From 23d128ff63ad897cf9345b1c147180e47938d5e1 Mon Sep 17 00:00:00 2001
From: Andras Schmelczer <andras@schmelczer.dev>
Date: Sat, 4 Apr 2026 09:49:06 +0100
Subject: [PATCH] Fix transport POIs

---
 pipeline/download/naptan.py | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/pipeline/download/naptan.py b/pipeline/download/naptan.py
index 750344d..a0c3dab 100644
--- a/pipeline/download/naptan.py
+++ b/pipeline/download/naptan.py
@@ -18,6 +18,7 @@ STOP_TYPES = {
     "BCE": "Bus station",
     "TXR": "Taxi rank",
     "TMU": "Metro or Tram stop",
+    "MET": "Metro or Tram stop",
 }
 
 
@@ -44,9 +45,30 @@ def download_naptan(output: Path) -> None:
             pl.col("StopType").replace(STOP_TYPES).alias("category"),
             pl.col("Latitude").alias("lat"),
             pl.col("Longitude").alias("lng"),
+            pl.col("NptgLocalityCode").alias("locality"),
         )
     )
 
+    before = len(df)
+
+    # Deduplicate: one record per name+category+locality
+    # (merges entrances, bus stop pairs on opposite sides of the road, etc.)
+    has_loc = df.filter(
+        pl.col("locality").is_not_null() & (pl.col("locality") != "")
+    )
+    no_loc = df.filter(
+        pl.col("locality").is_null() | (pl.col("locality") == "")
+    )
+    cols = ["id", "name", "category", "lat", "lng"]
+    deduped = has_loc.group_by("name", "category", "locality").agg(
+        pl.col("id").first(),
+        pl.col("lat").mean(),
+        pl.col("lng").mean(),
+    )
+    df = pl.concat([deduped.select(cols), no_loc.select(cols)])
+
+    print(f"Deduplicated {before:,} → {len(df):,} stops (by name+category+locality)")
+
     df.write_parquet(output)
     size_mb = output.stat().st_size / (1024 * 1024)
     print(f"Wrote {output} ({size_mb:.1f} MB, {len(df):,} stations)")