encoding incompatibility? help...

Tak

New member
Joined
Mar 31, 2004
Messages
3
Location
Tokyo,Japan
Hello all,

I've used DA in Japanese language.

I tried to write vacation message and comments,autoresponder etc...

but the messages would be displayed badly as follows...

ex. html source
PHP:
<textarea rows=15 cols=60 name=text>&#62 &#131&#92&#131t&#131e&#131&#139 ..... </textarea>

Is a multi-byte character supported?
if there are those who are using it in other languages, please let me know.

# a skin is Mr. trushkin's persephone. I've used putting it into Japanese.

Thank you.
 
Any characters that arn't a-zA-Z are converted to encoded html string.. I don't have any experience with writing japanese, but I'm wondering if an encoding tag at the top of the skin html would fix the problem? ..

John
 
Thank you for reply John,

I've added already charset,but the problem were not solved.

PHP:
<meta http-equiv="Content-Type" content="text/html; charset=shift_jis">


then i saw

/etc/virtual/domain/reply/***.msg

file is no problem.

Ummm...
 
Hello,

do these characters work on another webpage?
PHP:
<textarea rows=15 cols=60 name=text>&#62 &#131&#92&#131t&#131e&#131&#139 ..... </textarea>
Could you post a sample from the /etc/virtual/domain/reply/***.msg files? I never really used those characters so I'm not quite sure what to expect them to look like :)

John
 
Hi John.

do these characters work on another webpage?
PHP:
<textarea"&#147&#250&#150&#123&#140&#234 </textarea>

No, the code is not displayed correctly.


The content of an original file (/etc/virtual/domain/***.msg) is
"“ú–{Œê"(Sorry,it won't be displayed)


However, it has changed to
"&#147&#250&#150&#123&#140&#234" in html sauce.

The DA is very nice things!
I hope it'll become popularity over the world, and I also want to help.

T.Ogawa
 
Ok,

Looks like I'll have to figure out whcih charactes are safe not to html encode. There were issues with having some charactes not encoded before, so we currently have 2 conflicting requirements to make it work for both cases. I'll need to read up on the proper way to show special characters, so that they actually show up properly, but still maintain proper functionality. Not sure what the outcome will be.

John
 
Just an example.
1) Creating a new file in filemanager.
2) Pasting string
Code:
ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþÿ¨¸
into textarea.
3) Clicking "Save As"
4) Viewing file thru FTP - String saved correctly. No problems with cyrillic charset.
5) Again in FileManager: Clicking "Edit" (test.txt)
6) Now string in textarea displayed incorrectly:
Code:
ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþÿ¨¸
7) Clicking "Save As"
8) Clicking "Edit"
9) Now string in textarea changed to:
PHP:
ÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞßàáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþÿ¨¸

Yeah... Something wrong with that... :(
 
Last edited:
One more note.
Something wrong with converting characters to html-entities.
For example,
PHP:
Character 'À' must be printed as 'А', but DA print '&#192'
 
Hmm.

When I paste &#38#192, I get À, but when I try out &#38#1040 I just get A (without the \ ).

I'm also going from http://www.asciitable.com, then click HTML Codes. Go to 192, and you'll see "A, grave" which is À.

Where does &#38#1040 come from? Do you have a chart somewhere? Also, what encoding is your browser set to? I'm on UTF-8

John
 
Originally posted by DirectAdmin Support Hmm.

When I paste &#38#192, I get À, but when I try out &#38#1040 I just get A (without the \ ).

I'm also going from http://www.asciitable.com, then click HTML Codes. Go to 192, and you'll see "A, grave" which is À.
'À' in cyrillic (win) encoding is 'А' not 'À' or 'À'
There is no "Agrave" character in cyrillic codepages.

The problem is - when displaying file in textarea for editing directadmin converts characters to htmlentities, but when converted to htmlentitie character becomes independent from browser charset, so 'À' always displayed as "LATIN CAPITAL LETTER A WITH GRAVE" regardless of browser charset selection.

DirectAdmin MUST NOT convert characters to htmlentities before displaing, because he can't determine original codepage of single-byte characters in file.

(Same problem with displaying ticket content and subject in message system, and with some other pages.)

Where does &#38#1040 come from? Do you have a chart somewhere?
http://www.kostis.net/charsets/cp1251.htm for example.
Also, what encoding is your browser set to? I'm on UTF-8
John
Cyrillic (Windows) or cp1251 - most common russian codepage.
 
Last edited:
Hello John,

I have been trying to get Hebrew working on the Autoresponder and am having the same problem.
 
Hello,

I'll take a gander at everything and see about getting it into the next release. It's tricky for me to figure it all out because I can't tell what's it's supposed to look like (don't know other languages) :) I'll figure it out.

John
 
We are waiting too.
1.22.4 - still not fixed :(
This problem is very very big problem!
Fix it as soon as possible, either DA - hosting panel not for Russia :(
 
Hello,

I'll look into finally getting this done. I just need to test out the effects (if any) of not encoding the characters in the text areas.

John
 
Back
Top