[whatwg] "Content-Disposition" property for <a> tags

[Julian Reschke]

On 2011-06-03 14:23, Dennis Joachimsthaler wrote:
> Am 03.06.2011, 10:23 Uhr, schrieb Eduard Pascual <herenvardo at gmail.com>:
>
>> On Thu, Jun 2, 2011 at 10:09 PM, Dennis Joachimsthaler
>> <dennis at efjot.de> wrote:
>>> By the way, another point that we have to discuss:
>>>
>>> Which tag should a browser favor. The one in HTTP or the other one in
>>> HTML?
>>
>> Is that really worth discussing? HTTP >> HTML: whomever provides the
>> file should have the last say about how the file needs to be served,
>> regardless of what a site referencing to it may suggest.
>>
>> Furthermore, when links point to URIs with any scheme other than
>> "http:", whatever the scheme defines about how to deliver the file
>> takes precedence.
>>
>> Thus, only in the lack of an actual Content-Disposition header, or its
>> equivalent on some other scheme, would the attribute given by the link
>> be used, just like an additional fallback step before whatever the
>> UA's default behaviour would be.
>
> I agree that I shouldn't even have asked since this is actually a no-
> brainer. I can't think of any good reason to overwrite the http header
> with the html attribute.
>
> Alright, so, moving on...
>
>> This grants the ability for any content provider to use an explicit
>> "Content-Disposition: inline" HTTP header to effectively block
>> "download links" from arbitrary sources.
>
> True. Is it still so that some browsers ignore the "filename" part
> of a content-disposition if an "inline" disposition is used?

Yes, see <http://greenbytes.de/tech/tc2231/#inlwithasciifilename>. 
Apparently only Firefox gets this right.

>  ...

Best regards, Julian