At a certain point during an investigation, data will be generated that would add value to a timeline but are not formated or covered in Create timeline from JSON/JSONL/CSV file.
For such a case, assume all your data is in a file called raw_data and it is not a CSV or a JSON that matches the timesketch format and is not a plaso output file that could be uploaded according to the documentation.
In first place you might be triggered to look for:
raw_data --> timesketch
The idea from timesketch is not to copy parsing logic that is already covered in other tools, so the first look for parsing capabilities should be plaso parsing.
The ideal data-processing flow is
raw_data --> plaso --> timesketch
The plaso parsers are listed within the plaso documentation
If there is no plaso parser for your file format yet, please have a look at how to write a plaso parser
Of course the following could work as well
raw_data --> proprietary_raw_data2csv_parser --> timesketch
Where proprietary_raw_data2csv_parser would not end up in timesketch repository.