test: guard v1.3.28 website launch (#3410)

Co-authored-by: zhifu gao <[email protected]>

scripts/check_funasr_website_static.py

@@ -4,6 +4,7 @@
 from __future__ import annotations
 
 import argparse
+from html.parser import HTMLParser
 import re
 import sys
 import time
@@ -19,6 +20,7 @@ class PageContract:
     required: tuple[str, ...]
     forbidden: tuple[str, ...] = ()
     visible_required: tuple[str, ...] = ()
+    required_links: tuple[str, ...] = ()
 
 
 @dataclass(frozen=True)
@@ -212,6 +214,52 @@ PAGE_CONTRACTS: dict[str, PageContract] = {
         ),
         forbidden=("funasr==1.3.26", "runtime-llamacpp-v0.1.1"),
     ),
+    f"{BASE_URL}/blog/funasr-v1-3-28-realtime-websocket-subtitles.html": PageContract(
+        required=(
+            "funasr==1.3.28",
+            "funasr-realtime-server",
+            "VAD",
+            "runtime-llamacpp-v0.1.9",
+            "https://github.com/modelscope/FunASR/releases/tag/v1.3.28",
+            "/donors.html",
+        ),
+        visible_required=(
+            "精确合并源码通过 118 项聚焦回归测试",
+            "实时 WebSocket 文件还独立通过 60 项测试",
+            "覆盖 STOP 最终解码",
+            "SenseVoice 用户通过同一次包升级获得字幕对齐修复",
+        ),
+        required_links=(
+            "https://github.com/modelscope/FunASR",
+            "https://github.com/FunAudioLLM/Fun-ASR",
+            "https://github.com/FunAudioLLM/SenseVoice",
+            "https://github.com/modelscope/FunClip",
+        ),
+        forbidden=("funasr==1.3.27", "runtime-llamacpp-v0.1.1"),
+    ),
+    f"{BASE_URL}/en/blog/funasr-v1-3-28-realtime-websocket-subtitles.html": PageContract(
+        required=(
+            "funasr==1.3.28",
+            "funasr-realtime-server",
+            "VAD",
+            "runtime-llamacpp-v0.1.9",
+            "https://github.com/modelscope/FunASR/releases/tag/v1.3.28",
+            "/en/donors.html",
+        ),
+        visible_required=(
+            "The exact merged source passed 118 focused regression tests",
+            "The realtime WebSocket file also passed all 60 tests independently",
+            "including STOP final decode",
+            "SenseVoice users receive the subtitle alignment fix",
+        ),
+        required_links=(
+            "https://github.com/modelscope/FunASR",
+            "https://github.com/FunAudioLLM/Fun-ASR",
+            "https://github.com/FunAudioLLM/SenseVoice",
+            "https://github.com/modelscope/FunClip",
+        ),
+        forbidden=("funasr==1.3.27", "runtime-llamacpp-v0.1.1"),
+    ),
 }
 
 
@@ -235,6 +283,27 @@ def extract_visible_text(html: str) -> str:
     return re.sub(r"\s+", " ", body).strip()
 
 
+class _LinkCollector(HTMLParser):
+    def __init__(self) -> None:
+        super().__init__()
+        self.links: set[str] = set()
+
+    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)
+
+
+def extract_links(html: str) -> set[str]:
+    parser = _LinkCollector()
+    parser.feed(html)
+    return parser.links
+
+
 def validate_pages(pages: dict[str, str]) -> list[str]:
     failures: list[str] = []
     for url, contract in PAGE_CONTRACTS.items():
@@ -250,6 +319,11 @@ def validate_pages(pages: dict[str, str]) -> list[str]:
             for needle in contract.visible_required:
                 if needle not in visible_text:
                     failures.append(f"{url}: visible text missing `{needle}`")
+        if contract.required_links:
+            links = extract_links(text)
+            for target in contract.required_links:
+                if target not in links:
+                    failures.append(f"{url}: missing link `{target}`")
         for needle in contract.forbidden:
             if needle in text:
                 failures.append(f"{url}: forbidden `{needle}`")

tests/test_check_funasr_website_static.py

@@ -147,6 +147,38 @@ def test_website_contract_accepts_current_public_copy():
             https://github.com/modelscope/FunClip
             <a href="/en/donors.html">Thanks</a>
         """,
+        "https://www.funasr.com/blog/funasr-v1-3-28-realtime-websocket-subtitles.html": """
+            funasr==1.3.28
+            funasr-realtime-server
+            VAD STOP SenseVoice
+            精确合并源码通过 118 项聚焦回归测试
+            实时 WebSocket 文件还独立通过 60 项测试
+            覆盖 STOP 最终解码
+            SenseVoice 用户通过同一次包升级获得字幕对齐修复
+            runtime-llamacpp-v0.1.9
+            https://github.com/modelscope/FunASR/releases/tag/v1.3.28
+            <a href="https://github.com/modelscope/FunASR">FunASR</a>
+            <a href="https://github.com/FunAudioLLM/Fun-ASR">Fun-ASR</a>
+            <a href="https://github.com/FunAudioLLM/SenseVoice">SenseVoice</a>
+            <a href="https://github.com/modelscope/FunClip">FunClip</a>
+            <a href="/donors.html">功德榜</a>
+        """,
+        "https://www.funasr.com/en/blog/funasr-v1-3-28-realtime-websocket-subtitles.html": """
+            funasr==1.3.28
+            funasr-realtime-server
+            VAD STOP SenseVoice
+            The exact merged source passed 118 focused regression tests.
+            The realtime WebSocket file also passed all 60 tests independently
+            including STOP final decode
+            SenseVoice users receive the subtitle alignment fix
+            runtime-llamacpp-v0.1.9
+            https://github.com/modelscope/FunASR/releases/tag/v1.3.28
+            <a href="https://github.com/modelscope/FunASR">FunASR</a>
+            <a href="https://github.com/FunAudioLLM/Fun-ASR">Fun-ASR</a>
+            <a href="https://github.com/FunAudioLLM/SenseVoice">SenseVoice</a>
+            <a href="https://github.com/modelscope/FunClip">FunClip</a>
+            <a href="/en/donors.html">Thanks</a>
+        """,
     }
 
     assert checker.validate_pages(pages) == []
@@ -187,6 +219,117 @@ def test_website_contract_includes_v1327_launch_articles():
     )
 
 
+def test_website_contract_includes_v1328_launch_articles():
+    checker = _load_module()
+
+    assert (
+        "https://www.funasr.com/blog/funasr-v1-3-28-realtime-websocket-subtitles.html"
+        in checker.PAGE_CONTRACTS
+    )
+    assert (
+        "https://www.funasr.com/en/blog/funasr-v1-3-28-realtime-websocket-subtitles.html"
+        in checker.PAGE_CONTRACTS
+    )
+
+
+def test_v1328_contract_rejects_language_swapped_body():
+    checker = _load_module()
+    url = "https://www.funasr.com/blog/funasr-v1-3-28-realtime-websocket-subtitles.html"
+    pages = {
+        url: """
+            <body>
+            funasr==1.3.28 funasr-realtime-server VAD STOP SenseVoice
+            The exact merged source passed 118 focused regression tests.
+            The realtime WebSocket file also passed all 60 tests independently.
+            runtime-llamacpp-v0.1.9
+            https://github.com/modelscope/FunASR/releases/tag/v1.3.28
+            <a href="https://github.com/modelscope/FunASR">FunASR</a>
+            <a href="https://github.com/FunAudioLLM/Fun-ASR">Fun-ASR</a>
+            <a href="https://github.com/FunAudioLLM/SenseVoice">SenseVoice</a>
+            <a href="https://github.com/modelscope/FunClip">FunClip</a>
+            <a href="/donors.html">Thanks</a>
+            </body>
+        """
+    }
+
+    failures = checker.validate_pages(pages)
+
+    assert any(
+        url in failure
+        and "visible text missing `精确合并源码通过 118 项聚焦回归测试`" in failure
+        for failure in failures
+    )
+
+
+def test_v1328_contract_rejects_css_test_count_and_indirect_repo_link():
+    checker = _load_module()
+    url = "https://www.funasr.com/en/blog/funasr-v1-3-28-realtime-websocket-subtitles.html"
+    pages = {
+        url: """
+            <style>.proof { font-weight: 600; }</style>
+            <body>
+            funasr==1.3.28 funasr-realtime-server VAD STOP SenseVoice
+            The exact merged source passed 118 focused regression tests.
+            runtime-llamacpp-v0.1.9
+            <a href="https://github.com/modelscope/FunASR/releases/tag/v1.3.28">Release</a>
+            <a href="https://github.com/FunAudioLLM/Fun-ASR">Fun-ASR</a>
+            <a href="https://github.com/FunAudioLLM/SenseVoice">SenseVoice</a>
+            <a href="https://github.com/modelscope/FunClip">FunClip</a>
+            <a href="/en/donors.html">Thanks</a>
+            </body>
+        """
+    }
+
+    failures = checker.validate_pages(pages)
+
+    assert any(
+        url in failure
+        and "visible text missing `The realtime WebSocket file also passed all 60 tests independently`"
+        in failure
+        for failure in failures
+    )
+    assert any(
+        url in failure
+        and "missing link `https://github.com/modelscope/FunASR`" in failure
+        for failure in failures
+    )
+
+
+def test_v1328_contract_requires_visible_stop_and_sensevoice_copy():
+    checker = _load_module()
+    url = "https://www.funasr.com/en/blog/funasr-v1-3-28-realtime-websocket-subtitles.html"
+    pages = {
+        url: """
+            <head><meta name="keywords" content="STOP"></head>
+            <body>
+            funasr==1.3.28 funasr-realtime-server VAD
+            The exact merged source passed 118 focused regression tests.
+            The realtime WebSocket file also passed all 60 tests independently.
+            runtime-llamacpp-v0.1.9
+            https://github.com/modelscope/FunASR/releases/tag/v1.3.28
+            <a href="https://github.com/modelscope/FunASR">FunASR</a>
+            <a href="https://github.com/FunAudioLLM/Fun-ASR">Fun-ASR</a>
+            <a href="https://github.com/FunAudioLLM/SenseVoice">Repository</a>
+            <a href="https://github.com/modelscope/FunClip">FunClip</a>
+            <a href="/en/donors.html">Thanks</a>
+            </body>
+        """
+    }
+
+    failures = checker.validate_pages(pages)
+
+    assert any(
+        url in failure and "visible text missing `including STOP final decode`" in failure
+        for failure in failures
+    )
+    assert any(
+        url in failure
+        and "visible text missing `SenseVoice users receive the subtitle alignment fix`"
+        in failure
+        for failure in failures
+    )
+
+
 def test_static_asset_contract_rejects_non_png_payloads():
     checker = _load_module()
     banner_url = "https://www.funasr.com/img/banner.4f436d19.png"