Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

adding a JavaScript/Jinja example to dropdowns.md #4886

Open
wants to merge 10 commits into
base: doc-prod
Choose a base branch
from
96 changes: 96 additions & 0 deletions doc/python/dropdowns.md
Original file line number Diff line number Diff line change
Expand Up @@ -444,5 +444,101 @@ fig.update_layout(title_text="Yahoo")
fig.show()
```

### Creating several independent graphs and using Jinja to insert them into a JavaScript enabled webpage
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
### Creating several independent graphs and using Jinja to insert them into a JavaScript enabled webpage
### Embedding Multiple Graphs in a Webpage using Jinja

We can shorten this if possible, so it takes up less space on the left sidebar

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good suggestion. What would you think of "Using a Dropdown to Select a Graph Using Jinja" as a short title?


It is straight forward to create each potential view as a separate graph and then use Jinja to insert each potential view into a div on a JavaScript enabled webpage with a drop down that chooses which div to display. This approach produces code that requires little customization or updating as you e.g. add, drop, or reorder views or traces, so it is particularly compelling for prototyping and rapid iteration. It produces web pages that are larger than the webpages produced through the built in method which is a consideration for very large figures with hundreds or thousands of data points in traces that appear in multiple selections. This approach requires both a Python program and a Jinja template file. The documentation on [using Jinja templates with Plotly](https://plotly.com/python/interactive-html-export/#inserting-plotly-output-into-html-using-a-jinja2-template) is relevant background.

rl-utility-man marked this conversation as resolved.
Show resolved Hide resolved
#### Python Code File

```
import plotly.express as px
from jinja2 import Template
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think this needs to be marked as a python block.
Right now it's rendering as text:
image

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

agreed! Thanks!

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This probably actually needs the same <!-- #region --> blocks as well as being marked as Python, as in:

https://github.com/plotly/plotly.py/blob/doc-prod/doc/python/interactive-html-export.md?plain=1#L59-L102

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Added in commit 5; there's something funky with the preview screen -- neither the example -- which demonstrably renders correctly when live -- nor my new commit convert e.g. > to > in preview mode. I hope I've done everything right, but it's difficult to tell.

import collections
# Load the gapminder dataset
df = px.data.gapminder()

#create a dictionary with Plotly figures as values
rl-utility-man marked this conversation as resolved.
Show resolved Hide resolved
fig_dict = {}

# we need to fill that dictionary with figures. this example assumes that each figure has a title and that
# we want to use the titles as descriptions in the drop down
# This example happens to fill the dictionary by creating a scatter plot for each continent using the 2007 Gapminder data
for continent in df['continent'].unique():
# Filter data for the current continent
continent_data = df[(df['continent'] == continent) & (df['year'] == 2007)]

fig_dict[continent] = px.scatter(continent_data, x='gdpPercap', y='lifeExp',
title=f'GDP vs Life Expectancy for {continent}',
labels={'gdpPercap': 'GDP per Capita (USD)', 'lifeExp': 'Life Expectancy (Years)'},
hover_name='country',size="pop", size_max=55
)
#Standardizing the axes makes the graphs easier to compare
fig_dict[continent].update_xaxes(range=[0,50000])
fig_dict[continent].update_yaxes(range=[25,90])


# Create a dictionary, data_for_jinja with two entries:
# the value for the "dropdown_entries" key is a string containing a series of <option> tags, one tag for each item in the drop down
# the value for the "divs" key is a string with a series of <div> tags, each containing the content that appears only when the user selects the corresponding item from the dropdown
# in this example, the content of each div is a figure and descriptive text.
data_for_jinja= collections.defaultdict(str)
text_dict = {}
for n, figname in enumerate(fig_dict.keys()):
text_dict[figname]=f"Here is some custom text about the {figname} figure" #This is a succinct way to populate text_dict; in practice you'd probably populate it manually elsewhere
data_for_jinja["dropdown_entries"]+=f"<option value='{figname}'>{fig_dict[figname].layout.title.text}</option>"
#YOU MAY NEED TO UPDATE THE LINK TO THE LATEST PLOTLY.JS
fig_html = fig_dict[figname].to_html(full_html=False, config=dict(responsive=False, scrollZoom=False, doubleClick=False), include_plotlyjs = "https://cdn.plot.ly/plotly-2.35.2.min.js")
initially_hide_divs_other_than_the_first = "style=""display:none;"""*(n>0)
data_for_jinja["divs"]+=f'<div id="{figname}" class="content-div" {initially_hide_divs_other_than_the_first}>{fig_html}{text_dict[figname]}</div>'

# Insert data into the template and write the file to disk
# YOU WILL LIKELY NEED TO CUSTOMIZE THESE PATHS
input_template_path=r"C:\data\demo_template.jinja"
output_html_path=r"C:\data\demo_result.html"
rl-utility-man marked this conversation as resolved.
Show resolved Hide resolved

with open(output_html_path, "w", encoding='utf-8') as output_file:
with open(input_template_path) as template_file:
j2_template = Template(template_file.read())
output_file.write(j2_template.render(data_for_jinja))
```

#### Jinja HTML Template

```
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">

</head>
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm not certain why, but this renders like this in the generated page. We probably want to mark it as txt. Do you edit the files in a Jupyter notebook or the md file directly

image

Copy link
Contributor Author

@rl-utility-man rl-utility-man Dec 2, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for all these good edits. I edited the .md directly. There's an example of embedding HTML into the .md here: https://plotly.com/python/interactive-html-export/

The gist of it is to use e.g. lt; rather than <:

<!-- #region -->

[three backticks]
&lt;!DOCTYPE html&gt;
&lt;html&gt;
&lt;head&gt;

I can make the requisite edits in the next day or two

Copy link
Contributor Author

@rl-utility-man rl-utility-man Dec 3, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I just committed changes that resolve this problem. The preview now looks good to me. Thank you for flagging this!

<body>
<div class="container">
<h1>Select an analysis</h1>
<select id="dropdown" class="form-control">
{{ dropdown_entries }}
</select>


{{ divs }}

</div>

<script>
document.getElementById('dropdown').addEventListener('change', function() {
const divs = document.querySelectorAll('.content-div');
divs.forEach(div => div.style.display = 'none');

const selectedDiv = document.getElementById(this.value);
if (selectedDiv) {
selectedDiv.style.display = 'block';
}
});
</script>
</body>
</html>
```



#### Reference
See https://plotly.com/python/reference/layout/updatemenus/ for more information about `updatemenu` dropdowns.