From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Eric Schulte" Subject: Re: Feature request: IDs on anything Date: Fri, 06 Mar 2009 06:23:14 -0800 Message-ID: <87ab7y3f3h.fsf@gmail.com> References: <20524da70903051728m4005f584p6b7b247e3b29936e@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1Lfax3-00074V-T4 for emacs-orgmode@gnu.org; Fri, 06 Mar 2009 09:23:01 -0500 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1Lfax1-00073F-Vj for emacs-orgmode@gnu.org; Fri, 06 Mar 2009 09:23:01 -0500 Received: from [199.232.76.173] (port=34397 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1Lfax1-00073C-Rs for emacs-orgmode@gnu.org; Fri, 06 Mar 2009 09:22:59 -0500 Received: from wf-out-1314.google.com ([209.85.200.175]:51608) by monty-python.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1Lfax1-0000IG-EV for emacs-orgmode@gnu.org; Fri, 06 Mar 2009 09:22:59 -0500 Received: by wf-out-1314.google.com with SMTP id 28so519178wfc.24 for ; Fri, 06 Mar 2009 06:22:57 -0800 (PST) In-Reply-To: <20524da70903051728m4005f584p6b7b247e3b29936e@mail.gmail.com> (Samuel Wales's message of "Thu, 5 Mar 2009 18:28:39 -0700") List-Id: "General discussions about Org-mode." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: emacs-orgmode-bounces+geo-emacs-orgmode=m.gmane.org@gnu.org Errors-To: emacs-orgmode-bounces+geo-emacs-orgmode=m.gmane.org@gnu.org To: Samuel Wales Cc: emacs-orgmode@gnu.org Hi, Lots of interesting ideas, and cool syntax options. My one proposed modification would be to only allow linking down to the table or list level, and then use existing reference syntax combined with the link to reference a particular cell or range inside of a table, or a particular element of a list. Maybe something like $[id B7423F4D-2E8A-471B-8810-C40F074717E9 range: @1$3..@3$3] Is this wholly new link syntax, or did I miss something? Thanks -- eric Samuel Wales writes: > Now seems like an ideal time to post this. > > I have been thinking that it would be useful to be able to slap org IDs on > anything. This includes plain list items, table cells, and > specific words in long sections of text.[1] Links to > these markers will never be broken and will go to their > exact locations. > > I am calling them =ID markers=. The syntax looks like > this.[2] > > $[id B7423F4D-2E8A-471B-8810-C40F074717E9] > > Here is an example: > > - this is a plain list > - example $[id B7423F4D-2E8A-471B-8810-C40F074717E9] > - the above can safely be linked to > > You can label markers to make them prettier: > > $[id B7423F4D-2E8A-471B-8810-C40F074717E9 :label "foo"] > this is a marker labeled "foo" (similarly to how links > are labeled). > > $[id B7423F4D-2E8A-471B-8810-C40F074717E9 :label ""] > now the marker is invisible unless you set links to be > visible or go to and edit the marker.[3] > > A key aspect of this feature is that it is extensible[2] > in various[4] ways. > > > I have more notes, including applications, but also want to > gauge interest in the basic idea. > > Is this appealing? > > > Footnotes: > > [1] This might also work for Charles Cave's thread, "My > Python solution [...]", which seeks IDs or the equivalent in > headlines. > > ID markers should work in non-org files (provided that org > is told about their existence via a user variable). Thus, > you can safely link to source code. > > [2] This syntax is motivated in a thread on the org > mailing list ( > [http://www.google.com/search?num=100&hl=en&ie=UTF-8&oe=UTF-8&q=%22extensible+syntax%22+%22parsing+risk%22+%22samuel+wales%22&btnG=Search] > ) named "extensible syntax". Some benefits: > > 1. You can add /new/ org features. > - This is done by reserving a new first element. > - For example, the keyword for the ID marker feature > is "id". > - If you want to add a new org feature for, say, > changing the color of a region of text, you would > use the keyword "color". > - You can do this with no new lexing code or syntax > debugging. > 2. You can extend /existing/ features. > - This is done with a keyword argument (plist key). > - For example, ID markers accept a :label keyword. > - To make the label be different in the exported text, > the key would be :export-label. > - To turn an ID marker into a link, the key would > be :link and its argument would be the link itself. > - I will motivate this and its applications in > another thread. It enables the user to create > arbitrary graph-theoretic structures, including > bidirectional links and tours through a table, by > pointing ID markers to one another. More later. > - No new lexing code or syntax debugging is necessary. > > A bonus: in principle, the facility can be opened up to the > users, who can then experiment with new features in their > .emacs files (without modifying org code) then spring them > on the rest of us. :) However, this is not essential to the > idea. > > [3] I am not sure, but it is possible that running M-x > visible-mode would also work. Or perhaps a standard org > command could do it. > > [4] For example, to make the label be different in the > exported text, it could look like this: > > $[id B7423F4D-2E8A-471B-8810-C40F074717E9 > :label "foo" > :export-label "bar"] > the exported version is labeled "bar", while your source > is still nicely labeled "foo". > > $[id B7423F4D-2E8A-471B-8810-C40F074717E9 > :label "foo" > :export-label ""] > now it is invisible when exported. but it can still be > pointed to. > > Or to make it easy to remember ID markers with a short > number: > > $[id B7423F4D-2E8A-471B-8810-C40F074717E9 :label :file-unique] > this is a marker labeled with a small, automatically > generated number that is only guaranteed to be unique > for the current file. > > My point in this footnote isn't that these are needed > subfeatures, but that with extensible syntax we can do this > kind of thing.