[Scidvspc-users] Nov 2012 update

[Steven <ste...@ya...>]

4.8 seems to have seem pretty solid, so i've been spending less time on Scid vs. PC.
Still managing occasional updates to subversion though. The main new feature is a  "Move Search".


Today I noticed the Tree->Fill Cache With Game feature was broken (and in Scid also).
It's been broken for ages - since the Board Update became more flexible,
and no longer chugged slowly along. I committed a fix to svn, so please test it.
It also cancels FillCacheWithBase if cache gets full, which seems sensible. :)

Other recent changes in svn include 


* Move search feature (eg 'h6 Bxh6')
* Delete key now deletes moves in game/variation *after* the current move
*  Score Graphs are now bargraphs instead of lines
* 'Game Save' dialog now remembers any custom tags you add to a game, making them easy to recall

* Bind statusbar->middle button to 'switch base'.
* Some clock improvements for FICS and Play against UCI engine.
* FICS game offers now show more information, and FICS has a premove feature
* Crosstable: handle games with a year-only date more appropriately.
* Add a few new board textures (from Ed Collins)
* Allow the Name editor to glob '*' for Site, Event and Round fields

Steve






>________________________________
> From: Gregor Cramer <gc...@gm...>
>To: Steven <ste...@ya...> 
>Sent: Monday, 26 November 2012 10:17 PM
>Subject: Re: Scid FillCacheWithGame
> 
>Hi Steve,
>
>> Scid already has a tree cache feature.... so i'm reluctant to add
>> anything like this, which seems to duplicate the idea to me.
>
>It's not exactly the same idea. The tree caching works for specific positions:
>
># ::tree::prime
>#   Primes the tree for this database, filling it with a number of
>#   common opening positions.
>
>But Shiv's idea is to have an index for all positions in database.
>
>> And I found myself a bug ! Scid[vspc]'s tree->FillCacheWithGame  
>> feature is broke since Fulvio's interuptible tree code got added.
>> I've found the place to do a vwait of sorts (patch attached) but is a
>> vwait on a global var (triggered when sc_tree_search completes) the
>> best solution ?
>
>Another idea is to prevent the interruptions in this case, only one additional parameter to "sc_tree search" is required.
>
>> Shiv *may* write me a direct implemntation once his converter is
>> finished.
>
>Why not?
>
>By the way, do you know an internet source for Bughouse openings? I do not mean game collections, I'm interested in structured articles about Bughouse openings.
>
>Best,
>Gregor
>-- 
>Gregor Cramer
>email: <gc...@gm...>
>
>
>
>