add support for sdvx nabla news site

4 files changed, 83 insertions, 1 deletions

diff --git a/bemani/sdvx.py b/bemani/sdvx.py
index 9d5a33b..5a7d25c 100644
--- a/bemani/sdvx.py
+++ b/bemani/sdvx.py
@@ -49,3 +49,79 @@ def parse_exceed_gear_news_site(html: str):
         })
 
     return entries
+
+def parse_nabla_news_site(html: str):
+    base_url = "https://p.eagate.573.jp"
+    soup = BeautifulSoup(html, 'html.parser')
+    news_list = soup.select('#news-inner ul.news li')
+
+    entries = []
+    for li in news_list:
+        strong_tags = li.select('strong')
+        if not strong_tags:
+            continue
+
+        date = strong_tags[0]
+        date_str = date.text.strip()
+        try:
+            dt = datetime.strptime(date_str, "%Y.%m.%d")
+            timestamp = int(dt.timestamp())
+        except ValueError:
+            timestamp = None
+
+        headline_text = None
+        if len(strong_tags) > 1:
+            headline_text = strong_tags[1].text.strip()
+
+        for tag in li.select('font, b, u, span'):
+            tag.unwrap()
+
+        content_parts = []
+        for node in li.contents:
+            if hasattr(node, 'name'):
+                if node.name == 'strong':
+                    continue
+                elif node.name == 'br':
+                    content_parts.append('\n')
+                elif node.name == 'a' and 'link-text' in node.get('class', []):
+                    content_parts.append(node.text.strip())
+                elif node.name not in ['img']:  # Skip image tags for content
+                    content_parts.append(node.get_text(strip=True))
+            else:
+                text = str(node).strip()
+                if text and text not in [date_str, headline_text]:
+                    content_parts.append(text)
+
+        content = '\n'.join(filter(None, content_parts)).strip()
+
+        images = []
+        for img in li.select('img'):
+            src = img.get('data-original') or img.get('src')
+            if not src or (isinstance(src, str) and src.startswith('data:')):
+                continue
+            if isinstance(src, str):
+                src = urljoin(base_url, src)
+                parent = img.find_parent('a')
+                href = None
+                if parent and hasattr(parent, 'get') and parent.get('href'):
+                    href_val = parent.get('href')
+                    if isinstance(href_val, str):
+                        href = urljoin(base_url, href_val)
+
+                image_entry = {'image': src, 'link': href}
+                if image_entry not in images:
+                    images.append(image_entry)
+
+        entries.append({
+            'date': date_str,
+            'identifier': 'SOUND_VOLTEX',
+            'type': None,
+            'timestamp': timestamp,
+            'headline': headline_text,
+            'content': content,
+            "url": None,
+            'images': images,
+            'is_ai_summary': False
+        })
+
+    return entries
diff --git a/constants.py b/constants.py
index 6125147..22567a9 100644
--- a/constants.py
+++ b/constants.py
@@ -3,6 +3,7 @@ from enum import Enum
 DAYS_LIMIT=14
 
 SOUND_VOLTEX_EXCEED_GEAR_NEWS_SITE ="https://p.eagate.573.jp/game/sdvx/vi/news/index.html"
+SOUND_VOLTEX_NABLA_NEWS_SITE="https://p.eagate.573.jp/game/sdvx/vii/news/index.html"
 IIDX_PINKY_CRUSH_NEWS_SITE="https://p.eagate.573.jp/game/2dx/32/info/index.html"  # legacy should not be used,  eamuse feed is more verbose
 DDR_WORLD_NEWS_SITE="https://p.eagate.573.jp/game/ddr/ddrworld/info/index.html"
 POLARIS_CHORD_NEWS_SITE="https://p.eagate.573.jp/game/polarischord/pc/news/news.html"
diff --git a/generate.py b/generate.py
index ab7d968..555b2d6 100644
--- a/generate.py
+++ b/generate.py
@@ -165,7 +165,7 @@ def generate_iidx_news_file(eamuse_feed: bool=False):
     return news
 
 def generate_sdvx_news_file():
-    news = generate_news_file("sdvx_news", constants.SOUND_VOLTEX_EXCEED_GEAR_NEWS_SITE)
+    news = generate_news_file("sdvx_news", constants.SOUND_VOLTEX_NABLA_NEWS_SITE)
     attempt_broadcast_notifications(news, "New Information for SOUND VOLTEX","sdvx")
     return news
 
diff --git a/news_feed.py b/news_feed.py
index c962e82..805aecf 100644
--- a/news_feed.py
+++ b/news_feed.py
@@ -66,6 +66,11 @@ def get_news(news_url: str, version=None) -> list:
         news_posts = sorted(sound_voltex.parse_exceed_gear_news_site(site_data), key=lambda x: x['timestamp'], reverse=True)
         news_posts = translate.add_translate_text_to_en(news_posts, overrides=[("ボルテ", "SDVX")])
 
+    elif news_url == constants.SOUND_VOLTEX_NABLA_NEWS_SITE:
+        site_data = download_site_as_html(news_url)
+        news_posts = sorted(sound_voltex.parse_nabla_news_site(site_data), key=lambda x: x['timestamp'], reverse=True)
+        news_posts = translate.add_translate_text_to_en(news_posts, overrides=[("ボルテ", "SDVX")])
+
     elif news_url == constants.IIDX_PINKY_CRUSH_NEWS_SITE:
         site_data = download_site_as_html(news_url)
         news_posts = sorted(iidx.parse_pinky_crush_news_site(site_data), key=lambda x: x['timestamp'], reverse=True)