table of styled pandas frame renders differently in pdf

[richard-mevis]

I have:

• searched the issue tracker for similar issues
• installed the latest version of Quarto CLI
• formatted my issue following the Bug Reports guide

Bug description

When rendering a pandas data frame with style as a table in pdf, the output produces a second "Table" heading without a title. Although with same table number. The problem does not arise in a html rendering.

table-problem.pdf

Steps to reproduce

---
format:
    html:
        theme: cosmo
    pdf: default
---

# A Table Problem

I think i found a problem with tables from styled (pandas) data frames 
from python codes, when rendering them in pdf. The following two code 
cells will render as expected to html: two numbered tables with 
captions (see @fig-html).

```{python}
#| label: tbl-results
#| tbl-cap: Results.

import pandas as pd

results = {
    "Daten": ["x_train", "x_val", "x_test", "x_values"],
    "ACC": [1.0, 0.9, 0.8, 0.7],
    "ERR": [1, 2, 3, 4],
}

results = pd.DataFrame(results)
results
```


And now with a little style in the data frame representation.

```{python}
#| label: tbl-results-styled
#| tbl-cap: Results with style.

results.style.hide(axis="index").format({"ACC": "{:.3f}", "ERR": "{:.0f}"})
```

But if one renders to pdf, the second styled table has a second 
"Table 2" heading without a caption. 

::: {#fig-html}
![](table-problem-html.pdf)

HTML Rendering.
:::

[table-problem-html.pdf](https://github.com/user-attachments/files/21813180/table-problem-html.pdf)

Actual behavior

As mentioned before, the pdf output for the styled pandas data frame has a second "Table" heading.

Expected behavior

I would like the same output as in the html: just the one table heading with caption for a styled pandas data frame.

Your environment

• VSCode 1.103.1
• Windows 10
• Quarto 1.6.39

Quarto check output

PS C:\Users\rrascher-friesenhaus\Documents\quarto\quarto-table> quarto check
Quarto 1.6.39
[>] Checking environment information...
Quarto cache location: C:\Users\rrascher-friesenhaus\AppData\Local\quarto
[>] Checking versions of quarto binary dependencies...
Pandoc version 3.4.0: OK
Dart Sass version 1.70.0: OK
Deno version 1.46.3: OK
Typst version 0.11.0: OK
[>] Checking versions of quarto dependencies......OK
[>] Checking Quarto installation......OK
Version: 1.6.39
Path: C:\Users\rrascher-friesenhaus\AppData\Local\Programs\Quarto\bin
CodePage: 1252

[>] Checking tools....................OK
TinyTeX: (not installed)
Chromium: (not installed)

[>] Checking LaTeX....................OK
Using: Installation From Path
Path: C:\texlive\2024\bin\windows
Version: 2024

[>] Checking basic markdown render....OK

[>] Checking Python 3 installation....OK
Version: 3.12.3 (Conda)
Path: C:/Users/rrascher-friesenhaus/AppData/Local/anaconda3/python.exe
Jupyter: 5.7.2
Kernels: python3

[>] Checking Jupyter engine render....OK

[>] Checking R installation...........OK
Version: 4.5.0
Path: C:/PROGRA1/R/R-451.0
LibPaths:
- C:/Users/rrascher-friesenhaus/AppData/Local/R/win-library/4.5
- C:/Program Files/R/R-4.5.0/library
knitr: 1.50
rmarkdown: 2.29

[>] Checking Knitr engine render......OK

[mcanouil]

@richard-mevis Please upgrade to the latest stable version of Quarto. Thank you.

Also, use keep-tex: true and keep-md: true to investigate what you are producing with your code. It seems you are producing a table LaTeX environment and not just the table itself.
Finally, be sure that your code does not emit a table with an identifier.

[mcanouil]

It seems Pandas style function applies an ID to every element thus blocking Quarto to add the "right" one to the parsed table which in the end result in a LaTeX longtable with two captions:

\caption{\label{tbl-results-styled}Results with style.}

\tabularnewline

\caption{}\label{T_2346a}\tabularnewline

It seems Quarto uses the ID as label and does not override it with the ID from the cross-reference.

[richard-mevis]

Thanks a lot for the response. You identified the table id as the possible factor for the second table heading. A workaround is the following: one has to add .set_table_attribute('id=""') to the data frame style. This overrites the pandas given id and then the pdf rendering is ok.
Thanks again for the hint !