Item8085: TinyMCE converting international characters into HTML entities

Priority: Enhancement
Current State: Closed
Released In: 1.1.3
Target Release: patch
Applies To: Extension
Component: TinyMCEPlugin
Reported By: Foswiki:Main.GregorRudolf
Waiting For: Main.MichaelTempest
Last Change By: KennethLavrsen
Some slavic letters (Latin S with caron - Š and š) are transformed by TinyMCE editor into &+Scaron; which makes text search unusable. Raw editor works fine.

A quick fix is to remove "352,Scaron,353,scaron," from entities definition in /pub/System/TinyMCEPlugin/tinymce/jscripts/tiny_mce/tiny_mce.js and tiny_mce_src.js

-- GregorRudolf - 25 Mar 2009

Hm. It work's fine for me. Maybe it depends on your encoding settings?

-- SvenHess - 16 Jul 2009

Yes, it could be. Pages are encoded in ISO-8859-2, site locale is set to "sl_SI.iso-8859-2", charset to "iso-8859-2".

-- GregorRudolf - 07 Sep 2009

This can be worked around in a TinyMCE setting. You might choose to set entity_encoding : "raw" as per the moxiecode docs. I have set up a sandbox topic with this in effect, see TinyMCEPluginWithoutEntityEncoding (actually that's a lie, it still encodes XML chars)

Gregor, Sven: can you share your locale settings in configure, and the web browser you're testing with

-- PaulHarvey - 30 Nov 2009

Upgraded to 1.0.9, the behavior stays the same - TinyMCE converts Š and š into &+Scaron; when saving page, rendering the search and topic previews (Recently changed topics section in WebHome, WebIndex) unusable.

Site locale is set to "sl_SI.iso-8859-2", charset to "iso-8859-2", all tested browsers (Chrome 4, Firefox 3.6, IE 8) render the pages as "text/html; charset=iso-8859-2".

IMO this isn't a bug in either Foswiki or TinyMCE but a somewhat unfortunate default TinyMCE configuration.

The "entity encoding=raw" or fixed "entities:" entry in TINYMCEPLUGIN_INIT setting are a much better solution than manually editing tiny_mce.js on each upgrade. smile Thanks!

-- GregorRudolf - 05 Feb 2010

1.0.9's WysiwygPlugin can corrupt numeric entities in the range 127-255 with charsets other than ISO-8859-1, so entity_encoding: "numeric" should not be used on 1.0.9.

Trunk's WysiwygPlugin now handles numeric entities without corruption, so either entity_encoding: "raw" or entity_encoding: "numeric" would be acceptable on trunk.

I am uncertain about what might break if we change the TINYMCEPLUGIN_INIT setting to entity_encoding: "raw" - some (older) browsers might misbehave. That makes me lean towards numeric on trunk, and adding to the installation notes for TMCE on the 1.0.x release branch.

-- MichaelTempest - 21 Jun 2010

This was set WaitingFor me, so here is my feedback: I think for trunk, the suggestion of entity_encoding: "numeric" sounds like a reasonable way forward.

As nobody else is responding, Michael, I've set this to you because although I could make the changes, you are more familiar with encoding issues in wyswiyg and what might break smile

-- PaulHarvey - 30 Aug 2010

Fixed in Item9973. Added a new setting TINYMCEPLUGIN_ENTITY_ENCODING so users don't have to go through the pain of managing TINYMCE_INIT so closely.

GregorRudolf, I added you in the WaitingFor on that task.

-- PaulHarvey - 22 Nov 2010

ItemTemplate edit

Summary TinyMCE converting international characters into HTML entities
ReportedBy Foswiki:Main.GregorRudolf
Codebase 1.1.2, 1.1.1, 1.1.0, 1.0.9, trunk
SVN Range Foswiki-1.0.0, Thu, 08 Jan 2009, build 1878
AppliesTo Extension
Component TinyMCEPlugin
Priority Enhancement
CurrentState Closed
WaitingFor MichaelTempest
TargetRelease patch
ReleasedIn 1.1.3
Topic revision: r9 - 16 Apr 2011, KennethLavrsen - This page was cached on 27 Mar 2017 - 08:40.

The copyright of the content on this website is held by the contributing authors, except where stated elsewhere. See Copyright Statement. Creative Commons License