[Orgmode] Re: Generate list of all tags in use?

[Matt Lundin]

Uriel Avalos <address@hidden> writes:

> Is there a way to generate a list of all tags in use in all agenda files?

Yes. Please see the documentation for the function
org-global-tags-completion-table, as suggested in the other recent
thread on this topic:

http://permalink.gmane.org/gmane.emacs.orgmode/33995

The function above returns a list suitable for use with, say,
completing-read.

> I'm thinking of something like a tag cloud. 

Could you please clarify in which context you'd like this list generated
(in a buffer, as a list returned to another function, etc.)?

The following expression will return a list of strings (without text
properties) containing all the tags in your agenda files:

--8<---------------cut here---------------start------------->8---
(mapcar (lambda (tag) 
          (substring-no-properties (car tag))) 
        (org-global-tags-completion-table))
--8<---------------cut here---------------end--------------->8---

If you'd like to insert an alphabetical list of tags in an org buffer,
you could evaluate a source block such as the following:

--8<---------------cut here---------------start------------->8---
#+begin_src emacs-lisp 
  (mapconcat 'identity
             (sort (mapcar 
                    (lambda (tag) 
                      (substring-no-properties (car tag))) 
                    (org-global-tags-completion-table))
                   'string<)
             "\n")
#+end_src
--8<---------------cut here---------------end--------------->8---

Finally, you could write a function (or, say, a quick perl script) to
find all tags in your org files, sort them by the number of times they
appear, and spit out the results. I use the following:

--8<---------------cut here---------------start------------->8---
#!/usr/bin/perl
use strict;
use warnings;

my %tags;

while (<>) {
  next unless (/^\*+\s/);
  if (/\s:([\w:]+):\s/) {
    for my $tag (split(/:/, $1)) {
      $tags{$tag} += 1;
    }
  }
}

for my $tag (sort {$tags{$b} <=> $tags{$a}} keys %tags) {
  print "$tag ($tags{$tag})\n";
}
--8<---------------cut here---------------end--------------->8---

Save this script and name orgtags.pl or the like. Then simply run
orgtags.pl on the desired org files:

% perl orgtags.pl ~/org/*.org

And you should receive a listing of tags that looks like this:

home (185)
email (93)
read (75)
think (73)
errands (62)
phone (59)
CONTACT (49)
ATTACH (45)
yard (37)
desk (36)
BIB (34)
NEXT (28)
...
[snip]

Obviously, you could do a lot more, but this is an example of a "quick
and dirty" tag cloud.

HTH,
Matt