build_code_export does not escape curly brackets that are used as part of syntax code

[lironle6]

dtale version: 2.4.0

Hi,
Just started using this wonderful package a few days ago and ran into this issue.
replication:

df = pd.read_csv('https://raw.githubusercontent.com/datasciencedojo/datasets/master/titanic.csv')
dtale.show(df)

then on the website use replacements and
search for: male replace with 0 and click +
search for: female replace with 1 and click +
click replace
now click on "correlations" and it shows:

Traceback (most recent call last):
  File "/home/sta/miniconda3/envs/JLab/lib/python3.9/site-packages/dtale/views.py", line 120, in _handle_exceptions
    return func(*args, **kwargs)
  File "/home/sta/miniconda3/envs/JLab/lib/python3.9/site-packages/dtale/views.py", line 2815, in get_correlations
    matrix_data = build_correlations_matrix(
  File "/home/sta/miniconda3/envs/JLab/lib/python3.9/site-packages/dtale/views.py", line 2749, in build_correlations_matrix
    matrix_code.format(
KeyError: "\n\t'male'"

This happens because the replace code has curly brackets and this wasn't escaped so the next format breaks.
This is the matrix_code before the format breaks:

s = df['Sex']
s = s.replace({
	'male': '0',
	'female': '1'})
df.loc[:, 'Sex'] = s
s = df['Sex']
s = s.replace({
	'male': '0',
	'female': '1'})
df.loc[:, 'Sex'] = s
corr_cols = [
	'{corr_cols}'
]
corr_data = df[corr_cols]
{str_encodings}corr_data = corr_data.corr(method='pearson')

The second line has curly brackets so python tries to format it but doesn't find the relevant key and errors out.
I would suggest either escaping prior code curly brackets, or better yet format the code before joining.

Thanks and keep up the good work!

[aschonfeld]

@shnitzel6 thanks for bringing this to my attention. I just committed a fix and will let you know when a new version is available

[lironle6]

Thanks @aschonfeld
Looking forward to it

[aschonfeld]

@shnitzel6 just released v2.5.1 with this fix. Also, if you haven't already, please toss your ⭐ on the repo. Thanks 🙏