Two issues in the sub-package store:

• In the module ldr, the function load_pickle() may fail to work after upgrading Pandas.
• In the module svr, the functions save_spreadsheet() and save_spreadsheets() may not work properly in some occasions.

When using the MSSQL class on Windows OS, the database connection URL and address may incorporate encoded characters like %5C instead of the expected backslash \. For example:

>>> from pyhelpers.dbms import MSSQL
>>> mssql = MSSQL()
>>> mssql.address  # Expected: computer_name\username@localhost:1433/master
'computer_name%5Cusername@localhost:1433/master'

Related function:

• save_spreadsheet()

When saving data to a .csv file using the save_spreadsheet() function with the index=False (default) parameter (which is supposed to exclude the index column), the index of the data is unexpectedly still saved in the resulting file.

To enhance the functionality of the save_fig() function in the svr module, it is recommended to include a parameter that specifies the figure object, like so: save_fig(data, path_to_file, ...).

Related functions:

• load_pickle()
• load_data()

When attempting to load pickle files that were saved previously, both load_pickle() and load_data() functions encounter an error and fail to read the files. The error message displayed is as follows:

TypeError: Argument 'placement' has incorrect type (expected pandas._libs.internals.BlockPlacement, got numpy.ndarray)

Related functions:

• wgs84_to_osgb36()
• osgb36_to_wgs84()

When passing an array as an argument to these functions, a warning message appears:

Expected type 'int | float | Iterable[int, float]', got 'ndarray[Any, dtype]' instead".

When saving data to a spreadsheet in an Excel file using save_spreadsheet() and save_spreadsheets(), we would expect the column widths to be automatically fitted.

Related functions:

• _check_loading_path()
• _check_saving_path()

It is desired to allow users to customise the printing messages for tasks related to checking loading and saving paths. This could involve providing more optional arguments/parameters, which can be helpful for making the process more adaptable to users' needs.

Related functions:

• _check_rel_pathname()
• _get_rel_pathname()

The two functions within the _cache module are duplicated. One of them may be removed or refactored to avoid confusion and mixed usage.