https://bugs.freedesktop.org/show_bug.cgi?id=55570
tommy27 changed:
What|Removed |Added
See Also||https://bugs.freedesktop.or
https://bugs.freedesktop.org/show_bug.cgi?id=55570
--- Comment #18 from Michael Meeks ---
a minor code cleanup with no significant improvement in overall speed :-)
--
You are receiving this mail because:
You are the assignee for the bug.
___
Libreoffi
https://bugs.freedesktop.org/show_bug.cgi?id=55570
--- Comment #17 from tommy27 ---
@Caolan McNamara
could you please try to explain me in simple words (I'm not a developer and did
not understand the committ infos) what your latest patch is going to do?
is that an additional improvement to Miche
https://bugs.freedesktop.org/show_bug.cgi?id=55570
--- Comment #15 from tommy27 ---
(In reply to comment #7)
retested with latest daily build with Micheal Meeks fixes.
Version 4.0.0.0.beta1+ (Build ID: 63f68ccd72d9d6dc847f1b6ebf19ee2f8019f54)
>
> I redid the test with current LibO 4.0 daily bui
https://bugs.freedesktop.org/show_bug.cgi?id=55570
Michael Meeks changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
https://bugs.freedesktop.org/show_bug.cgi?id=55570
--- Comment #9 from Michael Meeks ---
Amazingly it seems that a big chunk of our time is spent by the (banal) copy
constructor of the std::set - we seemingly duplicate the initial plain
SvxAutocorrect instance into an SwAutoCorrect sub-class (typ
https://bugs.freedesktop.org/show_bug.cgi?id=55570
--- Comment #8 from Michael Meeks ---
working on this in my idle cycles. It -seems- that a big chunk of the slowness
is down to the sorted std::set in there - which looks un-necessary (to me at
least).
Am re-factoring the SvxAutocorrWordList and
https://bugs.freedesktop.org/show_bug.cgi?id=55570
--- Comment #7 from tommy27 ---
(In reply to comment #0)
>
> LibO 3.6.1
>
> acor_.dat (65K entries)
> 5 seconds
>
> acor_.dat 65K entries + acor_it-IT.dat 55K entries
> 9 seconds
>
> -
https://bugs.freedesktop.org/show_bug.cgi?id=55570
Roman Eisele changed:
What|Removed |Added
Attachment #68055|application/octet-stream|application/zip
mime type|
https://bugs.freedesktop.org/show_bug.cgi?id=55570
--- Comment #5 from tommy27 ---
Created attachment 68055
--> https://bugs.freedesktop.org/attachment.cgi?id=68055&action=edit
2 large autocorrect database (65K entries each)
---
STEPS TO REPRODUCE
--
https://bugs.freedesktop.org/show_bug.cgi?id=55570
--- Comment #4 from Tomaz Vajngerl ---
(In reply to comment #3)
> maybe your "16bit to 32bit" committ is not guilty.
> as suggested by Korrawit I'm going to test a LOdev 3.7 build without your
> patch to see if the performance decline took place
https://bugs.freedesktop.org/show_bug.cgi?id=55570
--- Comment #3 from tommy27 ---
@Michael
sorry for messing the 3.5 MABs and thanks for filing this follow-up bug.
@Tomaz
maybe your "16bit to 32bit" committ is not guilty.
as suggested by Korrawit I'm going to test a LOdev 3.7 build without yo
https://bugs.freedesktop.org/show_bug.cgi?id=55570
--- Comment #2 from Tomaz Vajngerl ---
> I suspect that the longer freeze of LOdev 3.7 is a side effect of the new
> "32-bit autocorrect databases" introduced by this committ:
I don't think this has any impact on performance - the change is onl
https://bugs.freedesktop.org/show_bug.cgi?id=55570
Michael Meeks changed:
What|Removed |Added
Status|UNCONFIRMED |NEW
CC|
14 matches
Mail list logo
