[Development] Qml mime types

[Mohamed Fawzi]

If there are no other objections I will go ahead an propose this on the mime mailing list, and prepare a patch for this.
As most uses are qt specific, probably  we should be able to still change things without too many problems if we really see that it does not work.

Fawzi
On 10 Jan 2013, at 07:44, Mohamed Fawzi <Fawzi.Mohamed at digia.com> wrote:

> 20000 obviously needs more  than 256, and magic solving in general needs more resources, so that the mime resolution algorithm tries to avoid it if possible, and soem applications might not do it at all (if they don't have the content of the file, or only very slowly).
> By having qtquick1 and qtquick2 separated (or maybe not so much see latest comment by Alan in the Qml Runtime thread) but using a single extension, we force a correct resolution to to always use magic sniffing.
> 
> Still the normal case (linux desktop, or qt-creator) if one loads up less than 32K I think OS optimizations should keep the performance hit still in the negligible range.
> 
> Still maybe the need to treat qtquick1 and 2 differently will become less now
> 
> Fawzi
> ________________________________________
> From: apoe at hrz.tu-chemnitz.de [apoe at hrz.tu-chemnitz.de] on behalf of André Pönitz [andre.poenitz at mathematik.tu-chemnitz.de]
> Sent: Thursday, January 10, 2013 12:24 AM
> To: Mohamed Fawzi
> Cc: Rutledge Shawn; development at qt-project.org
> Subject: Re: [Development] Qml mime types
> 
> On Tue, Jan 08, 2013 at 05:44:05PM +0000, Mohamed Fawzi wrote:
>> That is a good point, currently the license is 1393 characters, so
>> I think 20000 should be enough (it doen't look like there is any
>> limit in the range, other than being implicitly 32 bit in some
>> places...
> 
> Would 20000 have any relevant performance implications?
> 
> Andre'
> _______________________________________________
> Development mailing list
> Development at qt-project.org
> http://lists.qt-project.org/mailman/listinfo/development