TAB as delimiter in Export search result to a CSV file?
Posted: 17 Nov 2017, 06:35
From the results window: "Save search results" > "Export search result to a CSV file" uses a semicolon as delimiter to separate the results, but a Tab would be more practical, as a semicolon can be part of a filename and Excel (2003) expects a Tab character by default as delimiter.
There are some more complications to this. In a CSV (Comma-separated values) file values that contain a comma by themselves should be enclosed in quotation marks. So a filename like abc,def.txt should be written as "abc,def.txt" in the export result. However even if I changed the semicolon into a comma using Notepad++, the exported CSV file still does not split at the comma. It turns out that AllDup exports the results as an UCS-2 L(ittle) E(ndinan) BOM encoded file. Now this encoding does not works in Excel (2003), but converting the CSV file to an UTF-8 or ANSI file solves it!
So I think the Export to a CSV file should
1) use a comma instead of a semicolon
2) wrap values containing a comma between quotes
3) use UTF-8 or ANSI encoding instead of USC-2
[running Windows 7 x64]
There are some more complications to this. In a CSV (Comma-separated values) file values that contain a comma by themselves should be enclosed in quotation marks. So a filename like abc,def.txt should be written as "abc,def.txt" in the export result. However even if I changed the semicolon into a comma using Notepad++, the exported CSV file still does not split at the comma. It turns out that AllDup exports the results as an UCS-2 L(ittle) E(ndinan) BOM encoded file. Now this encoding does not works in Excel (2003), but converting the CSV file to an UTF-8 or ANSI file solves it!
So I think the Export to a CSV file should
1) use a comma instead of a semicolon
2) wrap values containing a comma between quotes
3) use UTF-8 or ANSI encoding instead of USC-2
[running Windows 7 x64]