Pandas Styler set_sticky help

[edumarcato]

Description

Python: 3.12.6
quarto-cli==1.5.57

Hello, first of all - great working with quarto!

I am hitting some issues to render a panda data frame using the pandas styler / set_sticky .
https://pandas.pydata.org/docs/reference/api/pandas.io.formats.style.Styler.set_sticky.html

Expected: based on what's observed on a jupyter notebook, for a dataframe styled with set_sticky(axis='index') the index column should be frozen/fixed, so when scrolling left/right on a data frame the index can always be seen without overlapping with the other columns on the data frame - with the same code used on Scenario 2 below.

Observed behaviour:
On the qmd I have attached, on the Scenario 0 the index column either overlaps with the last hidden scrolled column. On scenarios 1 and 2, only the index name is frozen while the index values are not.

I have also attached the generated sample.hml.md
sample.html.md

---
title: "Set Sticker Sample"
format: 
  html:
    embed-resources: true
    echo: false
    theme:
      light: minty
      dark: slate
jupyter: python3
keep-md: true
---


```{python}
# | output: true
# | tbl-cap: Scenario 0 - Markdown - Embedding a table with panda styler local
import numpy as np
import pandas as pd
from IPython.display import Markdown, HTML
np.random.seed(0)
df2 = pd.DataFrame(np.random.randn(1,20), columns=[str(i) for i in range(20)]).set_index('0').style.set_sticky(axis='index')

df2_str = df2.to_html()
Markdown(df2_str)
```

```{python}
# | output: true
# | tbl-cap: Scenario 1 - HTML - Embedding a table with panda styler local
HTML(df2_str)
```


```{python}
# | output: true
# | tbl-cap: Scenario 2 - Using display - Embedding a table with panda styler local
display(df2)
```

[mcanouil]

TL;DR:

• This is a code issue not a Quarto CLI or related to Quarto CLI issue
• Use html-table-processing: none to avoid Quarto HTML table processing (it is what's happening for the two last code cells)

---

If you look at the Jupyter Notebook what you think is Markdown from your first case is actually not markdown at all.

Note that you did that:

df2_str = df2.to_html()
Markdown(df2_str)

I don't know what you expect by telling Quarto HTML is in fact Markdown here.

This is unrelated to Quarto.
The last two cells especially. You can see the same results using Jupyter Notebook so without Quarto CLI.

Quarto only uses what's the Python libraries produce.

Quarto has the ability to handle HTML table, you can opt-out of this feature with html-table-processing: none which will make all your cells behave the same. If that output is not what you want, then you need to figure out that on the Python code itself.

The scrolling issue is also coming from your code, not Quarto.
Your code produces the following CSS which is the "issue":

<style type="text/css">
#T_12c76 thead tr th:nth-child(1) {
  position: sticky;
  background-color: inherit;
  left: 0px;
  z-index: 3 !important;
}
#T_12c76 tbody tr th:nth-child(1) {
  position: sticky;
  background-color: inherit;
  left: 0px;
  z-index: 1;
}
</style>