---
jupyter:
  jupytext:
    text_representation:
      extension: .Rmd
      format_name: rmarkdown
      format_version: '1.2'
      jupytext_version: 1.4.1
  kernelspec:
    display_name: Python 3
    language: python
    name: python3
---

```{python nbsphinx=hidden}
import pandas as pd
pd.set_option("display.max_rows", 5)
```

## Joins

```{python}
from siuba import _, inner_join, left_join, full_join

df1 = pd.DataFrame({'id': [1,2], 'x': ['a', 'b']})
df2 = pd.DataFrame({'id': [2,2,3], 'y': ['l', 'm', 'n']})
```

```{python}
df1
```

```{python}
df2
```

**⚠️Note on piping:** Currently, when you use a join in a pipe, you need to pass `_` as the first argument. This is because it requires two DataFrames. For single DataFrame verbs it is optional.

```{python}
df1 >> inner_join(_, df2, on = "id")
```

### Inner join

```{python}
inner_join(df1, df2, on = "id")
```

### Left join 

```{python}
left_join(df1, df2, on = "id")
```

### Full join

```{python}

full_join(df1, df2, on = "id")
```

### Semi and anti join

```{python}
from siuba import semi_join, anti_join

semi_join(df1, df2, on = "id")
```

```{python}
# TODO: implement
#anti_join(df1, df2, on = "id")
```