ci: guard FunClip v2.1.0 launch pages (#3417)

Add robust live-site contracts for the bilingual FunClip v2.1.0 launch pages, release routes, donor navigation, and interface image. Hidden evidence and zero-area links cannot satisfy the contract.
2026-10-12 02:19:18 +00:00 · 2026-07-25 00:04:08 +08:00
parent 8a34247dc5
commit a28ee6b39d
2 changed files with 395 additions and 49 deletions
@@ -22,11 +22,13 @@ class PageContract:
    visible_required: tuple[str, ...] = ()
    visible_patterns: tuple[tuple[str, str], ...] = ()
    required_links: tuple[str, ...] = ()
+    required_images: tuple[str, ...] = ()


@dataclass(frozen=True)
 class StaticAssetContract:
    signature: bytes
+    media_type: str
    min_bytes: int = 128


@@ -59,9 +61,7 @@ PAGE_CONTRACTS: dict[str, PageContract] = {
            ("FunASR 官方插件 0.1.1", r"FunASR 官方插件 0\.1\.1(?![\w.+-])"),
            ("最大 25 MB 音频上传", r"最大 (?<!\d)25 MB 音频上传"),
        ),
-        required_links=(
-            "https://marketplace.dify.ai/plugin/langgenius/funasr",
-        ),
+        required_links=("https://marketplace.dify.ai/plugin/langgenius/funasr",),
        forbidden=("16K+",),
    ),
    f"{BASE_URL}/en/ecosystem.html": PageContract(
@@ -76,9 +76,7 @@ PAGE_CONTRACTS: dict[str, PageContract] = {
            ("FunASR plugin 0.1.1", r"FunASR plugin 0\.1\.1(?![\w.+-])"),
            ("25 MB uploads", r"(?<!\d)25 MB uploads"),
        ),
-        required_links=(
-            "https://marketplace.dify.ai/plugin/langgenius/funasr",
-        ),
+        required_links=("https://marketplace.dify.ai/plugin/langgenius/funasr",),
        forbidden=("16K+",),
    ),
    f"{BASE_URL}/donors.html": PageContract(
@@ -275,48 +273,216 @@ PAGE_CONTRACTS: dict[str, PageContract] = {
        ),
        forbidden=("funasr==1.3.27", "runtime-llamacpp-v0.1.1"),
    ),
+    f"{BASE_URL}/blog/funclip-v2-1-0-video-clipping-release.html": PageContract(
+        required=(
+            "FunClip v2.1.0",
+            "FunClip-2.1.0.tar.gz",
+            "FunClip-2.1.0.zip",
+            "SHA256SUMS",
+            "funasr>=1.3.29",
+            "TwelveLabs Pegasus",
+        ),
+        visible_required=(
+            "FunClip v2.1.0",
+            "FunClip-2.1.0.tar.gz",
+            "FunClip-2.1.0.zip",
+            "SHA256SUMS",
+            "funasr>=1.3.29",
+            "TwelveLabs Pegasus",
+            "为什么发布的是源码归档，而不是 wheel",
+            "v2.1.0 的精确合并源码通过 50 项测试，另有 1 项跳过",
+            "干净 Python 3.12 环境中的 6 项发布契约测试",
+            "GitHub Actions 首次发布和主动重跑均成功",
+            "Star FunClip 仓库",
+        ),
+        required_links=(
+            "https://github.com/modelscope/FunClip",
+            "https://github.com/modelscope/FunClip/releases/tag/v2.1.0",
+            "https://github.com/modelscope/FunClip/releases/download/v2.1.0/FunClip-2.1.0.tar.gz",
+            "https://github.com/modelscope/FunClip/releases/download/v2.1.0/FunClip-2.1.0.zip",
+            "https://github.com/modelscope/FunClip/releases/download/v2.1.0/SHA256SUMS",
+            "https://huggingface.co/spaces/FunAudioLLM/FunClip",
+            "https://github.com/modelscope/FunASR/releases/tag/v1.3.29",
+            "/donors.html",
+        ),
+        required_images=("/img/funclip-v2-1-0-interface.jpg",),
+        forbidden=("funasr>=1.3.28",),
+    ),
+    f"{BASE_URL}/en/blog/funclip-v2-1-0-video-clipping-release.html": PageContract(
+        required=(
+            "FunClip v2.1.0",
+            "FunClip-2.1.0.tar.gz",
+            "FunClip-2.1.0.zip",
+            "SHA256SUMS",
+            "funasr>=1.3.29",
+            "TwelveLabs Pegasus",
+        ),
+        visible_required=(
+            "FunClip v2.1.0",
+            "FunClip-2.1.0.tar.gz",
+            "FunClip-2.1.0.zip",
+            "SHA256SUMS",
+            "funasr>=1.3.29",
+            "TwelveLabs Pegasus",
+            "Why these are source archives, not a wheel",
+            "The exact v2.1.0 merge passed 50 tests with 1 skipped",
+            "Six release-contract tests also passed in a clean Python 3.12 environment",
+            "Both the initial GitHub Actions release and a deliberate rerun succeeded",
+            "star the FunClip repository",
+        ),
+        required_links=(
+            "https://github.com/modelscope/FunClip",
+            "https://github.com/modelscope/FunClip/releases/tag/v2.1.0",
+            "https://github.com/modelscope/FunClip/releases/download/v2.1.0/FunClip-2.1.0.tar.gz",
+            "https://github.com/modelscope/FunClip/releases/download/v2.1.0/FunClip-2.1.0.zip",
+            "https://github.com/modelscope/FunClip/releases/download/v2.1.0/SHA256SUMS",
+            "https://huggingface.co/spaces/FunAudioLLM/FunClip",
+            "https://github.com/modelscope/FunASR/releases/tag/v1.3.29",
+            "/en/donors.html",
+        ),
+        required_images=("/img/funclip-v2-1-0-interface.jpg",),
+        forbidden=("funasr>=1.3.28",),
+    ),
 }


 PNG_SIGNATURE = b"\x89PNG\r\n\x1a\n"
+JPEG_SIGNATURE = b"\xff\xd8\xff"

 STATIC_ASSET_CONTRACTS: dict[str, StaticAssetContract] = {
-    f"{BASE_URL}/img/banner.4f436d19.png": StaticAssetContract(PNG_SIGNATURE),
-    f"{BASE_URL}/logo.png": StaticAssetContract(PNG_SIGNATURE),
+    f"{BASE_URL}/img/banner.4f436d19.png": StaticAssetContract(
+        PNG_SIGNATURE,
+        "PNG",
+    ),
+    f"{BASE_URL}/logo.png": StaticAssetContract(PNG_SIGNATURE, "PNG"),
+    f"{BASE_URL}/img/funclip-v2-1-0-interface.jpg": StaticAssetContract(
+        JPEG_SIGNATURE,
+        "JPEG",
+        min_bytes=100_000,
+    ),
 }


-def extract_visible_text(html: str) -> str:
-    body = html.split("<body", 1)[-1]
-    body = re.sub(
-        r"<script[\s\S]*?</script>|<style[\s\S]*?</style>",
-        " ",
-        body,
-        flags=re.IGNORECASE,
+_HIDDEN_TAGS = frozenset(("head", "script", "style", "template", "noscript"))
+_VOID_TAGS = frozenset(
+    (
+        "area",
+        "base",
+        "br",
+        "col",
+        "embed",
+        "hr",
+        "img",
+        "input",
+        "link",
+        "meta",
+        "source",
+        "track",
+        "wbr",
    )
-    body = re.sub(r"<[^>]+>", " ", body)
-    return re.sub(r"\s+", " ", body).strip()
+)


-class _LinkCollector(HTMLParser):
+@dataclass
+class _ElementState:
+    tag: str
+    hidden: bool
+    link: str | None = None
+    has_visible_content: bool = False
+
+
+class _VisibleContentCollector(HTMLParser):
    def __init__(self) -> None:
        super().__init__()
+        self.text: list[str] = []
        self.links: set[str] = set()
+        self.images: set[str] = set()
+        self._stack: list[_ElementState] = []

-    def handle_starttag(
-        self, tag: str, attrs: list[tuple[str, str | None]]
-    ) -> None:
-        if tag.lower() != "a":
-            return
-        for name, value in attrs:
-            if name.lower() == "href" and value is not None:
-                self.links.add(value)
+    @property
+    def _parent_hidden(self) -> bool:
+        return bool(self._stack and self._stack[-1].hidden)
+
+    def _mark_link_content(self) -> None:
+        for state in reversed(self._stack):
+            if state.link is not None:
+                state.has_visible_content = True
+                return
+
+    @staticmethod
+    def _has_hidden_attribute(attrs: dict[str, str | None]) -> bool:
+        style = re.sub(r"\s+", "", attrs.get("style") or "").lower()
+        return (
+            "hidden" in attrs
+            or (attrs.get("aria-hidden") or "").lower() == "true"
+            or "display:none" in style
+            or "visibility:hidden" in style
+            or "opacity:0" in style
+        )
+
+    def handle_starttag(self, tag: str, attrs: list[tuple[str, str | None]]) -> None:
+        tag = tag.lower()
+        attributes = {name.lower(): value for name, value in attrs}
+        hidden = (
+            self._parent_hidden
+            or tag in _HIDDEN_TAGS
+            or self._has_hidden_attribute(attributes)
+        )
+        if not hidden:
+            if tag == "img" and attributes.get("src"):
+                self.images.add(attributes["src"])
+                self._mark_link_content()
+        if tag not in _VOID_TAGS:
+            link = attributes.get("href") if tag == "a" and not hidden else None
+            self._stack.append(_ElementState(tag=tag, hidden=hidden, link=link))
+
+    def handle_startendtag(self, tag: str, attrs: list[tuple[str, str | None]]) -> None:
+        tag = tag.lower()
+        attributes = {name.lower(): value for name, value in attrs}
+        hidden = (
+            self._parent_hidden
+            or tag in _HIDDEN_TAGS
+            or self._has_hidden_attribute(attributes)
+        )
+        if not hidden and tag == "img" and attributes.get("src"):
+            self.images.add(attributes["src"])
+            self._mark_link_content()
+
+    def handle_endtag(self, tag: str) -> None:
+        tag = tag.lower()
+        for index in range(len(self._stack) - 1, -1, -1):
+            if self._stack[index].tag == tag:
+                state = self._stack[index]
+                if state.link is not None and state.has_visible_content:
+                    self.links.add(state.link)
+                del self._stack[index:]
+                return
+
+    def handle_data(self, data: str) -> None:
+        if not self._parent_hidden:
+            self.text.append(data)
+            if data.strip():
+                self._mark_link_content()
+
+
+def _collect_visible_content(html: str) -> _VisibleContentCollector:
+    parser = _VisibleContentCollector()
+    parser.feed(html)
+    parser.close()
+    return parser
+
+
+def extract_visible_text(html: str) -> str:
+    parser = _collect_visible_content(html)
+    return re.sub(r"\s+", " ", " ".join(parser.text)).strip()


 def extract_links(html: str) -> set[str]:
-    parser = _LinkCollector()
-    parser.feed(html)
-    return parser.links
+    return _collect_visible_content(html).links
+
+
+def extract_images(html: str) -> set[str]:
+    return _collect_visible_content(html).images


 def validate_pages(pages: dict[str, str]) -> list[str]:
@@ -345,6 +511,11 @@ def validate_pages(pages: dict[str, str]) -> list[str]:
            for target in contract.required_links:
                if target not in links:
                    failures.append(f"{url}: missing link `{target}`")
+        if contract.required_images:
+            images = extract_images(text)
+            for target in contract.required_images:
+                if target not in images:
+                    failures.append(f"{url}: missing image `{target}`")
        for needle in contract.forbidden:
            if needle in text:
                failures.append(f"{url}: forbidden `{needle}`")
@@ -358,7 +529,7 @@ def validate_assets(assets: dict[str, bytes]) -> list[str]:
        if body is None:
            failures.append(f"{url}: asset was not fetched")
        elif not body.startswith(contract.signature) or len(body) < contract.min_bytes:
-            failures.append(f"{url}: response is not a valid PNG")
+            failures.append(f"{url}: response is not a valid {contract.media_type}")
    return failures