TYPO3 Core

Please explain: I cannot reproduce this problem.

What version of htmlArea RTE? Pasting from? Please attach a file to paste from to reproduce the problem.

I cannot reproduce this problem with the attached example on a utf-8 backend with htmlArea RTE version 1.2

Reopened on request from Jörg Wagner.

I can confirm the effect on several windows based servers (two WinXP testing servers, one Win2003 live server).
The clients (all Windows) use IE and FF. On some client machines the problem can be reproduced using both browsers, on others only FF reproduces the problem.
The problem vanishes when RTE.default.enableWordClean is set 0.
I affects all non ASCII characters such as umlauts, accented characters, typographical quotes. The affected chars are replaced by combinations of two characters (looking very much like UTF-8 combinations).

As Stanislas could not reproduce it, it might be either dependant on the client OS or even the server OS. I will send access codes to an affected system by personal email to Stanislas.

I attach a screenshot that shows the problem.

Problem confirmed on these systems:
htmlArea 1.2.0
Typo3 beta3 & RC1
Windows XP / 2003
Clients Windows FF & IE

I removed plugin TYPO3HtmlParser from the list of available htmlArea plugins. Since no parsing configuration is provided, it should not have been activated. I will check why it was.

Then, I have to see why the character set conversion occurs when the plugin is used.

This is now fixed in TYPO3core cvs, branch TYPO3_4-0.