From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marc-Oliver Ihm Subject: Re: [dev] New version of org-index.el --- A personal index for org and beyond Date: Sat, 21 Sep 2013 23:12:53 +0200 Message-ID: <877ge998ve.fsf@think.site> References: Mime-Version: 1.0 Content-Type: text/plain Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:41894) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VNUUJ-00036g-DF for emacs-orgmode@gnu.org; Sat, 21 Sep 2013 17:13:17 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1VNUUD-00082T-Hn for emacs-orgmode@gnu.org; Sat, 21 Sep 2013 17:13:11 -0400 Received: from plane.gmane.org ([80.91.229.3]:38577) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VNUUD-00082K-Af for emacs-orgmode@gnu.org; Sat, 21 Sep 2013 17:13:05 -0400 Received: from list by plane.gmane.org with local (Exim 4.69) (envelope-from ) id 1VNUUB-0000Cn-KT for emacs-orgmode@gnu.org; Sat, 21 Sep 2013 23:13:03 +0200 Received: from 91-67-33-113-dynip.superkabel.de ([91.67.33.113]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sat, 21 Sep 2013 23:13:03 +0200 Received: from marc by 91-67-33-113-dynip.superkabel.de with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sat, 21 Sep 2013 23:13:03 +0200 List-Id: "General discussions about Org-mode." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: emacs-orgmode-bounces+geo-emacs-orgmode=m.gmane.org@gnu.org Sender: emacs-orgmode-bounces+geo-emacs-orgmode=m.gmane.org@gnu.org To: emacs-orgmode@gnu.org Hi Alan Alan Schmitt writes: > > I'm trying org-index, and I let the assistant build the table for me. At > the end of the set, this error occurred: > > ,---- > | Saved org-index-id '50E94CE9-06B3-452E-96B6-FA47AA08B9D6' to /Users/schmitta/.emacs.d/custom.el > | let: No catch for tag: created-new-index, nil > `---- > > Is this problematic? > No, not really ... The assistant has finished its work properly and this tag is just thrown to bail out without signaling an error. However, you still get an error, which is an annoyance to address. Thanx for reporting ! > Here are a few notes I'm taking as I'm playing with it. > > First, the table seems to be different from the one on worg. Here are > the columns I have: > > | | | | | | comment | > | ref | link | created | count;s | last-accessed | ;c | > | | <4> | | | | | > > I guess "comment" includes Type, description, and Keywords? Or can I > extend this table as I see fit? > Yes, just add any further columns you need. As long as the do not have a semicolon, they are ignored. You can also reorder columns according to your needs. > Second, I don't see how to update a link in a table so that it points > somewhere else. Should I just copy the "Property" drawer to the new place? Yes, this is possible; or you could copy the id-property from the new place into the index-table. > Also, I noticed it's impossible to create a link to a file. If a file > does not have any org entry, then org-index link will fail with "Before > first headline at position 35591 in buffer ...". I can go around this by > creating an headline, but then I cannot use saveplace or something like > that to put me in the same place in the file (typically at the very end > of a long data table, to input more data). Currently org-index uses the function org-id-get-create to create the link. This function needs an org-mode nodes and whence org-index has the same limitation. However, it might be possible to create more general links here, that would be able to point to non-org files. I will check this for the next version. > Finally, what is the workflow to create a new reference? The way I do it > is "link" from the place to link, then call "+fill" in the table. Is > there a simpler way (that would do both in one stroke)? Definitely: With the initial prompt just choose "ref"; this will only create a new reference, but not a link. If you want both, then you currently have to use the workflow you described. However, having a feature, that allows both, would be easy to implement, if needed. best regards, Marc