free-short-video 7.0.0 → 7.0.2

package/.env.example

@@ -36,11 +36,15 @@ HOST=0.0.0.0
 PORT=8765
 
 # 注：Agnes API 走哪个站点（国际站 / 国内站）**不通过环境变量配置**
-# （AGNES_BASE_URL 已废弃，代码中不再读取）。请在 Web 设置页的 API Key 面板
-# 选择域名，或用「自动探测域名」逐 Key 补全匹配域名（每个 Key 可绑定各自域名）。
+# （AGNES_BASE_URL 已废弃，代码中不再读取）。请在 Web 设置页的 API Key 面板选择全局默认域名。
+# ⚠️ 重要：通过本文件 / 环境变量写入的 Key 属于 env 来源，**无法逐个绑定域名，也不参与
+#    Web 页的「自动探测域名」**——它们一律使用你在 Web 页设定的「全局默认域名」。因此：
+#      - 单站点 Key：把全局默认域名设成与该 Key 站点一致即可；
+#      - 混用不同站点的 Key：请改到 Web 配置页逐个添加（config 来源），再逐 Key 绑定域名
+#        或点「自动探测域名」，才能各走各的站点（env Key 做不到这点）。
+#    跨站使用 Key（如国际站 Key 打到国内站专属域名）会得到 HTTP 401 / 无效的令牌。
 # 可选域名：com = apihub.agnes-ai.com（国际站）、cn = api.agnes-ai.cn（国内站，
-# 需国内站专属 Key）、cn_bak = apihub.agnes-ai.cn（备用，国际站 Key 亦可用）。
-# 跨站使用 Key 会得到 HTTP 401 / 无效的令牌。
+# 需国内站专属 Key）、cn_bak = apihub.agnes-ai.cn（备用，国内与国际 Key 均可用）。
 
 
 # ── 4. 限速（可选）─────────────────────────────────────────────────

package/README.md

@@ -1,56 +1,23 @@
 ---
 
-# What's New in v7.0.0
+# What's New in v7.0.2
 
 ## What's New
 
 ### Features & Improvements
 
-- **Custom text model providers (OpenAI-compatible / Anthropic-compatible).** Configure any
-  LLM endpoint with `base URL + API key` and two supported wire protocols
-  (`openai-completions`, `anthropic-messages`). Once selected, every text call — screenplay
-  breakdown, scene planning, poetry segmentation, image-prompt rewriting — is routed to your
-  provider automatically. The built-in Agnes provider is always available, cannot be deleted,
-  and keeps the previous behaviour when no custom provider is selected.
-- **Provider management panel.** A dedicated settings section lists all text providers, lets
-  you add, edit and delete them, and keeps the Agnes key/domain editing inside the provider
-  modal. Keys are masked in the UI, and the provider's model list is a single editable list:
-  fetch it from the endpoint or type model ids by hand. Model selection is a two-level
-  provider → model picker that merges the models of all providers.
-- **Interface-language aware backend messages.** Task progress messages, failure panels, HTTP
-  error details and the diagnostics report are now localized according to the UI language
-  instead of being hard-coded Chinese. The UI language travels with every request
-  (`X-Agnes-UI-Lang`, falling back to `Accept-Language`), is snapshotted into each task at
-  creation time, and therefore does not drift if the user switches language later. Chinese and
-  English are covered; the remaining languages fall back to the existing behaviour.
-- **Clearer network failure diagnosis.** TLS handshake resets (`connection reset`) are now
-  attributed to local-connection problems, and the diagnostics report shows the interface
-  language of the reporting user, making failure reports easier to triage.
-
-### Refactoring & Optimizations
-
-- **Decoupled text-generation layer.** A new protocol-client layer
-  (`core/api/providers/` → base / OpenAI / Anthropic) plus a provider factory replaces the two
-  hard-coded Agnes chat client constructions, so pipelines and the screenwriter are untouched
-  while the underlying model source becomes pluggable. Custom providers reuse the existing
-  shared rate limiter and exponential-backoff retry policy, and failures are recorded in the
-  same error log as before.
-- **Backend i18n runtime.** A new translation runtime plus request-scoped language middleware
-  gives the FastAPI side the same localization guarantees the frontend already had, with a
-  safe fallback chain that never turns a missing translation into a request failure.
+* **Clearer key/domain setup to prevent `401`.** The API Key panel now states, per source, that keys added in the web UI can be bound to their own domain (and are covered by *Auto-detect domains*), while `.env` / environment-variable keys always follow the single **global default domain** and cannot be bound individually. Each `.env` key now shows a neutral "follows global domain" badge with the resolved endpoint. A new **multi-key** note explains that a single mismatched or expired key makes tasks intermittently fail with `401` whenever that key is picked. Getting Started and the FAQ gained a "match the key to the right site" section, and `.env.example` now spells out the `.env`-key behavior — so single-site users set the domain and mixed-site users move keys into the web config page.
+* **True 16:9 landscape.** Landscape (16:9) video now renders at the correct aspect ratio, and the portrait / large-landscape resolution option labels were fixed so the listed size matches what is actually generated.
+* **Refreshed language switcher.** The native browser language dropdown was replaced with a site-styled language switcher, consistent with the rest of the UI across all 22 locales.
 
 ### Bug Fixes
 
-- Model pulling for an already saved provider now reuses its stored credentials instead of
-  falling back to the Agnes key.
-- The built-in Agnes provider no longer appears twice in the provider picker and is now
-  labelled as text-only.
-- Non-Chinese interfaces no longer receive hard-coded Chinese network-diagnosis and task
-  messages (issue #64).
+* Fixed the config page recommending "bind a matching domain for each key" on `.env`-sourced keys, which have no domain dropdown and cannot be bound — the misleading hint that repeatedly sent users down the wrong path when chasing `401` errors. The *Auto-detect domains* tooltip now correctly states it only probes web-UI keys.
+* Corrected the Docker/compose image reference to the real flat GHCR namespace, so documented pull commands actually resolve.
 
 ---
 
-No action is required after upgrading.
+No action is required after upgrading. If you were hitting `401`, re-check your key against its site in the API Key panel or run *Auto-detect domains*.
 
 ---
 

package/core/api/agnes_video.py

@@ -15,10 +15,10 @@ from core.api.error_collector import collect_error, collect_error_from_exception
 from core.api.key_manager import get_key_ring
 from core.api.rate_limiter import get_rate_limiter, get_video_submit_limiter
 from core.config import (
-    VIDEO_ASPECT_RATIOS,
     get_agnes_api_root,
     get_base_url_for_key,
     is_v25_video_model,
+    width_height_to_aspect_ratio,
 )
 from utils.image_normalizer import normalize_reference_path
 from utils.video import download_video
@@ -271,7 +271,7 @@ class AgnesVideoAPI:
             return 961   # 480p tier
 
     def _get_frame_config(self, duration: Optional[int] = None,
-                          width: int = 1152, height: int = 768) -> tuple:
+                          width: int = 1280, height: int = 720) -> tuple:
         d = duration or self.default_duration
         max_nf = self._get_max_frames(width, height)
         if d in DURATION_PRESETS:
@@ -563,8 +563,8 @@ class AgnesVideoAPI:
         prompt: str,
         reference_image_paths: List[str] = [],
         duration: Optional[int] = None,
-        width: int = 1152,
-        height: int = 768,
+        width: int = 1280,
+        height: int = 720,
         seed: Optional[int] = None,
         negative_prompt: Optional[str] = None,
         progress_callback=None,
@@ -587,8 +587,8 @@ class AgnesVideoAPI:
         prompt: str,
         reference_image_paths: List[str] = [],
         duration: Optional[int] = None,
-        width: int = 1152,
-        height: int = 768,
+        width: int = 1280,
+        height: int = 720,
         seed: Optional[int] = None,
         negative_prompt: Optional[str] = None,
         **kwargs,
@@ -649,31 +649,21 @@ class AgnesVideoAPI:
 
     @staticmethod
     def _width_height_to_aspect_ratio(width: int, height: int) -> str:
-        """将像素宽高映射到最接近的 2.5 系列 aspect_ratio 枚举。
+        """将像素宽高映射到 2.5 系列 aspect_ratio 枚举（委托 ``core.config``）。
 
-        找不到精确比例时取误差最小的档位（默认 16:9）。
+        映射规则已上提到 ``core.config.width_height_to_aspect_ratio``，供回归
+        校验脚本共用（保证「提交」与「校验」同一套规则）；此方法保留为类内
+        调用入口，行为与上提前完全一致。
         """
-        if not width or not height:
-            return "16:9"
-        ratio = width / height
-        best = "16:9"
-        best_dist = float("inf")
-        for ar in VIDEO_ASPECT_RATIOS:
-            w, h = ar.split(":")
-            target = int(w) / int(h)
-            dist = abs(ratio - target)
-            if dist < best_dist:
-                best_dist = dist
-                best = ar
-        return best
+        return width_height_to_aspect_ratio(width, height)
 
     async def _submit_video_v25(
         self,
         prompt: str,
         reference_image_paths: List[str],
         duration: Optional[int] = None,
-        width: int = 1152,
-        height: int = 768,
+        width: int = 1280,
+        height: int = 720,
         seed: Optional[int] = None,
         **kwargs,
     ) -> str:

package/core/config.py

@@ -18,7 +18,7 @@ CONFIG_FILE = os.path.join(CONFIG_DIR, "config.json")
 # ═══════════════════════════════════════════════════
 # 应用版本号（v6.1 新增：发版时同步更新，见 docs/dev/release_process.md）
 # ═══════════════════════════════════════════════════
-APP_VERSION = "7.0.0"
+APP_VERSION = "7.0.2"
 
 # 未配置 API Key 时的统一报错文案（含免费获取与在线体验兜底，全站路由共用）
 #
@@ -903,7 +903,7 @@ def set_watermark_config(enabled: bool = None, language: str = None):
 
 VIDEO_RESOLUTION_PRESETS = {
     "portrait": {"width": 768, "height": 1152, "label": "竖屏 9:16"},
-    "landscape": {"width": 1152, "height": 768, "label": "横屏 16:9"},
+    "landscape": {"width": 1280, "height": 720, "label": "横屏 16:9"},
     "square": {"width": 1024, "height": 1024, "label": "方形 1:1"},
 }
 
@@ -1085,6 +1085,63 @@ def is_v25_video_model(model: str) -> bool:
 # 支持的画幅比例（2.5 系列 aspect_ratio 枚举）
 VIDEO_ASPECT_RATIOS = ["21:9", "16:9", "4:3", "1:1", "3:4", "9:16"]
 
+# ── 项目标准像素档 → aspect_ratio 显式映射（语义优先于数值最近邻）──
+# 背景：2.5 系列只接受上面 6 档比例枚举，而非标准比例走「数值最近邻」会归错档：
+#   · 1152x768（3:2, 1.500）距 4:3(1.333) 0.167 < 距 16:9(1.778) 0.278 → 误判 4:3
+#   · 768x1152（2:3, 0.667）距 3:4(0.750) 0.083 < 距 9:16(0.563) 0.104 → 误判 3:4
+# 改度量方式（比值 / 对数距离）救不了这两个值——2:3 在几何上确实更贴近 3:4。
+# 但它们在前端就是「竖屏 9:16」「横屏 16:9」两个语义明确的惯例像素档，故显式查表，
+# 不让数值距离决定。表内其余档位与最近邻结果一致，仅作显式化，不改变既有行为。
+STANDARD_PIXEL_TO_RATIO = {
+    # 竖屏
+    (768, 1152): "9:16",
+    (720, 1280): "9:16",
+    (768, 1344): "9:16",
+    (1024, 1792): "9:16",
+    # 横屏
+    (1152, 768): "16:9",
+    (1280, 720): "16:9",
+    (1344, 768): "16:9",
+    (1792, 1024): "16:9",
+    # 方形与超宽
+    (1024, 1024): "1:1",
+    (960, 720): "4:3",
+    (720, 960): "3:4",
+    (1680, 720): "21:9",
+}
+
+
+def width_height_to_aspect_ratio(width: int, height: int) -> str:
+    """将像素宽高映射到 2.5 系列 aspect_ratio 枚举。
+
+    映射顺序：
+
+    1. 项目标准像素档显式查表（``STANDARD_PIXEL_TO_RATIO``）——语义优先，避免
+       ``768x1152``（竖屏 9:16）/ ``1152x768``（横屏 16:9）这类项目惯例档被
+       数值最近邻误判成 ``3:4`` / ``4:3``；
+    2. 未命中时取误差最小的档位（默认 ``16:9``）。
+
+    供 API 客户端（提交时推导 aspect_ratio）与回归校验脚本（校验时推导期望
+    比例）共用，保证「提交」与「校验」使用同一套映射规则。
+    """
+    if not width or not height:
+        return "16:9"
+    exact = STANDARD_PIXEL_TO_RATIO.get((width, height))
+    if exact:
+        return exact
+    ratio = width / height
+    best = "16:9"
+    best_dist = float("inf")
+    for ar in VIDEO_ASPECT_RATIOS:
+        w, h = ar.split(":")
+        target = int(w) / int(h)
+        dist = abs(ratio - target)
+        if dist < best_dist:
+            best_dist = dist
+            best = ar
+    return best
+
+
 # 2.5 系列时长档位（seconds 字符串 "4"–"12"）
 VIDEO_25_DURATIONS = [4, 5, 6, 8, 10, 12]
 
@@ -1104,7 +1161,7 @@ VIDEO_MODEL_CAPABILITIES = {
             "type": "pixels",
             "options": [
                 {"value": "768x1152", "label": {"zh": "竖屏 768x1152", "en": "Portrait 768x1152"}},
-                {"value": "1152x768", "label": {"zh": "横屏 1152x768", "en": "Landscape 1152x768"}},
+                {"value": "1280x720", "label": {"zh": "横屏 1280x720", "en": "Landscape 1280x720"}},
                 {"value": "1024x1024", "label": {"zh": "方形 1024x1024", "en": "Square 1024x1024"}},
             ],
         },