Wolfgang Jörissen wrote:
Birgit, I might be stupid (in any case, I am not Java-savvy), but isn't this the way they are displayed in these files anyway and once the applet is run, they will display just fine?
Yes, they will be displayed correctly.
And even if not:
It should be feasible to get all those 1500 files imported into an application that can load them and perform a search and replace action on the whole batch. There are not too many Umlauts, so that should be a matter of minutes.
Every advanced search and replace tool will do it in a breeze.
E.g. Search&Replace, BK Replace em etc.
I suppose Deja Vu X can do this.
The global search and replace operations in DVX may be too slow because of the GUI, a simple S&R tool will be faster.
BTW, I already used DVX as a perverse code page converter
Contact me in private if you need me to try this.
AFAIR in DVX you can export diacritics as letters or as entities, as you like.