From mboxrd@z Thu Jan 1 00:00:00 1970 From: Carsten Dominik Subject: Re: Re: skip entry with inherited tags Date: Thu, 22 Jul 2010 09:38:46 +0200 Message-ID: References: <20100518074232.GA10524@mteege.de> <4BF2778D.9070702@os.inf.tu-dresden.de> <87mxvxdsv7.fsf@fastmail.fm> <4C082225.6020007@os.inf.tu-dresden.de> <4C08ED73.1000104@os.inf.tu-dresden.de> <4C343FB2.3080204@os.inf.tu-dresden.de> <4C3CA546.3060000@os.inf.tu-dresden.de> <3BFF4D35-CD16-4A41-B0A4-383538586592@gmail.com> <4C3CCDE4.9000600@os.inf.tu-dresden.de> <6694F386-1367-43D6-9587-7F26ECBC9C57@gmail.com> Mime-Version: 1.0 (Apple Message framework v936) Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Return-path: Received: from [140.186.70.92] (port=59291 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1ObqMn-0003Sb-Ch for emacs-orgmode@gnu.org; Thu, 22 Jul 2010 03:38:54 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69) (envelope-from ) id 1ObqMl-00036A-Px for emacs-orgmode@gnu.org; Thu, 22 Jul 2010 03:38:53 -0400 Received: from mail-ey0-f169.google.com ([209.85.215.169]:37930) by eggs.gnu.org with esmtp (Exim 4.69) (envelope-from ) id 1ObqMl-00035t-Dz for emacs-orgmode@gnu.org; Thu, 22 Jul 2010 03:38:51 -0400 Received: by eydd26 with SMTP id d26so2132585eyd.0 for ; Thu, 22 Jul 2010 00:38:50 -0700 (PDT) In-Reply-To: <6694F386-1367-43D6-9587-7F26ECBC9C57@gmail.com> 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: Carsten Dominik Cc: Matt Lundin , Martin Pohlack , emacs-orgmode@gnu.org Hi Martin, On Jul 21, 2010, at 4:32 PM, Carsten Dominik wrote: > Hi Martin, > > I just looked at your patch. > > If I have a normal agenda (i.e. *not* a block agenda), then your > patch will cause the preset filter *not* to be applied. OK, that was obviously incorrect, sorry about that. However, it still does not work correctly, here is the counter example: ----------------------------------------------------------------------------- * TODO test 1 :a: * TODO test 2 :b: * at 1 :a: SCHEDULED: <2010-07-22 Thu> * at 1 :b: SCHEDULED: <2010-07-22 Thu> ------------------------------------------------------------------------------ With this custom command: (setq org-agenda-custom-commands '(("x" "testmartin" ((agenda "" ((org-agenda-filter-preset '("+a")))) (alltodo "" ((org-agenda-filter-preset '("+b"))))) nil nil))) the result is incorrect, both with and without your patch. The internal logic of the filter and the preset filter is such that it applies to the entire view, and you should not set in the local options for a command that is part of a block agenda view. I have now documented this limitation. Sorry that this took so long - thank you for your patience. Best wishes - Carsten > > - Carsten > > On Jul 13, 2010, at 10:34 PM, Martin Pohlack wrote: > >> Hi Carsten and Matt, >> >> On 13.07.2010 20:48, Carsten Dominik wrote: >>> On Jul 13, 2010, at 7:41 PM, Martin Pohlack wrote: >>>> Hard to respond to this vague situation :-). >>>> >>>> It would be great if you could point me at more specific situations >>>> that >>>> might break or if others could test the patch. I have been using >>>> it >>>> for >>>> more than a month now without problems. >>> >>> The specific one I meant is if you use r or g to rebuild the agenda, >>> if you do something like a refile command which does an automatic >>> rebuild, >>> do you get back the view you expected? >> >> Yes, I regularly use 'r'. It works for me. Refiling a single >> entry did >> not show a problem. >> >>> Also, if you apply other filter commands, either with "/", or >>> narrowing >>> the filter with "/", does that give the expected results while you >>> are >>> in your block agenda? >> >> I quickly applied a tag filter using '/' 'tab'. Only the relevant >> entries were shown. Some block agendas become empty as expected. >> Clearing the filter restores the expected full view. >> >> I am not entirely sure that the patch does not have problems though. >> >> Matt: could you give the patch a quick test at your end? >> >>>>> I will only be able to study this more closely after the release. >>>> >>>> Take your time, I will wait for more details to emerge or feedback >>>> from >>>> other testers. >>>> >>>> One more note here: The current situation for block agendas is a >>>> bit >>>> problematic as their limitations are not documented afaik. >>>> >>>> Writing custom agendas is not easy in itself. If things don't >>>> work, >>>> it >>>> is really hard to distinguish between driver errors and >>>> limitations of >>>> the block agendas. >>> >>> Do you have suggestions on how to improve the situation? >> >> * Well, the best thing to do would be to remove the limitations of >> the >> block agendas, obviously :-), which I hope this patch does. >> >> * The second best thing is to document them. >> >> I am aware of this tag filter limitation, but no others from the >> top of >> my head. Do others know more? >> >> Cheers, >> Martin > > - Carsten > > > - Carsten