I'd vote for: figure out if this can be done properly (e.g. rely on
file system date/time file and HTTP), and if necessary add a hack to
support browsers like IE6.

On Feb 12, 3:54 pm, Ross Mellgren <[email protected]> wrote:
> It does normally, but not for CSS resources. I just did a quick google and I 
> found one page that said it will send a HEAD for a cached CSS but only when 
> the browser session is restarted. I know that at my work we've had to do the 
> filename hack because nothing else works with IE6.
>
> And IE6 may be dying out to some degree, but it's still popular enough (and 
> particularly in certain industries) that it has to be supported for many 
> companies :-/
>
> -Ross
>
> On Feb 12, 2010, at 3:51 PM, Alex Black wrote:
>
> > 1. Luckily IE6 is dying out :) unlesshttp://saveie6.com/works
> > 2. surely even IE6 obeys expires headers or some caching rules?
>
> > On Feb 12, 3:48 pm, Ross Mellgren <[email protected]> wrote:
> >> I believe IE6 does not follow the correct process you describe and will 
> >> always cache CSS files of the same name.
>
> >> -Ross
>
> >> On Feb 12, 2010, at 3:48 PM, Alex Black wrote:
>
> >>> hey guys, I love the enthusiasm, but putting a unique value on the css
> >>> filenames seems like a hack, surely we can do better?
>
> >>> Whats supposed to happen is:
> >>> - browser requests resource (e.g. styles.css) with a conditonal get
> >>> (if newer than X)
> >>> - server checks to see if resource is newer than X
> >>> - if it is new than x then: return resource
> >>> - if it is not newer than x, then return 304 not modified
>
> >>> - Alex
>
> >>> On Feb 12, 2:35 pm, Marius <[email protected]> wrote:
> >>>> On 12 feb., 21:31, Jeppe Nejsum Madsen <[email protected]> wrote:
>
> >>>>> On Fri, Feb 12, 2010 at 8:20 PM, Marius <[email protected]> wrote:
> >>>>>> Jeppe probably we can combine the two proposals.
>
> >>>>> Yes, that would be natural
>
> >>>>>> Perhaps something like:
>
> >>>>>> <lift:css name="mycss.css, some_other.css. /classpath/baz.css" />
>
> >>>>>> thus Lift could generate:
>
> >>>>>> <link rel="stylesheet" type="text/css" href="compound_2434rfe34534.css?
> >>>>>> i784yrfiuhferfhweir57=_"/>
>
> >>>>>> compound_2434rfe34534.css is a synthetic name that would contain the
> >>>>>> mycss.css, some_other.css. /classpath/baz.css concatenated. Same thing
> >>>>>> for JS. This content could potentially be compressed.
>
> >>>>> One thing that I think will be important (at some point :-) is to do
> >>>>> combining of individual tags. If a page is constructed from several
> >>>>> snippets/widgets, each emitting different js files (think jQuery
> >>>>> plugins) and css files, these need to be combined somehow. This means
> >>>>> that each page will get it's own unique synthetic css/js file. This
> >>>>> probably needs to be configurable in some way :-)
>
> >>>> Yeah that is a slightly different use-case that require more noodling.
> >>>> But would worth considering in the future.
>
> >>>>>> I can open a ticket and start looking into this.
>
> >>>>> Awesome! I'll watch from the sideline!
>
> >>>>> /Jeppe
>
> >>> --
> >>> You received this message because you are subscribed to the Google Groups 
> >>> "Lift" group.
> >>> To post to this group, send email to [email protected].
> >>> To unsubscribe from this group, send email to 
> >>> [email protected].
> >>> For more options, visit this group 
> >>> athttp://groups.google.com/group/liftweb?hl=en.
>
> > --
> > You received this message because you are subscribed to the Google Groups 
> > "Lift" group.
> > To post to this group, send email to [email protected].
> > To unsubscribe from this group, send email to 
> > [email protected].
> > For more options, visit this group 
> > athttp://groups.google.com/group/liftweb?hl=en.

-- 
You received this message because you are subscribed to the Google Groups 
"Lift" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/liftweb?hl=en.

Reply via email to