Hi everyone,
we've been using CKEditor for a while. Somewhere during last year, we upgraded our version (we were using a really old version). We started having some problems I will describe later.
Last friday, I installed the very last version, and now I have even more problems:
Inputing a line break, or using the bold/italic/underline button will write: ​
White spaces are replaced by: Â
I tried changing the entire site's encoding (files were in ANSI, charset windows-1252, tried all utf-8). It doesn't seem to fix the problem (or caused other problems, such as the language file having question marks symbols instead of accented characters.
Do you know what would be the better way to fix all this? We have a lot of websites using CKEditor, having more and more problem as people switch to IE10 (even if my actual problem is on every browsers).
If you have a solution, I would be truly grateful.
Thanks a lot, Antoine
I have found the same issue.
I have found the same issue.
When I enters space It created new character  tailered by space.
This isssue happens only some time.
I am using CKEditor 4.0
Try updating to the latest
Try updating to the latest version. There were some issues with Chrome but they've been fixed in subsequent releases. HTML Purifier also used to cause this.
Customer and Community Manager, CKSource
Follow us on: Facebook, Twitter, LinkedIn
If you think you found a bug in CKEditor, read this!
I have the same problem, only
I have the same problem, only with Chrome.
I have the latest version, my cache is clear. What can I do ?
Thanks
urgent: having this problem as well
We have version 4.2 installed. The extra  characters are being added whenever text is saved. It happens in Firefox and Chrome. We've cleared the cache and restarted the web service. We have many applications running CKEditor.
Help!
Jeannine
Please tell us the most
Please tell us the most important first - when does it happen?
When you type in editor or when you save data and load it again. If possible all of you please reply.
Could you also tell us:
1. which server do you use?
2. Have you checked encoding on server - some servers need to have encoding set or changed from default ISO.
3. What does your application use - does it encode requests, streams etc. in utf-8
4. Do your DB's have UTF-8 set?
5. Does HTML page (I assume you are using HTML5) where editor is set has utf-8 set as well (<meta charset="utf-8">) ?
Appearance of Cap A with circumflex accent
This is more input on this problem many others seem to be having.
OUR SETTINGS
1. We use Apache 2 server.
2. Server has CharacterSet encoding set to "windows-1252". This setting has been confirmed for all locations: Server, PHP, MySQL, and every other place I can find to set it. Browsers set to same.
3. Website has Metatag setting CharacterSet=windows-1252.
4. DB's set to window-1252.
5. Website specifies HTML4 -- too many problems with much of our coding not being supported by HTML5. Website uses CharacterSet=1252. Using utf-8 is not an option for us, as it causes many other problems with the wrong characters being displayed.
THE PROBLEM
When our website displays text that has been entered with CKEditor, there are numerous places where a special character appears when it was never entered originally. The character is a Cap A with circumflex accent. This often appears after the closing period of a sentence, or a question mark, or often with a line break.
After studying this carefully for quite some time, it appears that:
1. Sometimes text that has been copied from another word processor program and pasted into the CKEditor will have extra non-breaking spaces after sentence periods, etc. and the CKEditor will replace these with & #194 ; followed by & #160 ; (which are the entities for Cap A, circumflex accent and non-breaking space in succession).
2. Whenever I insert a non-breaking space, using & nbsp ; the CKEditor replaces it with the   sequence (described in 1.). I tested this several times by clicking on the Source button. When clicked back and forth between normal and source, the CKEditor replaces it each time, resulting in a series of Cap A, circumflex accent character and one non-breaking space. This also occurs when I insert the entity & Acirc ; as well.
3. This transformation seems to occur when I click the Source button for sure, but it may happen at other times which I have not studied and documented. I know it is the CKEditor somehow, because we are using it for text fields in our website forms, and when I click the Source button and manually remove all the & #194 ; and & #160 ; characters, then SAVE (submit) the form, the text is entered into the database without those characters. If I click the Source a second time to go back to normal display before I save the data, then the characters are reinserted into the text and saved to the database.
4. I have tested this with Internet Explorer 11.0.9600, Firefox 29.0.1, and Chrome 34.0.1847.137 and this problem exists with all of them.
We would really be interested in getting a fix for this soon. Thanks
The only reasonable solution
The only reasonable solution is using UTF-8, which is a standard used by a great majority of websitest. If used correctly, unlike many other encodings, it will work perfectly for all possible language in the world.
Piotrek (Reinmar) Koszuliński
CKEditor JavaScript Developer
--
CKSource - http://cksource.com
--
Follow CKEditor on: Twitter | Facebook | Google+
Not Unreasonable Requirements
We are not concerned with any other language other than English. Our websites have been created over the last eight years using the windows-1252 charset. So much of our sites is dependent upon codes and characters (which do not exist the same in UTF-8) that we cannot use UTF-8 without completely remaking our websites from scratch. That "ain't gonna happen!"
HTML5 gives us the same problem, because it does not support features of HTML4 that are used extensively in our site.
All current browsers can deal with both the windows-1252 charset and HTML4. We need an editor module that will work perfectly with them too. If CKEditor is not designed to do so, and will not be corrected to accommodate those requirements, can you recommend one which does work?
Thanks,
No other solution is going to
No other solution is going to work when not served correctly by you server. You don't need to serve your all your content in UTF-8, but at least CKEditor files. It's totally doable by using .htaccess.
Piotrek (Reinmar) Koszuliński
CKEditor JavaScript Developer
--
CKSource - http://cksource.com
--
Follow CKEditor on: Twitter | Facebook | Google+
What Do You Mean?
The text content is created within the CKEditor and saved to the database. No files are involved except the data table files. So, I have to assume you mean adding a directive to an .htaccess file in the /ckeditor/ directory that states: AddDefaultCharset utf-8 so it will override the directive I have in the / root directory which has: AddDefaultCharset windows-1252, causing the server to serve files from the /ckeditor/ directory in utf-8. Correct? If this in not correct, what do you mean? Specifically. Nothing in your documentation indicated something like this would be necessary, but I can easily create that type of .htaccess file if you tell me that is what is needed.
CKEditor documentation does
CKEditor documentation does not cover this topic because it covers CKEditor, not servers. It mentions only that you should use UTF-8, because there are many servers and each of them has its own documentation. I don't know the details of configuring Apache too. Hopefully, other users will be able to help you.
Piotrek (Reinmar) Koszuliński
CKEditor JavaScript Developer
--
CKSource - http://cksource.com
--
Follow CKEditor on: Twitter | Facebook | Google+
Understood
I would think a program like CKEditor would contain in its documentation some details about known problems on specific servers and how to work around them. But I understand if the company cannot spend the time and resources doing this.
Thank you for your time and information as far as you were able to go. I do appreciate that. I will try the specific idea I had about placing an .htaccess file with the directive AddDefaultCharset windows-1252 in it in the /ckeditor/ directory, and report back if that solves this or not.
Test Failed
Nope. That didn't work. If it is of any value to you Reinmar, the CKEditor seems to replace & #194; with & Acirc; and leaves & Acirc; alone (spaces added here to avoid any code replacements in this comment). It also replaces & #160; with & Acirc;& #160; and replaces & nbsp; with & Acirc;& #160; as well. Obviously, CKEditor wants to do its own thing, no matter what I specify I want. Maybe this will help someone else, too.
Found the solution to remove special character while parsing
Hi
I was having same issue with ckeditor. I have fixed issue by replacing special character with empty string.
How I solved the issue is as follow:
1. I beautified the ckeditor minified file using jsbeautifier.org
2. after beautified file, I found the function where hrml is parsed. Below is the function. (search for "htmlParser.text.prototype" in your file)
a.htmlParser.text.prototype = {
type: 3,
writeHtml: function(l, m) {
var n = this.value;
if (m && !(n = m.onText(n, this))) return;
l.text(n);
}
};
3. I have replaced the line from above function
var n = this.value;
with
var n = this.value.replace(/[^(\x20-\x7F)]*/g, '');
This line replaces the special character with empty. This solved my issue. This might be a hack but it worked for me.
Hope this saves somebody's time.
Note: To keep file miinified, search for the exact code without beatifying file and add your code.
Thanks all