andras/perfect-postcode
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages 2 Wiki Activity Actions

Add POI groups

Browse source
This commit is contained in:
Andras Schmelczer 2026-02-01 11:09:01 +00:00
parent a45495269e
commit 8e615c6aad
1 changed files with 7 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

7
pipeline/transform/transform_poi.py
View file

@ -624,13 +624,20 @@ def transform(input_path: Path, naptan_path: Path | None = None) -> pl.LazyFrame
if missing_emojis:
raise ValueError(f"Empty emojis for: {missing_emojis}")
# Derive group from the first component of the raw category key, title-cased
group_mapping = {
k: k.split("/")[0].replace("_", " ").title() for k in CATEGORY_MAP
}
lf = lf.with_columns(
pl.col("category").replace_strict(group_mapping).alias("group"),
pl.col("category").replace_strict(name_mapping).alias("category"),
pl.col("category").replace_strict(emoji_mapping).alias("emoji"),
)
naptan = pl.scan_parquet(naptan_path).with_columns(
pl.col("category").replace_strict(NAPTAN_EMOJIS).alias("emoji"),
pl.lit("Public Transport").alias("group"),
)
return pl.concat([lf, naptan], how="diagonal_relaxed")