Handling Duplicate File Names by Adding Unique Identifier in C#

Question

We are facing an issue in a C# project where we are trying to save files with a unique name. These files are generated from an mbox reader, and we're using the email's subject as the base of the filename. The problem arises when there are duplicate subjects; the expected behavior is that the filename should contain the email subject followed by a unique identifier (GUID) to avoid any collisions. However, in cases of duplicates, the filename doesn't include the GUID, and only the subject is seen. This causes overwriting of the file and we lose the data of the duplicate files.

What We Have Tried:

1. We attempted to add a unique identifier (GUID) to the filename. This works for unique files but fails to add the GUID for files with duplicate subjects.

2. We tried to add a counter to the filename when duplicates are detected. The filename should ideally look like "subject_GUID". In the case of duplicates, it should be "subject_1_GUID", "subject_2_GUID", and so forth. However, it only gives us "subject.eml".

3. We attempted to truncate the subject part to ensure that there are no issues related to path length, but the issue persisted.

4. We tried saving the file with a delay of 1 second, hoping it might help with the file detection process, but it didn't resolve the issue.

5. We also tried checking if the file exists before saving it and if it does, we tried to append the counter to the filename and then save it. But it resulted in an incomplete process where the progress bar stopped at 2/3 with no exception.

We are now seeking a robust solution to handle this duplicate file naming issue without losing any data and preserving the uniqueness of the file names.

string guid = Guid.NewGuid().ToString();

string[] subjectParts = new string[] { subject };

foreach (var subjectPart in subjectParts)
{
    var truncatedSubjectPart = subjectPart.Length > 30 ? subjectPart.Substring(0, 30) : subjectPart;
    var newVariable = truncatedSubjectPart + "_" + guid;
    var fullSubjectPart = newVariable;

    var filePath = Path.Combine(flspath, Path.ChangeExtension(SanitizeFileName(fullSubjectPart), ".eml"));
    FileInfo fileInfo = new FileInfo(filePath);

    message.Save(filePath, SaveOptions.DefaultEml);
}

"Although appending a unique identifier (GUID) before the subject part resolves the duplicate issue as in var fullSubjectPart = guid+"_"+subjectPart;, this doesn't match our desired filename format."

Consider a scenario where a folder contains three items: Item1, Item2, and another Item2 (a duplicate). In the process of conversion, the resultant file names I obtain are Item1_guid.eml and Item2.eml, totaling to only two files. This outcome is not as expected, as the original folder contained three items, not two.

Interestingly, when I reverse the order of the elements in the file name such as var newVariable = guid + "_" + truncatedSubjectPartguid;, I end up with the anticipated number of files after conversion. This results in three files: guid_Item1.eml, guid_Item2.eml, and guid_Item2.eml. Despite the last two items being duplicates, they are rendered unique due to the preceding unique identifier (GUID). This peculiar situation is the problem at hand that needs addressing. I've included the code which generates a unique identifier, or GUID, for each file.