Skip to content

CRAN: No MacOS X binary builds since January 7

5 messages · Brian Ripley, Simon Urbanek, Henrik Bengtsson

#
FYI,

no MacOS X binaries have been built for CRAN since 2010-01-07:
pattern <- ".*([0-9]{2}-[A-Za-z]{3}-[0-9]{4}) ([0-9]{2}:[0-9]{2}).*";
y <- grep(pattern, x, value=TRUE);
y <- grep("PACKAGE", y, invert=TRUE, value=TRUE);
z <- gsub(pattern, "\\1", y);
z <- unique(z);
z <- as.Date(z, "%d-%b-%Y");
z <- sort(z);
print(z);
[1] "2009-07-18" "2009-07-19" "2009-10-02" "2009-10-16" "2009-10-27"
 [6] "2009-10-28" "2009-10-29" "2009-10-30" "2009-10-31" "2009-11-02"
[11] "2009-11-04" "2009-11-05" "2009-11-06" "2009-11-07" "2009-11-09"
[16] "2009-11-10" "2009-11-11" "2009-11-12" "2009-11-13" "2009-11-14"
[21] "2009-11-16" "2009-11-17" "2009-11-18" "2009-11-19" "2009-11-20"
[26] "2009-11-21" "2009-11-23" "2009-11-24" "2009-11-25" "2009-11-26"
[31] "2009-11-27" "2009-11-28" "2009-11-30" "2009-12-01" "2009-12-02"
[36] "2009-12-03" "2009-12-04" "2009-12-07" "2009-12-08" "2009-12-10"
[41] "2009-12-11" "2009-12-12" "2009-12-14" "2009-12-15" "2009-12-16"
[46] "2009-12-17" "2009-12-18" "2009-12-19" "2009-12-21" "2009-12-22"
[51] "2009-12-23" "2009-12-24" "2009-12-25" "2009-12-27" "2009-12-29"
[56] "2009-12-30" "2009-12-31" "2010-01-02" "2010-01-03" "2010-01-04"
[61] "2010-01-05" "2010-01-06" "2010-01-07"
1  2  3 11 14 75
46 12  1  1  1  1

/Henrik
#
Not an issue for *this* list!

Please report to the maintainer and perhaps cc R-sig-mac.  Note that 
you are looking at the (old) Tiger binaries and not the more current 
Leopard ones, which were last updated yesterday,

In any case, binary packages are a privilege and you can always 
install from the sources (in the vast majority of cases with no extra 
tools other than Xcode).
On Sun, 17 Jan 2010, Henrik Bengtsson wrote:

            

  
    
#
On Mon, Jan 18, 2010 at 1:36 AM, Prof Brian Ripley
<ripley at stats.ox.ac.uk> wrote:
I used this list to share this with package developers - not
particularly MacOS X users.  As a package provider  I'd like to know
when packages are not available on all platforms.  It seems like a
errors, because the records show that packages are typically built
every day.
Maintainer has been notified repeatably without response (==don't know
if my messages even gets through).
Thanks for this note.  As a non-OSX user, I wasn't aware of this.  It
made me find:

http://cran.r-project.org/bin/macosx/leopard/contrib/r-release/

This is not the directory that are used for the MacOS X links when
going to package pages under http://cran.r-project.org/web/packages/,
e.g.

http://cran.r-project.org/web/packages/aroma.core/

The MacOS X links is:

http://cran.r-project.org/bin/macosx/universal/contrib/r-release/aroma.core_1.3.1.tgz

Another example is here:

  http://cran.r-project.org/web/packages/biglars/

Then, do the MacOS X links on the CRAN package pages need to be
updated/complemented?

If this is already well known, that is all I need to hear.  (I
understand that install.packages() takes care of the installation).

/Henrik
#
On Jan 18, 2010, at 7:53 , Henrik Bengtsson wrote:

            
I wonder where you were sending your notifications -- quite apparently  
not to the right place as I didn't get any ...

Anyway, now that it reached me (through suboptimal channels as Brian  
pointed out) it should be fixed for the next run.
I think so -- there is currently a discrepancy - the checks show  
Leopard builds but that link is to the Tiger build -- Kurt, can you,  
please, check?
1 day later
#
Thanks,

the "universal" MacOS X binaries are now updated/built, which are what
the CRAN package pages refer to.

Just for others information: I've been emailing offline with Simon U
("maintainer"), and it turns out that my initial posts to him
regarding these issues went into his spam folder.  Without wanting to
raise expectations of his responsiveness, he basically said if you
don't get a reply within a few days, your post most likely ended up in
his spam folder.  May be good to know.

/Henrik

On Mon, Jan 18, 2010 at 6:54 AM, Simon Urbanek
<simon.urbanek at r-project.org> wrote: