Update app.py

app.py

@@ -4,6 +4,8 @@ import gzip
 import io
 import xml.etree.ElementTree as ET
 import gradio as gr
+import re
+import difflib
 
 # URLs
 CHANNELS_JSON_URL = "https://raw.githubusercontent.com/pigzillaaa/daddylive/main/channels.json"
@@ -28,11 +30,40 @@ def fetch_epg(url):
         cid = channel.get("id")
         display_name = channel.findtext("display-name")
         if cid:
-            epg_map[cid.strip().upper()] = cid.strip()
+            epg_map[cid.strip()] = cid.strip()
         if display_name:
-            epg_map[display_name.strip().upper()] = cid.strip() if cid else None
+            epg_map[display_name.strip()] = cid.strip() if cid else None
     return epg_map
 
+# ---------------- Matching Helpers ----------------
+def normalize(s):
+    return re.sub(r'[^a-z0-9]', '', s.lower())
+
+def find_best_match(old_id, ch_name, epg_map):
+    if not old_id and not ch_name:
+        return None
+
+    # Build normalized map
+    epg_norm_map = {normalize(k): v for k, v in epg_map.items() if k}
+
+    old_norm = normalize(old_id)
+    name_norm = normalize(ch_name)
+
+    # Try normalized id
+    if old_norm in epg_norm_map:
+        return epg_norm_map[old_norm]
+
+    # Try normalized name
+    if name_norm in epg_norm_map:
+        return epg_norm_map[name_norm]
+
+    # Fuzzy match on normalized values
+    all_norm_keys = list(epg_norm_map.keys())
+    for candidate in difflib.get_close_matches(name_norm, all_norm_keys, n=1, cutoff=0.85):
+        return epg_norm_map[candidate]
+
+    return None
+
 # ---------------- Main Processing ----------------
 def fix_channels():
     # Load sources
@@ -49,16 +80,10 @@ def fix_channels():
         new_id = None
 
         if group == "UNITED STATES":
-            if old_id.upper() in us_epg:
-                new_id = us_epg[old_id.upper()]
-            elif ch_name.upper() in us_epg:
-                new_id = us_epg[ch_name.upper()]
+            new_id = find_best_match(old_id, ch_name, us_epg)
 
         elif group == "CANADA":
-            if old_id.upper() in ca_epg:
-                new_id = ca_epg[old_id.upper()]
-            elif ch_name.upper() in ca_epg:
-                new_id = ca_epg[ch_name.upper()]
+            new_id = find_best_match(old_id, ch_name, ca_epg)
 
         # Apply update if found
         if new_id:
@@ -76,7 +101,11 @@ def fix_channels():
     with open(updated_file, "w", encoding="utf-8") as f:
         json.dump(updated, f, indent=4, ensure_ascii=False)
 
-    log_text = "\n".join(log_lines)
+    # Sort log: show all ✅ first, ⚠️ unmatched at bottom
+    fixed = [l for l in log_lines if l.startswith("✅")]
+    unmatched = [l for l in log_lines if l.startswith("⚠️")]
+    log_text = "\n".join(fixed + ["", "---- Unmatched ----"] + unmatched if unmatched else fixed)
+
     return updated_file, log_text
 
 # ---------------- Gradio UI ----------------
@@ -85,10 +114,10 @@ demo = gr.Interface(
     inputs=[],
     outputs=[
         gr.File(label="Download Updated channels.json"),
-        gr.Textbox(label="Log (US & Canada only)", lines=20, interactive=False)
+        gr.Textbox(label="Log (US & Canada only)", lines=25, interactive=False)
     ],
     title="Project 2: Channels.json Updater",
-    description="Updates tvg_id for UNITED STATES and CANADA channels using new EPG XMLs. Outputs a single updated file + log."
+    description="Updates tvg_id for UNITED STATES and CANADA channels using new EPG XMLs. Uses fuzzy matching to handle naming differences (e.g., TSN1 ⇔ TSN.1). Outputs a single updated file + log."
 )
 
 if __name__ == "__main__":