Selecting the wrong encoding (code page) may display some characters correctly but others will be scrambled.

The first 256 characters in a mixed selection of encodings are displayed below.

Characters may display as a box denoting binary data, another character or even several other characters.

If you have a CSV file that is encoded with UTF-8 and contains entries in multibyte characters from East Asian languages such as Chinese, Japanese, Koreans, opening the file in Excel on MacOS may give you some surprise.

Instead of showing the East Asian characters, Excel will display garbled characters.

We will use Node JS to create a CSV file, encode the file using UTF-8 and then try to open it using Excel on MacOS.

There are two rows.

The first row only contains single-byte latin characters.

The second row contains mix of latin and multi-byte east asian characters that are represented by Korean phrases.

Each row is appended with newline character n so that the second row goes to the next line.

Instead of spreading the row fields into several columns, the row entry is shown in a single column.

We also see that a field in a row is separated to the next field by a tab.

It is important to note that Node JS will not add BOM when writing the file even though the encoding is specified in the option.

Also, the tab character has no effect in separating the fields in a row.

However, this does not hold true when the file is opened in Excel.

Nonetheless, the absence of BOM fails proper data display on Excel.

When such file is programmatically created by using Node JS, the BOM character must be manually prepended since Node JS does not automatically add BOM to the data.

Node JS Issue 11767: Add option to automatically add BOM to write methods.

Utf-8 Encoding For Excel Mac