Bug Report: ProfileReport fails if column has only empty strings

[BraginIvan]

Current Behaviour

    profile = ProfileReport(
        pd.DataFrame({
            "q": ["", None, " "],
            "v": [1, 2, 3]
        }),
        minimal=True,
    ).html

Fails with

    wordcloud = WordCloud(
  File "/wordcloud/wordcloud.py", line 410, in generate_from_frequencies
    raise ValueError("We need at least 1 word to plot a word cloud, "
ValueError: We need at least 1 word to plot a word cloud, got 0.

Expected Behaviour

Should create an html

Data Description

one of df columns should have only empty strings and Nones

Code that reproduces the bug

import pandas as pd 
from ydata_profiling import ProfileReport
profile = ProfileReport(
        pd.DataFrame({
            "q": ["", None, " "],
            "v": [1, 2, 3]
        }),
        minimal=True,
).html

pandas-profiling version

v4.3.1

Dependencies

pandas=1.5.3

OS

MacOs 13.4 (22F66)

Checklist

• There is not yet another bug report for this issue in the issue tracker
• The problem is reproducible from this bug report. This guide can help to craft a minimal bug report.
• The issue has not been resolved by the entries listed under Common Issues.

Hints

    pd.DataFrame({
        "q": ["", None, " "],
        "v": [1, 2, 3]
    })

• q type is Text in the df ^
• text summarizer culculates word_counts but remove whitespaces. So we have empry series
  ydata_profiling.model.pandas.describe_categorical_pandas.word_summary_vc
  words = word_lists.explode().str.strip(string.punctuation + string.whitespace)
• wordcloud cant work with empty series
  wordcloud.wordcloud.WordCloud.generate_from_frequencies
  if len(frequencies) <= 0:
  raise ValueError("We need at least 1 word to plot a word cloud, "
  "got %d." % len(frequencies))
• As I understood in the previous versions q would be categorical not Text

[Ge0f3]

Any update on this or a solution to bypass this?

[alexbarros]

Any update on this or a solution to bypass this?

Not yet, but you can bypass this by setting the variable as categorical using type_schema:

import pandas as pd 
from ydata_profiling import ProfileReport
profile = ProfileReport(
        pd.DataFrame({
            "q": ["", None, " "],
            "v": [1, 2, 3]
        }),
        minimal=True,
        type_schema={"q": "categorical"},
).html