Question

Hi, we are working in Germany, and we updated last Friday to the September 2001 version.

Until then, the german special characters were stored "as is". Now, when we load an old topic source with umlauts, they are changed to a 7-bit ascii character (e.g. "ä" -> "d", "ü" -> "|"). The result is hardly usable, as all instances of the 7-bit characters have to be replaced manually by the original umlauts.

When these special characters are saved, the are converted to CharacterEntities automatically. After clicking on Edit again, these character entities are visible in the topic source as e.g. ä
This is very difficult to edit, and decreases the popularity of TWiki in our company

Our problems in short:

• Old TWiki topic sources contain special characters as itself, and they are converted in a way which is unusable for us.
• Working with "correctly" stored TWiki topic sources is affected by unhandy HTML character entities like ä

Resulting questions:

• Can we switch back to the old way of special character handling?
• If not: is a plug-in available to convert the stored HTML character entity to the corresponding special character in the editor (and back)?

THNX in advance!

.

• TWiki version: September 2001
• Web server: Apache
• Server OS: Linux

-- JanPeterMohr - 24 Sep 2001

Answer

The same fix as the previous version seems to work for me, see: NoWarningWhenEnteringNonStdChars

-- ColasNahaboo - 25 Sep 2001