UTF8/ANSI encoding bug

[athei]

1.Create new php file, echo some utf8 chars, save as utf8 without boom.
In browser this characters are displaying ok.
2. Delete utf8 chars and echo non utf8 text. [CTRL+S]
3.Close and reopen file - this file is recognized as ANSI
4. echo some utf8 chars [CTRL+S]
In browser this characters aren't displaying ok, because file is threaten as ANSI not UTF8.
IMHO it's bug, because I shouldn't have to always after I close file and reopen go to "File/save as" and from list select encoding utf8...
This is happening also after I close an start program again.

[Karlis]

See this:
http://forums.blumentals.net/viewtopic.php?f=4&t=3333

[athei]

But i have encoding in this file

Code: Select all

header('Content-Type: text/html; charset=utf-8');

and when i add utf8 characters (for example: Ä™Ć³Å‚Å›Ä…Å¼Åŗćń ) again the utf8 isn't detected and file is still in ansi, so characters are not displayed properly.
Something like that isn't happening in program like notpead++, where utf8 is well supported and recognized all the time, and I don't have to manually select encoding.

[Karlis]

1) As you are saying that you have Unicode symbols included in the file and they are not detected (this should be impossible), please send me the file (software[ at ]latnet.lv). Unicode symbols are symbols that take two character spaces when viewed with non-unicode program and these should be detected 100%. I tried to create empty file with only "Ä™Ć³Å‚Å›Ä…Å¼Åŗćń", I saved the file as UTF-8 without BOM and it was detected properly. So these symbols DO get detected. The will not be detected if you have entered and saved them from a non-unicode program.

2) A workaround in your case could be placing

Code: Select all

<meta http-equiv="content-type" content="text/html; charset=utf-8" /> 

tag (even in a comment, will do) as

Code: Select all

header('Content-Type: text/html; charset=utf-8');

is currently not supported.

[athei]

Code: Select all

    header('Content-Type: text/html; charset=utf-8');

is currently not supported.

- and that's the reason!

1) As you are saying that you have Unicode symbols included in the file and they are not detected (this should be impossible),

- I'm afraid You don't understand me.
Read carefully my first post, I have

Code: Select all

<?php
header('Content-Type: text/html; charset=utf-8');
echo 'Ä™Ć³Å‚Å›Ä…Å¼Åŗćń';
?>

1.Save as utf8 without boom. - In browser this characters are displaying ok.
2. Delete utf8 chars and echo simple ascii chars so this file looks like

Code: Select all

<?php
header('Content-Type: text/html; charset=utf-8');
echo 'test';
?>

[CTRL+S] or save. File is still detected at bottom as utf8 without boom.
3.Close and reopen file (or restart program) - this file is recognized as ANSI
4. echo some utf8 chars so this file looks like

Code: Select all

<?php
header('Content-Type: text/html; charset=utf-8');
echo 'Ä™Ć³Å‚Å›Ä…Å¼Åŗćń';
?>

[CTRL+S] (or save and restart)
In browser this characters aren't displaying ok, because file is threaten as ANSI not UTF8. And I have to manually from menu save as and select utf8 encoding.

I hope You understand me now.
The solution is to have meta charset or manually save file from menu.

[Karlis]

I got you now Well, we could try to detect UTF-8 just from "charset=utf-8" but I wonder whhether it would cause any problems as it may happen that a non UTF-8 file accidentally has "charset=utf-8" string in it. So I am not sure whhether it would be good.