Re: [libreoffice-users] 98, 000+ word list British dictionary, hyphen, thesaurus dictionary

Hi :)
On the types of machines you sometimes use a performance hit would be very noticeable.  So, if 
there is no hit then the longer lists make a lot of sense.  The 102k difference is surely not a 
huge worry to anyone these days?  I had assumed that with a list twice as long the spell check 
might take about twice as long too.  It's good to hear i was so wrong :)

Many thanks for your hard work at all of this.  I'm sure a lot of people appreiate the results :)
Thanks and regards from
Tom :)


--- On Tue, 8/11/11, webmaster for Kracked Press Productions <webmaster@krackedpress.com> wrote:

From: webmaster for Kracked Press Productions <webmaster@krackedpress.com>
Subject: Re: [libreoffice-users] 98, 000+ word list British dictionary, hyphen, thesaurus dictionary
To: users@global.libreoffice.org
Date: Tuesday, 8 November, 2011, 13:43


I have not seen any "performance hits" with my use of either the 98K or the 217K word list 
dictionaries.  The actual time would difference would be really small.  I have not tried using a 
390K or 638K word lists though.  For a real large document that I used for testing sometimes, you 
should not see any difference for spell checker lookup.

I do not know what type of searches are used in LO to search the word lists for the word lookup, 
but for the old 80286 systems that I use to do this type of programming with  these 50K vs 638K 
list searches would take about a 1/10 of a second difference for the same 100,000 word document.  
This was what I generally got for my word list searches back in the 80's with the programming 
samples I wrote.  The professor wanted a timer included in the search software so he could see how 
efficient your code was.  College is where I got interested in dictionary and word list searches 
and functions for spell checking.  To re-learn C++ after my second stroke, I write a program to 
create word lists from e-book text and compare them with the current lists to see what new words I 
could find.

The fact the the original .dic files had control codes after each word requires the system to do 
the work to do the conversions and then use those options in its searching.  So having lists that 
do not need those control codes may make spelling searched faster.




On 11/08/2011 03:13 AM, Mark Stanton wrote:
> I'm always interested in "the most comprehensive".
> Presumably there's a performance hit related to the size of
> dictionary?
>
> Mark Stanton
> One small step for mankind...


-- For unsubscribe instructions e-mail to: users+help@global.libreoffice.org
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://listarchives.libreoffice.org/global/users/
All messages sent to this list will be publicly archived and cannot be deleted


-- 
For unsubscribe instructions e-mail to: users+help@global.libreoffice.org
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://listarchives.libreoffice.org/global/users/
All messages sent to this list will be publicly archived and cannot be deleted