It’s a while for now that we have problems with OTRS generated CSV files and Unicode. We are conducting an OTRS implementation in a Persian (Farsi) context. OTRS’s Stats (reports) module can generate CSV as output. There is 2 problems here. First, delimiter character used is Semicolon (;) not Comma (,) and second, because of our Unicode environment, CSV file format is UTF8. Following screenshot shows such a CSV file opened directly in MS Excel 2007, all of fields is just located in one column and Persian (Farsi) text are showed incorrectly as some meaningless characters:
While searching for the solution in OTRS mailing lists, I found a good and very simple solution. Resolving the issue is possible by importing (not opening directly) the CSV with some considerations. All necessary steps are illustrated in the following.
1. Select “Get External Data, From Text” from Data menu and select desired CSV file:
2. A wizard opens. In the first step select “65001 : Unicode (UTF-8)” as “File Origin” and click “Next”:
3. In the second step select “Semicolon” instead of “Comma” and click “Next”:
4. In the 3rd step of wizard click “Finish” and in the “Import Data” window click “OK”. After this you will see your correct data in correct cell (each field in a separate column) and correct language (Persian text not garbage one) in MS Excel 2007:
Please consider that all of these instructions are applicable in MS Excel 2003 too.