Mercurial 1.7.5 released!

David Champion dgc at uchicago.edu
Thu Feb 3 19:07:05 UTC 2011


* On 03 Feb 2011, Steve Borho wrote: 
> On Thu, Feb 3, 2011 at 12:06 PM, David Champion <dgc at uchicago.edu> wrote:
> > * On 03 Feb 2011, David Champion wrote:
> >>
> >> The site javascript looks at your browser user-agent string and decides
> >> which package best matches what you're browsing from.  If you're not
> >> browsing from a 64-bit (Windows?) platform then this is perhaps a bug
> >> in the javascript.  Regardless not everyone sees the same download link
> >
> > I apologize, there is a bug -- sort of.  The latest.dat file for the
> > TortoiseHG distribution site does not mark 32- and 64-bit versions
> > distinctly in its matching regexp.  This is a matter for the THG team
> > (Steve? Adrian?) to address.
> 
> It never has.  I have no idea what regexps would be appropriate.  I
> purposefully keep my http knowledge to a minimum.  Wouldn't want to
> lose the lyrics to all those 80s sitcoms.

Making your way in the world today takes everything you've got!

>From looking at some lists of common user-agents this may be impossible.
I'm not sure that Windows Universal builds declare the word size of
their platforms for any browser.  And I don't see anything consistently
in Navigator that could be bent to this purpose either.  Any advice from
someone who has experience with this stuff?  What browser artifacts can
we used to detect a Windows browser's word size?

If it is truly impossible to differentiate then the priority of 32-bit
could be raised above that of 64-bit in latest.dat to effect the
requested change -- if that's preferable.  I have no opinion on which
should be the default, myself.

-- 
David Champion  *  dgc at uchicago.edu  *  IT Services  *  University of Chicago



More information about the Mercurial mailing list