Tags
People
Badges
Help-
CORE FRAMEWORK
FEATURED COMPONENTS
-
RESOURCES
-
LEARN ZK
RESOURCES
Hi,
I've discovered that zk systematically removes or hides the charset information from Media objects containing plain text. I can't imagine why this is done, but I think this information has value, specially when it comes to download that very media object (the browser cannot tell which one to apply and is likely to do it wrong) or if, by whichever reason, you need to reproduce the actual bytes that were uploaded.
If you write a custom charset finder, the charset it yields will be used to turn bytes into characters and discarded afterwards (but for files big enough to be translated into ReaderMedia, which anyway does not publicly display the charset)
If you write a custom AuUploader which takes care of appending the charset detected by a custom charset finder to the content type, AMedia will systematically strip it off the content type in the inner setup method.
Being so, no choice remains but to write special media wrappers that preserve the charset information or display it by means of a getter, as explained here.
Wouldn't it be possible for text media objects to preserve charset info in the content type? Or at least display a getter method does grant access to this information?
Kind regards
Yes it's surely possible. But I don't think it makes sense,
When you have text documents this information automatically used to convert the binary information into a character String or Reader.
If it's binary, it's well ... binary and the charset information is ignored on purpose (or embedded into the document such as xml). Is there a case where a text/* with content type with an encoding is not transformed into a String/reader properly, that needs to be fixed?
According to the MIME standard the charset is only applicable to text content type
From https://www.ietf.org/rfc/rfc2045.txt
For example, the "charset" parameter is applicable to any subtype of "text", while the "boundary" parameter is required for any subtype of the "multipart" media type.
So adding a charset to a binary content type doesn't make sense, to me.
In the other post you vaguely mentioned such a case "If by whichever reason you cannot read the media object's content as characters and are forced to read them as raw bytes ...". Do you have a more concrete example. Then I'd be happy to help.
Asked: 2015-09-01 08:14:58 +0800
Seen: 34 times
Last updated: Sep 03 '15
