3.8.0
• Deployment: single-node Docker Compose (storaged0)
• Client: nebula-console (bundled with 3.8.0) and nebula3 Python client