Re: [O] [RFC] Do not declare drawers to use them
I think these patches are increasing Org load time by orders of magnitude. Finding 7 files now takes a minute and a half. -- The Kafka Pandemic: http://thekafkapandemic.blogspot.com The disease DOES progress. MANY people have died from it. ANYBODY can get it. Denmark: free Karina Hansen NOW.
Re: [O] [RFC] Do not declare drawers to use them
Beforehand, I could find them in a second or two max. On 11/2/13, Samuel Wales samolog...@gmail.com wrote: I think these patches are increasing Org load time by orders of magnitude. Finding 7 files now takes a minute and a half. -- The Kafka Pandemic: http://thekafkapandemic.blogspot.com The disease DOES progress. MANY people have died from it. ANYBODY can get it. Denmark: free Karina Hansen NOW.
Re: [O] [RFC] Do not declare drawers to use them
Hello, Samuel Wales samolog...@gmail.com writes: Beforehand, I could find them in a second or two max. On 11/2/13, Samuel Wales samolog...@gmail.com wrote: I think these patches are increasing Org load time by orders of magnitude. Finding 7 files now takes a minute and a half. Interesting. Could you provide some profiling information? Thank you. Regards, -- Nicolas Goaziou
Re: [O] [RFC] Do not declare drawers to use them
It is the difference between instantaneous and more than 10 seconds per Org file. If you tell me the symbols to elp profile, I will do so. I can't do the polling profiler. On 11/2/13, Nicolas Goaziou n.goaz...@gmail.com wrote: Hello, Samuel Wales samolog...@gmail.com writes: Beforehand, I could find them in a second or two max. On 11/2/13, Samuel Wales samolog...@gmail.com wrote: I think these patches are increasing Org load time by orders of magnitude. Finding 7 files now takes a minute and a half. Interesting. Could you provide some profiling information? Thank you. Regards, -- Nicolas Goaziou -- The Kafka Pandemic: http://thekafkapandemic.blogspot.com The disease DOES progress. MANY people have died from it. ANYBODY can get it. Denmark: free Karina Hansen NOW.
Re: [O] [RFC] Do not declare drawers to use them
Hello, Samuel Wales samolog...@gmail.com writes: It is the difference between instantaneous and more than 10 seconds per Org file. If you tell me the symbols to elp profile, I will do so. I can't do the polling profiler. Instrumenting package org would be useful. Regards, -- Nicolas Goaziou
Re: [O] [RFC] Do not declare drawers to use them
Instrumenting package org is not an option. On 11/2/13, Nicolas Goaziou n.goaz...@gmail.com wrote: Hello, Samuel Wales samolog...@gmail.com writes: It is the difference between instantaneous and more than 10 seconds per Org file. If you tell me the symbols to elp profile, I will do so. I can't do the polling profiler. Instrumenting package org would be useful. Regards, -- Nicolas Goaziou -- The Kafka Pandemic: http://thekafkapandemic.blogspot.com The disease DOES progress. MANY people have died from it. ANYBODY can get it. Denmark: free Karina Hansen NOW.
Re: [O] [RFC] Do not declare drawers to use them
Samuel Wales samolog...@gmail.com writes: Instrumenting package org is not an option. Hmmm. Could you send me an example file then? You can replace [a-zA-Z0-9] with x if needed. Regards, -- Nicolas Goaziou
Re: [O] [RFC] Do not declare drawers to use them
Hello, Carsten Dominik carsten.domi...@gmail.com writes: On 21.10.2013, at 12:03, Nicolas Goaziou n.goaz...@gmail.com wrote: Hello, Carsten Dominik carsten.domi...@gmail.com writes: I think one objection agains this patch is that, I think, MobileOrg relies on #+DRAWERS lines to be present to define drawers. The patch also contains org-mobile.el modifications. Maybe the MobileOrg maintainers for both Android and iOS can chime in here? Obviously, a double check cannot hurt. Otherwise I don't have a significant objection against this change. I am wondering if it is going to introduce spurious drawers and parsing errors if someone happens to have a string :xxx: on a line by itself in the buffer. This might be unlikely, but I am not really sure. Actually, as far as the parser goes, :xxx: is insufficient to define a drawer. Indeed, the drawer has to be complete, i.e. it has to contain the :END: part. An incomplete drawer does not trigger a parsing error: it is parsed as a mere paragraph. OK, sounds good to me, then. Applied. Regards, -- Nicolas Goaziou
Re: [O] [RFC] Do not declare drawers to use them
Hello, Achim Gratz strom...@nexgo.de writes: I suggest that completion should still look at configured drawer names. Since the patch removes DRAWER keyword, there is no place left to configure drawer names. Using DRAWER keyword only for completion would be confusing, because it would then behave very differently from similar keywords, e.g. TAGS. Another option would be to collect drawer names already used throughout the buffer and offer it as a completion basis. It should be reasonably fast once parser cache is implemented. Regards, -- Nicolas Goaziou
Re: [O] [RFC] Do not declare drawers to use them
Hello, Carsten Dominik carsten.domi...@gmail.com writes: I think one objection agains this patch is that, I think, MobileOrg relies on #+DRAWERS lines to be present to define drawers. The patch also contains org-mobile.el modifications. Maybe the MobileOrg maintainers for both Android and iOS can chime in here? Obviously, a double check cannot hurt. Otherwise I don't have a significant objection against this change. I am wondering if it is going to introduce spurious drawers and parsing errors if someone happens to have a string :xxx: on a line by itself in the buffer. This might be unlikely, but I am not really sure. Actually, as far as the parser goes, :xxx: is insufficient to define a drawer. Indeed, the drawer has to be complete, i.e. it has to contain the :END: part. An incomplete drawer does not trigger a parsing error: it is parsed as a mere paragraph. Regards, -- Nicolas Goaziou
Re: [O] [RFC] Do not declare drawers to use them
On 21.10.2013, at 12:03, Nicolas Goaziou n.goaz...@gmail.com wrote: Hello, Carsten Dominik carsten.domi...@gmail.com writes: I think one objection agains this patch is that, I think, MobileOrg relies on #+DRAWERS lines to be present to define drawers. The patch also contains org-mobile.el modifications. Maybe the MobileOrg maintainers for both Android and iOS can chime in here? Obviously, a double check cannot hurt. Otherwise I don't have a significant objection against this change. I am wondering if it is going to introduce spurious drawers and parsing errors if someone happens to have a string :xxx: on a line by itself in the buffer. This might be unlikely, but I am not really sure. Actually, as far as the parser goes, :xxx: is insufficient to define a drawer. Indeed, the drawer has to be complete, i.e. it has to contain the :END: part. An incomplete drawer does not trigger a parsing error: it is parsed as a mere paragraph. OK, sounds good to me, then. - Carsten Regards, -- Nicolas Goaziou
Re: [O] [RFC] Do not declare drawers to use them
Nicolas Goaziou writes: At the moment, in order to use a drawer, one has to declare its name through `org-drawers' variable or DRAWERS keyword first. I'd like to simplify a bit the process and let the user create drawers on the fly instead. A valid name includes any word constituent, hyphen and underscore characters. Case isn't significant. That would be very welcome. It also cleans the internals (i.e. some dynamically scoped variables are removed) and makes syntax more portable. But we lose completion for drawer names. I suggest that completion should still look at configured drawer names. Regards, Achim. -- +[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]+ SD adaptations for Waldorf Q V3.00R3 and Q+ V3.54R2: http://Synth.Stromeko.net/Downloads.html#WaldorfSDada
[O] [RFC] Do not declare drawers to use them
Hello, At the moment, in order to use a drawer, one has to declare its name through `org-drawers' variable or DRAWERS keyword first. I'd like to simplify a bit the process and let the user create drawers on the fly instead. A valid name includes any word constituent, hyphen and underscore characters. Case isn't significant. It also cleans the internals (i.e. some dynamically scoped variables are removed) and makes syntax more portable. But we lose completion for drawer names. This patch is backward compatible, as long as previously user-defined drawers didn't include forbidden characters within their names. WDYT? Regards, -- Nicolas Goaziou From 8a2dad690caec96e6daf86097ff0cb062ac348f8 Mon Sep 17 00:00:00 2001 From: Nicolas Goaziou n.goaz...@gmail.com Date: Sun, 20 Oct 2013 14:40:09 +0200 Subject: [PATCH] Remove need to declare drawers before using them * lisp/org-agenda.el (org-agenda-prepare): Remove reference to `org-drawers-for-agenda'. (org-agenda-show-1): Remove reference to `org-drawers'. * lisp/org-clock.el (org-clock-remove-empty-clock-drawer, org-clock-cancel): Apply signature change to `org-remove-empty-drawer-at'. * lisp/org-element.el (org-element--list-struct): Use `org-drawer-regexp' instead of `org-drawers'. * lisp/org-feed.el (org-feed-drawer): Update docstring according to change. * lisp/org-list.el (org-in-item-p, org-list-context): Use `org-drawer-regexp' instead of `org-drawers'. * lisp/org-mobile.el (org-mobile-create-index-file): Remove reference to `org-drawers-for-agenda'. * lisp/org-pcomplete.el (pcomplete/org-mode/drawer): Remove function. * lisp/org.el (org-drawer-regexp): Make variable global. (org-drawers): Remove variable. (org-set-regexps-and-options): Ignore DRAWER keyword. (org-cycle): Use `org-drawer-regexp' instead of `org-drawers'. (org-cycle-hide-drawers): Add an optional argument to ignore some drawers. (org-remove-empty-drawer-at): Remove second argument. Rewrite function. (org-clone-subtree-with-time-shift): Apply signature change to `org-remove-empty-drawer-at'. (org-toggle-ordered-property): Apply `org-delete-property' signature change. (org-map-entries): Remove reference to `org-drawers-for-agenda'. (org-entry-delete): Remove optional argument. Small refactoring. (org-insert-drawer): Remove reference to `org-drawers'. (org-delete-property): Apply `org-entry-delete' signature change. (org-in-drawer-p): Rewrite function. * testing/lisp/test-org-element.el: Update tests. * testing/lisp/test-org.el: Update tests. * testing/lisp/test-ox.el: Update tests. --- lisp/org-agenda.el | 15 ++--- lisp/org-clock.el| 6 +- lisp/org-element.el | 5 +- lisp/org-feed.el | 5 +- lisp/org-list.el | 25 ++- lisp/org-mobile.el | 4 +- lisp/org-pcomplete.el| 19 -- lisp/org.el | 141 ++- testing/lisp/test-org-element.el | 57 ++-- testing/lisp/test-org.el | 22 +++--- testing/lisp/test-ox.el | 28 11 files changed, 109 insertions(+), 218 deletions(-) diff --git a/lisp/org-agenda.el b/lisp/org-agenda.el index 0bfba27..fe4c2d4 100644 --- a/lisp/org-agenda.el +++ b/lisp/org-agenda.el @@ -3642,7 +3642,6 @@ generating a new one. (or org-agenda-multi (org-agenda-fit-window-to-buffer)) (throw 'exit Sticky Agenda buffer, use `r' to refresh)) (setq org-todo-keywords-for-agenda nil) -(setq org-drawers-for-agenda nil) (unless org-agenda-persistent-filter (setq org-agenda-tag-filter nil org-agenda-category-filter nil @@ -3682,7 +3681,6 @@ generating a new one. (org-uniquify org-todo-keywords-for-agenda)) (setq org-done-keywords-for-agenda (org-uniquify org-done-keywords-for-agenda)) - (setq org-drawers-for-agenda (org-uniquify org-drawers-for-agenda)) (setq org-agenda-last-prefix-arg current-prefix-arg) (setq org-agenda-this-buffer-name org-agenda-buffer-name) (and name (not org-agenda-name) @@ -8627,15 +8625,10 @@ if it was hidden in the outline. (run-hook-with-args 'org-cycle-hook 'subtree)) (message Remote: SUBTREE)) ((= more 4) - (let* ((org-drawers (delete LOGBOOK (copy-sequence org-drawers))) - (org-drawer-regexp - (concat ^[ \t]*:\\( - (mapconcat 'regexp-quote org-drawers \\|) - \\):[ \t]*$))) - (show-subtree) - (save-excursion - (org-back-to-heading) - (org-cycle-hide-drawers 'subtree))) + (show-subtree) + (save-excursion + (org-back-to-heading) + (org-cycle-hide-drawers 'subtree '(LOGBOOK))) (message Remote: SUBTREE AND LOGBOOK)) (( more 4) (show-subtree) diff --git a/lisp/org-clock.el b/lisp/org-clock.el index 3195dc1..1cdbdc1 100644 --- a/lisp/org-clock.el +++ b/lisp/org-clock.el @@ -1589,7 +1589,7 @@ to, overriding the existing value of `org-clock-out-switch-to-state'. (while (and (
Re: [O] [RFC] Do not declare drawers to use them
Hello Nicolas, I think one objection agains this patch is that, I think, MobileOrg relies on #+DRAWERS lines to be present to define drawers. Maybe the MobileOrg maintainers for both Android and iOS can chime in here? Otherwise I don't have a significant objection against this change. I am wondering if it is going to introduce spurious drawers and parsing errors if someone happens to have a string :xxx: on a line by itself in the buffer. This might be unlikely, but I am not really sure. Comments? - Carsten On 20.10.2013, at 15:38, Nicolas Goaziou n.goaz...@gmail.com wrote: Hello, At the moment, in order to use a drawer, one has to declare its name through `org-drawers' variable or DRAWERS keyword first. I'd like to simplify a bit the process and let the user create drawers on the fly instead. A valid name includes any word constituent, hyphen and underscore characters. Case isn't significant. It also cleans the internals (i.e. some dynamically scoped variables are removed) and makes syntax more portable. But we lose completion for drawer names. This patch is backward compatible, as long as previously user-defined drawers didn't include forbidden characters within their names. WDYT? Regards, -- Nicolas Goaziou 0001-Remove-need-to-declare-drawers-before-using-them.patch
