From 4f2591b50a1b08c10eef8aba492282a1ca2503b8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E7=8E=8B=E5=BE=B7=E5=AE=87?= <kvtodev@outlook.com>
Date: Tue, 22 Sep 2026 12:22:08 +0800
Subject: [PATCH] =?UTF-8?q?=E8=AF=B4=E6=98=8E=20GPT=20Image=202=20?=
 =?UTF-8?q?=E5=B0=BA=E5=AF=B8=E6=97=8F=E4=B8=8E=20provider=20=E7=99=BD?=
 =?UTF-8?q?=E5=90=8D=E5=8D=95=E7=9A=84=E5=8F=A3=E5=BE=84=E5=B7=AE=E5=BC=82?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

- 为 is_gpt_image_2_family_model 补充文档注释：尺寸钳制族刻意包含业务名 gpt-image-2.5
- 明确 resolve_image_provider 仍坚持只收 concrete key，两处白名单不同是设计意图而非遗漏
---
 .../platform-image/src/image_provider/protocol/request.rs   | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/server-rs/crates/platform-image/src/image_provider/protocol/request.rs b/server-rs/crates/platform-image/src/image_provider/protocol/request.rs
index f93f23705..23470a04e 100644
--- a/server-rs/crates/platform-image/src/image_provider/protocol/request.rs
+++ b/server-rs/crates/platform-image/src/image_provider/protocol/request.rs
@@ -130,6 +130,12 @@ pub fn resolve_image_provider(model: &str) -> Result<ImageProvider, &'static str
     }
 }
 
+/// 尺寸钳制族判定：回答「这个模型值按 GPT Image 2 的像素预算处理吗」。
+///
+/// 中文注释：这里刻意包含业务模型名 `gpt-image-2.5`，因为业务边界可能直接用它
+/// 计算尺寸；而 `resolve_image_provider` 回答的是「发给哪个 provider」，所以
+/// 必须先在任务边界把业务名解析成 concrete key、收到业务名即拒绝。两处白名单
+/// 口径不同是有意设计，不要为了“对齐”把业务名放进 provider 白名单。
 pub(crate) fn is_gpt_image_2_family_model(model: &str) -> bool {
     matches!(
         normalize_image_model(model),