Errata

My point is it's NOT an error but rather I think the code as it is paints an unfair picture of VBA's error handling.

For the example, error handling is not even required.

Instead, the code could be written as follows:

Sub ModifiedRecip(num As Variant)

Select Case IsNumeric(num)

Case True

Select Case num

Case 0

Debug.Print "Value is 0"

Case Else

Debug.Print 1 / num

End Select

Case False

Debug.Print "Value is not numeric"

End Select

End Sub

Note from the Author or Editor:
Thanks for your feedback and you're definitely right, if the code sample would be about how to write an algorithm to return the reciprocal, there would be better ways to solve it. However, the sample was explicitly constructed to present the error handling capabilities of VBA and calculating the reciprocal was just an excuse to have something easy to follow along with. I will add a note about this to clarify.

When I type as default shell, the entry "Terminal: Select Default Shell" is not found. Request to check and update. I think this is not available in latest version of VS code.

Note from the Author or Editor:
[p32 rather than p57] Correct, VS Code changed the Command Palette entry from Terminal: Select Default Shell to Terminal: Select Default Profile, accordingly, when you search for default shell, you won’t find the entry. Instead, type default profile.

I had this pointed out on the book's homepage (xlwings.org/book) but forgot to open an official errata here.

Just for clarity it should be mentioned that the comment in

```vba
If condition Then
' Do nothing
End If
```

is *optional*. Thus, this is equivalent to

```vba
If condition Then
End If
```

Note from the Author or Editor:
I agree, but would rather mention again explicitly that "Do Nothing" is a comment in both versions.

df.groupby(["continent"]).agg(lambda x: x.max() - x.min())

returns result properly, but Jupyter Notebook shows this warning:

/opt/anaconda3/lib/python3.9/site-packages/pandas/core/groupby/generic.py:303: FutureWarning: Dropping invalid columns in SeriesGroupBy.agg is deprecated. In a future version, a TypeError will be raised. Before calling .agg, select only columns which should be valid for the aggregating function.
results[key] = self.aggregate(func)

so I think a more future-proof version of this command is required.

Note from the Author or Editor:
Correct, this is caused by a newer version of pandas and the code sample should be changed to:

df.loc[:, ["age", "score", "continent"]].groupby(["continent"]).agg(lambda x: x.max() - x.min())

The second line of output for `ms.info()` displays "RangeIndex: 8622 entries, 0 to 8621", while In [10] displays "DatetimeIndex: 8622 entries, 1986-03-13 to 2020-05-27". I initially only saw the difference in columns and missed the importance in the difference of indexes until it was mentioned on page 128 1st paragraph. Explicitly stating line 2 of the output for `DataFrame.info()` and calling attention to the start and end values of the index should help readers make this connection: "As ++the second line of output for `info` reveals, what was an index of 0 to 8621 is now 1986-03-13 to 2020-05-27++, you are now ..."

Note from the Author or Editor:
Agreed that this makes it clearer.

Problem on page 146
Hello,
I’m getting an error on page 146 when I try running the script sales_report_pandas.py in VS Code.
The error message (traceback) to the terminal is quite long, it ends with “Missing optional dependency ‘xlrd’ ”.
I changed the filter “rglob("*.xls*")” to try to isolate the problem:
• “rglob("*.xlsx")” ran successfully;
• “rglob("*.xls")” failed, with the same message about xlrd as above.
It looks like the problem is with the .xls files only, not the .xlsx files.
Could you give me any advice on solving this? Up to this page, all the examples have worked perfectly for me.
My version of Python is 3.11.4
My version of pandas is 1.5.3
Thanks for any help you can give me.

Note from the Author or Editor:
Thanks for your message! It looks like the most recent versions of Anaconda don't have xlrd installed by default anymore. You can manually install it by running "pip install xlrd" on the Anaconda Prompt.

The first paragraph in page 171 reads like this:

"Since OpenPyXL doesn’t support the with statement, you will need to make sure to close the file again when you’re done."

If this is true, you should *always* close file manually.

But the second comment in following example states:

"# Required with read_only=True"

This sounds like if you don't use read_only option, OpenPyXL *automatically* closes the file. Confusing, I need your clarification.

Note from the Author or Editor:
You're right. It isn't pointed out clear enough that you only need to manually close the file with "read_only=True". For your reference, without "read_only=True", it is closed automatically as per comment on p.159.

In the printed book and the source notebook, it says:

# This makes just two cross-application calls
values = sheet2["A1:E30"].options(np.array).value
sheet2["A1"].value = values + 1

This works just fine. But I was wondering if that last line should be:

sheet2["A1:E30"].value = values + 1

to show that you can update all the values in the range with a single cross-application call.

Thanks,
Sean

Note from the Author or Editor:
Hi Sean. I agree, this will make the example easier to read. On top of that, I think it's possibly confusing that I am writing out the values first before manipulating them. I should be able to simplify the sample by turning it into "write by looping" vs. "write as array".