PyStarDict v0.4 released

[Serge Matveenko]

Hello all!

I've released PyStarDict v0.4 that have great speed improvements and
new dictionary methods implemented.

Look for it on sf.net download page:
https://sourceforge.net/project/platformdownload.php?group_id=247655

--
Serge Matveenko
mailto:se...@matveenko.ru
http://serge.matveenko.ru/

[cocobear]

On Fri, 2 Jan 2009 02:07:37 +0300
"Serge Matveenko" <se...@matveenko.ru> wrote:

>
> Hello all!
>
> I've released PyStarDict v0.4 that have great speed improvements and
> new dictionary methods implemented.
>

I think we don't need to unpack every word, we can use regex find every
word's position, when look up, we retrive data at the postion of this
word, just at this time we unpack .

[Serge Matveenko]

On Mon, Jan 5, 2009 at 9:01 AM, cocobear <cocob...@gmail.com> wrote:
> On Fri, 2 Jan 2009 02:07:37 +0300
> "Serge Matveenko" <se...@matveenko.ru> wrote:
>
>>
>> Hello all!
>>
>> I've released PyStarDict v0.4 that have great speed improvements and
>> new dictionary methods implemented.
>>
>
> I think we don't need to unpack every word, we can use regex find every
> word's position, when look up, we retrive data at the postion of this
> word, just at this time we unpack .

I feel a bit sceptical about this way. Looks like we cannot know
word's position until we unpack it.

Possibly we could find a specific word by regex but this will take as
much time on every word lookup as we spending for idx load now.

Please run your tests with PyStarDict v0.4. It seems you could get
acceptable results with this version.