Linguan un-escapes all texts it reads from strings files so that they can be displayed correctly. You don’t want to have to deal with backslashes in the editor. Now on writing I had thought it to be a good idea to also escape everything, including question marks, backslashes and single quotes. It was a bad idea to do that because this unduely changes the tokens and translations.
Translators (the sad kind without Linguan) would be confused by having to deal with lots of \’ in the text. And there where some reports of the localized string loading functions not being able to find the key. Not always, but this is clearly not what we wanted.
This update relaxes the escaping on writing to the extent that only double quotes and invisible characters are now being escaped.