Hi Achim
On Wed, Nov 27, 2013 at 11:06 PM, Achim Gratz wrote:
> I've had a brief look at this. It seems that some of the differences
> are due to elp instrumenting a much larger range of functions when
> org-mode has already initialized vs. when only org-version had been
> called.
Thank you for
Achim Gratz writes:
> I've had a brief look at this. It seems that some of the differences
> are due to elp instrumenting a much larger range of functions when
> org-mode has already initialized vs. when only org-version had been
> called.
> Once that is out of the equation, there isn't any diff
Michael Brand writes:
> It shows a difference, see attachements. It looks like but it was not
> me loosing some first lines when doing this, I cross-checked :-).
I've had a brief look at this. It seems that some of the differences
are due to elp instrumenting a much larger range of functions when
Hi Achim
On Wed, Nov 27, 2013 at 9:05 PM, Achim Gratz wrote:
> Michael Brand writes:
>> I always do "make cleanall uncompiled". Is this correct?
>
> I don't recommend it, but it should work if you're using a non-buggy
> Emacs (i.e. not Emacs 23, which never ignores site-lisp).
Not recommend whic
Michael Brand writes:
> I always do "make cleanall uncompiled". Is this correct?
I don't recommend it, but it should work if you're using a non-buggy
Emacs (i.e. not Emacs 23, which never ignores site-lisp).
> Before I tried without the --eval, but I can reproduce quadratic with
> the following t
Hi Achim
Thank you for looking into this.
On Wed, Nov 27, 2013 at 7:13 PM, Achim Gratz wrote:
> Michael Brand writes:
>> 2) choice:
>>- for quadratic do: M-x org-mode
>>- for linear do: M-x org-version
>
> You have a botched installation. Make sure that the autoloads
> are current,
I a
Michael Brand writes:
> 2) choice:
>- for quadratic do: M-x org-mode
>- for linear do: M-x org-version
You have a botched installation. Make sure that the autoloads are
current, that the Org install directory comes first in load-path and
that you require org-loaddefs before anything else
Michael Brand writes:
> On Wed, Nov 27, 2013 at 12:27 PM, Nicolas Richard
>> M-: (insert (mapconcat #'car load-history "\n"))
>
> It shows a difference, see attachements.
It seems from my own tests that in the `org-version' case, org-element
is in fact not used.
Example with org-element loaded:
Hi Nico
On Wed, Nov 27, 2013 at 12:27 PM, Nicolas Richard
wrote:
> Can you show the output of
> M-: (insert (mapconcat #'car load-history "\n"))
> in each case ? It will output the list of loaded files, which hopefuly
> shows a difference.
It shows a difference, see attachements. It looks like b
Hi Michael,
Michael Brand writes:
> 2) choice:
>- for quadratic do: M-x org-mode
>- for linear do: M-x org-version
Can you show the output of
M-: (insert (mapconcat #'car load-history "\n"))
in each case ? It will output the list of loaded files, which hopefuly
shows a difference.
--
N
Hi Nicolas
On Tue, Nov 26, 2013 at 11:19 PM, Nicolas Goaziou wrote:
> So, it is a linear increase here. Have you disabled cache?
No, but in the meantime I found accidentally how to get linear and
bisected it to be related with how Org is loaded:
1) start GNU Emacs with -L pointing to Org and wi
Hello,
Michael Brand writes:
> What I don't understand is the still quadratic increase but if you
> expect that for opening then it is ok for me.
No, it should be linear. See below.
> Uncompiled release_8.2.3c-302-g15246d on a not new machine:
>
> Open 400er:
> org-mode1
Hi Nicolas
On Tue, Nov 26, 2013 at 9:09 PM, Nicolas Goaziou wrote:
> Is opening the file still slow? I get:
What I don't understand is the still quadratic increase but if you
expect that for opening then it is ok for me. Uncompiled
release_8.2.3c-302-g15246d on a not new machine:
Open 400er:
or
Hello,
Michael Brand writes:
> Thank you. As I wrote in the other post my recipe is not a real use
> case any more for me so it doesn't really matter that opening the Org
> file is still slow.
Is opening the file still slow? I get:
org-mode 1 0.458615558 0.45861
Hi Nicolas
On Tue, Nov 26, 2013 at 6:28 PM, Nicolas Goaziou wrote:
> I implemented it.
>
> `org-element-at-point' runs in a mean constant time now (worst case is
> still linear, though).
Thank you. As I wrote in the other post my recipe is not a real use
case any more for me so it doesn't really
> Please note that the parsing process is inherently linear, since we're
> in a context-dependent grammar. Thus, there's not much to do about the
> parsing algorithm.
>
> However, the cache is always up-to-date. So, we are not bound to start
> again from headline if we can find a cached element bet
Hi Nicolas
On Sun, Nov 24, 2013 at 5:25 PM, Nicolas Goaziou wrote:
> I didn't look at your code, but can't you `defadvice' `org-mode'?
Yes, thank you for the advice to defadvice org-mode (with class
"before") to get the same effect like a hook would have at the
beginning of turning on org-mode.
Hello,
Bastien writes:
> I noticed this too -- Nicolas, was the rationale for the rewrite just
> to prevent drawers to get flagged when they find themselves in source
> code blocks? Or are there other cases the rewrite fixes?
I somehow overlooked your message.
Anyway, preventing folding in a
Hello,
Michael Brand writes:
> First, I have an understanding question:
>
> I tried to find where org-mode-hook is called within the function
> org-mode but could not see something similar to the expected
> "(run-hooks 'org-mode-hook)". So I added 'backtrace to the
> org-mode-hook but don't unde
Hi Nicolas
On Sat, Nov 23, 2013 at 11:52 AM, Nicolas Goaziou wrote:
> I pushed a fix that should halve the time,
I see this effect, thank you.
> but it will still be slow.
>
> `org-element-at-point' is linear by the number of elements before point
> in the current section. Therefore, parsing /a
Hello,
Michael Brand writes:
> On Mon, Nov 11, 2013 at 5:41 PM, Michael Brand
> wrote:
>> I noticed that to open an Org file with a minimal example generated
>> like this slowed down:
>>
>> #+BEGIN_SRC sh
>> #!/bin/sh
>> echo '* a'
>> for ((i = 0; i < 400; i++)); do
>> echo ' - b'
Hi Nicolas
On Mon, Nov 11, 2013 at 5:41 PM, Michael Brand
wrote:
> I noticed that to open an Org file with a minimal example generated
> like this slowed down:
>
> #+BEGIN_SRC sh
> #!/bin/sh
> echo '* a'
> for ((i = 0; i < 400; i++)); do
> echo ' - b'
> echo ':PROPERTIES:'
Hi Michael,
Michael Brand writes:
> release_8.2.1-162-gb392750
>
> commit b3927501081b1dab15540591d55f016ed4f9f948
> Author: Nicolas Goaziou
> Date: Sat Nov 2 15:48:36 2013 +0100
>
> Prevent flagging drawers in example blocks
>
> is slow. Today's release_8.2.2-192-ge3033d3
Hi all
I noticed that to open an Org file with a minimal example generated
like this slowed down:
#+BEGIN_SRC sh
#!/bin/sh
echo '* a'
for ((i = 0; i < 400; i++)); do
echo ' - b'
echo ':PROPERTIES:'
echo ':END:'
done
#+END_SRC
Bisecting shows that release_8.2.1-
24 matches
Mail list logo
