We are using EPPlus to read an Excel .xlsx file created by Google Sheets. The original Google Sheets document contains both "Comments' and 'Notes' in the cells of the header row. When EPPlus reads the .xlsx file exported by Google Sheets, it fails to deserialize the comments present in the .xlsx file. This appear to be occurring because of the form in which Google Sheets is serializing the comments. Note that, Excel is able to read and deserialize the .xlsx file create by Google Sheet and to display the comments correctly.
Google Sheets serialized the comments to the file 'comments1.Google.xml', which is attached. Each Google Sheets comment is serialized as a simple text value.
I created an Excel file which is essentially identical to the Google Sheets file. It contains the same column names and comments. Excel serialized these comments to the file 'comments.Excel.xml', which is also attached. Each Excel comment is serialized in what appears to be rich text form.
There may also be a problem in the RangeCollection method 'IndexOf'. I was able to circumvent the problem described above and successfully deserialize the comments from the .xlsx file created by Google Sheets. However, the 'IndexOf' method was unable to reliably return the comments because the '_cellIndex' list entries were not in order. The 'IndexOf' method assumes that the '_cellIndex' list entries are sorted. In my case, the entries were out of order, so the 'Array.BinarySearch' method would not reliable look up entries. It is possible that the entries are normally ordered correctly, and my changes to the deserialization logic disrupted that order.
Google Sheets serialized the comments to the file 'comments1.Google.xml', which is attached. Each Google Sheets comment is serialized as a simple text value.
I created an Excel file which is essentially identical to the Google Sheets file. It contains the same column names and comments. Excel serialized these comments to the file 'comments.Excel.xml', which is also attached. Each Excel comment is serialized in what appears to be rich text form.
There may also be a problem in the RangeCollection method 'IndexOf'. I was able to circumvent the problem described above and successfully deserialize the comments from the .xlsx file created by Google Sheets. However, the 'IndexOf' method was unable to reliably return the comments because the '_cellIndex' list entries were not in order. The 'IndexOf' method assumes that the '_cellIndex' list entries are sorted. In my case, the entries were out of order, so the 'Array.BinarySearch' method would not reliable look up entries. It is possible that the entries are normally ordered correctly, and my changes to the deserialization logic disrupted that order.