Update app.py

app.py

@@ -1,47 +1,92 @@
 import re
 import requests
 import gradio as gr
+from pathlib import Path
 
-def filter_m3u(url):
-    # Convert GitHub blob links to raw links
-    if "github.com" in url and "blob" in url:
-        url = url.replace("github.com", "raw.githubusercontent.com").replace("/blob/", "/")
-
-    resp = requests.get(url)
-    if resp.status_code != 200:
-        return f"Error fetching file: {resp.status_code}"
-
-    lines = resp.text.splitlines()
-
-    # Regex: keep only MT, remove UK & EST, strip commas
-    pattern = re.compile(
-        r"^(.*\()(?:.*?UK / )?(.*?MT)(?: / .*?EST)?\)$"
-    )
-
-    filtered_lines = []
-    for line in lines:
-        match = pattern.match(line)
-        if match:
-            cleaned = match.group(1) + match.group(2) + ")"
-            cleaned = cleaned.replace(",", "")
-            filtered_lines.append(cleaned)
+def _to_raw_github(url: str) -> str:
+    if "github.com" in url and "/blob/" in url:
+        return url.replace("github.com", "raw.githubusercontent.com").replace("/blob/", "/")
+    return url
+
+def _clean_parenthetical(text: str) -> str:
+    """
+    Inside the first (...) group, keep only the MT entry and remove commas from the date.
+    Examples handled:
+      (Aug 21, 2025 4:55 PM UK / Aug 21, 2025 9:55 AM MT / Aug 21, 2025 11:55 AM EST)
+      (Aug 21, 2025 9:55 AM MT)
+    Leaves non-matching lines untouched.
+    """
+    m = re.search(r"\(([^)]*)\)", text)
+    if not m:
+        return text
+
+    inside = m.group(1)
+    parts = [p.strip() for p in inside.split(" / ")]
+    # Prefer the segment that explicitly ends with MT (avoid matching team names etc.)
+    mt_seg = None
+    for p in parts:
+        if re.search(r"\bMT\b", p):
+            mt_seg = p
+            break
+    if not mt_seg:
+        return text  # no MT segment—don’t touch
+
+    # Remove commas from the date in the MT segment
+    mt_seg_clean = mt_seg.replace(",", "")
+    # Rebuild the string with only the MT segment
+    start, end = m.span()
+    return text[:start] + f"({mt_seg_clean})" + text[end:]
+
+def _filter_text(content: str) -> str:
+    out_lines = []
+    for line in content.splitlines():
+        # Only try to clean lines that actually contain timezones structure
+        if "(" in line and " / " in line and "MT" in line:
+            out_lines.append(_clean_parenthetical(line))
         else:
-            filtered_lines.append(line)
-
-    # Save to temporary file (HuggingFace gives a download link automatically)
-    output_file = "/tmp/DaddyLiveEventsFiltered.m3u8"
-    with open(output_file, "w", encoding="utf-8") as f:
-        f.write("\n".join(filtered_lines))
-
-    return output_file
-
-demo = gr.Interface(
-    fn=filter_m3u,
-    inputs=gr.Textbox(label="M3U URL", placeholder="Paste your .m3u8 link"),
-    outputs=gr.File(label="Download filtered M3U"),
-    title="M3U Cleaner",
-    description="Paste a .m3u8 URL, it will remove UK & EST timestamps, keep only MT, and clean commas."
-)
+            out_lines.append(line)
+    return "\n".join(out_lines)
+
+def filter_from_url(url: str):
+    if not url or not url.strip():
+        return "Please paste a URL."
+    url = _to_raw_github(url.strip())
+    try:
+        r = requests.get(url, timeout=20)
+    except Exception as e:
+        return f"Error fetching file: {e}"
+    if r.status_code != 200:
+        return f"Error fetching file: HTTP {r.status_code}"
+
+    filtered = _filter_text(r.text)
+    out_path = Path("/tmp/DaddyLiveEventsFiltered.m3u8")
+    out_path.write_text(filtered, encoding="utf-8")
+    return str(out_path)
+
+def filter_from_file(file_obj):
+    if file_obj is None:
+        return "Please upload a file."
+    text = Path(file_obj.name).read_text(encoding="utf-8", errors="ignore")
+    filtered = _filter_text(text)
+    out_path = Path("/tmp/DaddyLiveEventsFiltered.m3u8")
+    out_path.write_text(filtered, encoding="utf-8")
+    return str(out_path)
+
+with gr.Blocks(title="M3U Cleaner") as demo:
+    gr.Markdown("### Paste a `.m3u8` URL **or** upload a file. Keeps only **MT** time and removes commas in the date.")
+    with gr.Tab("From URL"):
+        url_in = gr.Textbox(label="M3U URL", placeholder="https://…/file.m3u8")
+        url_btn = gr.Button("Clean")
+        url_out = gr.File(label="Download filtered M3U")
+        url_btn.click(filter_from_url, inputs=url_in, outputs=url_out)
+    with gr.Tab("From File"):
+        file_in = gr.File(label="Upload .m3u or .m3u8")
+        file_btn = gr.Button("Clean")
+        file_out = gr.File(label="Download filtered M3U")
+        file_btn.click(filter_from_file, inputs=file_in, outputs=file_out)
+
+    # Queue is recommended on Spaces
+    demo.queue()
 
 if __name__ == "__main__":
     demo.launch()