[R-wiki] list of all the pages

Romain Francois francoisromain at free.fr
Sun Jul 23 01:08:36 CEST 2006

Le 23.07.2006 00:39, Paul Johnson a écrit :
> Attached is my most recent version of the R program that generates all
> tips as a single sheet.  Maybe it will help you see what can be done.
> There is also an example of another approach that is listed in the
> wiki itself. I think the title is something like "how can I get a list
> of all tips" or something like that.
> pj
Thanks, I'll take a look at the R script,

on the bash way, I could torture the changes.log file of the wiki to get 
the info i want :
(maybe it's not the most efficient way)


rm -f changes.log
wget http://wiki.r-project.org/rwiki/data/changes.log

rm -f authors.txt
for i in `cut -f3 changes.log | sort | uniq`;
 echo -n "$i " >> authors.txt
 grep $i changes.log | cut -f4 | sort | uniq | tr "\n" " " >> authors.txt
 echo >> authors.txt
sed -e "s/  / /g" authors.txt > authors2.txt
mv -f authors2.txt authors.txt
Now, the file authors.txt contains the name of the page followed by the 
name of the person involved.
The problem is that this contains pages that are not anymore in the wiki 
because thay have been removed.

Another approach, but I don't have the rights to do that is to use the 
directory structure, something like that from the data/pages directory :

$ find -name "*.txt" | grep .txt | sed -e "s/\.\/\(.*\)\.txt/\\1/g" | tr 
"/" ":"

gives the names of all the pages that exist.

With a little more imagination, combining the two scripts, I could get 
to something like an xml file :

  <lastmodif>Sun Jul 23 01:05:28 2006</lastmodif>

which would do the trick to what I have in mind .... but now I should 
really go to bed, we'll see tomorow.

Philippe, when I'm done with that, would you consider setting a cronjob 
to generate that file from time to time and put it somewhere we can read 
it ?


> On 7/22/06, Romain Francois <francoisromain at free.fr> wrote:
>> Hi,
>> I'm trying to learn the ways of XUL : mozilla Xml User interface
>> Language (basically that's how they do firefox extensions and stuff).
>> I'd like to know if that is possible to get the some info about *all*
>> the wiki pages in a simple ascii file (or maybe an xml file) :
>>  * id of the page, ie something like tips:misc:alltips
>>  * last modif
>>  * who is involved in that page
>>  * etc ....
>> Cheers,
>> Romain

|                         Romain FRANCOIS                           |
| R Graph Gallery : http://addictedtor.free.fr/graphiques           |
| I'm looking for a job, see my cv : http://francoisromain.free.fr  |

More information about the R-sig-wiki mailing list