- added the `create_histogram` method and the `Histogram` object. - added the `round` parameter to `trend_line` and `ray_line` - chart.set can now be given line data. Bug Fixes: - `NaN` values can now be given when setting data, and will leave a blank space in the data. - `resize` will now change the chart wrapper’s size as well as the chart itself.
1.4 KiB
1.4 KiB
Histogram
The `Histogram` object represents a `HistogramSeries` object in Lightweight Charts and can be used to create indicators. As well as the methods described below, the `Line` object also has access to:
[`horizontal_line`](#AbstractChart.horizontal_line), [`hide_data`](#hide_data), [`show_data`](#show_data) and [`price_line`](#price_line).
Its instance should only be accessed from [`create_histogram`](#AbstractChart.create_histogram).
___
```{py:method} set(data: pd.DataFrame)
Sets the data for the histogram.
When a name has not been set upon declaration, the columns should be named: `time | value` (Not case sensitive).
The column containing the data should be named after the string given in the `name`.
A `color` column can be used within the dataframe to specify the color of individual bars.
```
___
```{py:method} update(series: pd.Series)
Updates the data for the histogram.
This should be given as a Series object, with labels akin to the `histogram.set` method.
```
___
```{py:method} scale(scale_margin_top: float, scale_margin_bottom: float)
Scales the margins of the histogram, as used within [`volume_config`](#AbstractChart.volume_config).
```
___
```{py:method} delete()
Irreversibly deletes the histogram.
```