`pandas.DataFrame.to_html()` without `table border` and `tr style`

Question

As per title, is it possible -- and if so, how -- to produce "clean" HTML code from pandas.DataFrame.to_html()?

I have found that border=... and justify=... parameters control what is displayed there, but apparently no matter what value you put in there, you seems to always get them.

Here is a minimal working example:

import pandas as pd
import numpy as np

df = pd.DataFrame(data=np.arange(3 * 4).reshape(3, 4))
df.to_html(border=0, justify='inherit')

which produces:

<table border="0" class="dataframe">
  <thead>
    <tr style="text-align: inherit;">
    ...

However, I would have been expecting that:

import pandas as pd
import numpy as np

df = pd.DataFrame(data=np.arange(3 * 4).reshape(3, 4))
df.to_html(classes=None, border=None, justify=None)

would / should produce:

<table class="dataframe">
  <thead>
    <tr>
    ...

instead of:

<table border="1" class="dataframe">
  <thead>
    <tr style="text-align: right;">
    ...

since the introduction both border and style are appearance rather than structure tags and should be included through the css.

So, is there any way to get rid of border from table and style from tr inside the thead?

Answer 1

As you've already observed, df.to_html(classes=None, border=None, justify=None) disregards settings of None, inserting the default values regardless. There are open requests to modify this, but they are not yet in place. As it stands, the only way to remove these hard-coded styles is to manipulate the output string, like so:

html = re.sub(r'<tr.*>', '<tr>', df.to_html().replace('border="1" ', ''))

Removing class="dataframe" can be accomplished the same way, but this shouldn't impact most CSS if left in place.