[Rd] Fwd: Re: RSiteSearch, sos, rdocumentation.org, ...?

Jonathan Baron baron at upenn.edu
Thu Dec 22 00:22:08 CET 2016

Unfortunately, I am unable to get this search site working again. (The
message below explains why I had to rebuild it.)

The computer worked for the better part of a day downloading and
installing all the help files from all CRAN packages. Somehow it
failed to get the vignettes this time. But I pushed ahead and ran the
part of namazu that makes the search indices: mknmz. And you can see
the results if you go to finzi.psych.upenn.edu. The search works, but
then you can't get the actual help pages.

The problem seems to be that namazu search is ignoring the "Replace"
rules in the configuration file, namazurc. I have tried 4 versions of
that file, in combination with 3 versions of namazu-cgi, and nothing
works. Note that the last version of namzu was about 2009, the result
of some fixes by linux distribution maintainers, not by the original
creators, who gave up a few years before.

If someone wants to replace namazu with a different search engine, let
me know. I do have a server with all the help files.

And I can tell you how I made them if you want to do this somewhere
else. The trick is:




This works as is in the fedora version of R, which I think is modified
(for my benefit, about 10 years ago) from the distribution version,
and I think I know what the modification is. The first is for packages
I already have. The second can be used to build the site.

(I use the bioconductor repository because most of the time it doesn't
matter, but there were once a few packages that were only there, ones
that I used myself, like multtest.)

Of course, if there is a human being who reads this and wants to
fiddle with namazu, he or she should contact me.


On 12/17/16 15:32, Jonathan Baron wrote:
>Spencer and others.
>I am going to have to take down the server for RSiteSearch, which is
>finzi.psych.upenn.edu, for at least a couple of days starting Sunday
>morning. It has been hacked. And I have another server that has also
>been hacked, which is higher priority (sjdm.org). On Monday, I will
>probably have time to rebuild that one, but I may not have time to
>rebuild finzi for another week. I will try to get it all done in one
>day, but I don't know if I can.
>Sorry about this.
>I thought that there was an alternative to this site, namely
>but, as bad is my site is, that one, I think, is worse.

Jonathan Baron, Professor of Psychology, University of Pennsylvania
Home page: http://www.sas.upenn.edu/~baron
Editor: Judgment and Decision Making (http://journal.sjdm.org)

More information about the R-devel mailing list