Split to rows
Split to rows — explodeRows
Expand a column into multiple rows — one per value. The other columns are repeated on each new row.
Use cases
- Turn a
tagscell like"news;sports;tech"into one row per tag. - Unnest a list column produced upstream.
- Normalize multi-value fields before grouping or joining.
What it does
With a delimiter, the text is split on it first and then exploded
("a;b" → two rows). Without a delimiter, an existing list column is exploded
directly. Row count grows; all other columns are duplicated per value.
Configuration
| Config key | Type | Required | Description |
|---|---|---|---|
column | string | Yes | The column to expand |
delimiter | string | No | Split text on this delimiter first; empty = explode a list column |
Generated Python code
df_2 = df_1.assign(tags=lambda _d: _d['tags'].astype('string').str.split(';')).explode('tags').reset_index(drop=True)
Tips & common mistakes
- A cell with no delimiter present yields a single row (count unchanged for it).
- A trailing delimiter (
"a;") produces an empty-string row — clean up with Filter by expression if needed.