This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.
assumption in filing this here is that the values in project encoding dropdown are from common code or api. vs each project building own dropdown items. I don't see MS932 on the dropdown, yet its the encoding that nb shows when user is in ja locale either using en xp with regional settings to Japanese, or using actual localized windows for japanese. thus it seems that it must be a commonly used notation for japanese encoding on windows, but users would not be able to choose it if they want to change project encoding property. for chinese, on windows it reports GBK as the encoding and that is in the dropdown. if dropdown cannt have the ms932, then perhaps we can mention in docs about equivalent encoding to choose; shiftjis might be one; there are a lot of choices windows-.... in that dropdown.
From some references I found on web I assume that MS932 has some aliases, like Windows-31J or x-sjis or even SJIS. See: http://www.ingrid.org/java/i18n/encoding/table.html http://bugs.mysql.com/bug.php?id=7607 If any of those encodings are listed in the combo I think that there is no issue.
I think since we have many customers in Japan, and since nb itself reports MS932 to user, I think we need to have MS932 on the dropdown; and 2nd url below seems to say that sjis is not exactly the same as MS932. I agree that not each and every alias can be or should be on dropdown, but if one looks at the dropdown there seem to be many, many listings and some could be aliases for others. Can you consider adding just this one to the dropdown ? ken.frank@sun.com
IMHO we should not special-case any encodings. We should display a combo containing the canonical display name for every registered encoding, as reported by the JRE, and if necessary show known non-canonical aliases e.g. in a r/o text field next to it. According to http://java.sun.com/javase/6/docs/technotes/guides/intl/encoding.doc.html MS932 is also "windows-31j".
thats a good idea to show the non canonical in a r/o text field. its a little confusing on that web page since it seems to say that both columns are canonical names, although for different io packages. ken.frank@sun.com
I've printed all available charsets in JDK and 'windows-31j' is listed there, so if it's the same encoding as MS-932 I think that this issue can be closed or changed to enhancement to show other names for the one encoding, but I don't know how to get those names.
Try Charset.aliases().
Created attachment 46644 [details] available charsets and their aliases
Changing issue type to enhancement and summary to better describe actual problem. I've attached list of available charsets and their aliases on my windows machine. The list shows that there can be lot of aliases for one charset, so adding r/o text field might not be the best solution, probably some (multi-)line tooltip?
Created attachment 46646 [details] once more with correct MIME
aliases list looks good; but am wondering about fitting in the dropdown in the various project props windows dropdown items are supposed to show all their contents (resize dynamically) but since some items have many aliases, not sure all these would be seen, even at std 11 pt font but users can use larger fonts could the dropdown have a horizontal sbar ? or could there be a tooltip for each item that would show on top of dropdown and be able to be wider than it ? ken.frank@sun.com
is this something that can be done easily for 6.1 ? I don't want to suggest anything that might destabilize things since its a short release cycle and planning already done for it. ken.frank@sun.com
I'm afraid that we won't make it for 6.1 release.