From: Eric Cheymol <eric.cheymol <at> laposte <dot> net>
Date: 01 sep 2007
Subject: Re: Definition of PSN notation


(I aready have sent this mail to the list on the 27 august, but=20=20
without any success. So I send it again with a new sunscription email)

Hi Reijer,

> Related to work on an English version of Shogidokoro I have been=20=20
> looking into PSN notation. I was surprised that I couldn't find a=20=20
> proper definition of this notation anywhere despite the fact that a=20=
=20
> number of programs are using this notation (including my own=20=20
> program!). Anyone who has a definition file?
>
>

Here is a mail sent to shogi-l in 2002 including a copy of a first=20=20
draft (written in 1997) for the PSN notation :

http://www.shogi.net/shogi-l/Archive/2002/Nsep29-03.txt


> I am particularly interested in the following:
>
> * The possible headers. On my website, I am only using [Black ""],=20=20
> [White ""], [Event ""] and [Date ""], but I know there are many=20=20
> more like [Opening ""], [Site ""] and shogi-specific ones like=20=20
> [Handicap ""].
> * How is the handicap header defined? For example, is two piece=20=20
> handicap described as [Handicap "Two piece"], [Handicap "Rook,=20=20
> bishop"] or even differently?

I have added some comments in this mail :

http://www.shogi.net/shogi-l/Archive/2007/Njan30-02.txt

or details from other authors :

http://www.shogi.net/shogi-l/Archive/1997/Nsep01-04.txt
http://www.shogi.net/shogi-l/Archive/1998/Nnov06-02.txt


> Any help is appreciated.

Here is the headers I use. The following lines are a copy and paste=20=20
lines from MacShogi, sorry if I don't translate them !) :

	"Sente",		// GAM_SENTE		Nom du joueur Sente
	"SenteGrade",		// GAM_SENTEGRADE	Grade du joueur Sente
	"Gote",			// GAM_GOTE		Nom du joueur Gote
	"GoteGrade",		// GAM_GOTEGRADE	Grade du joueur Gote
	"Date",			// GAM_DATE		Date de la partie
	"Event",		// GAM_EVENT		Nom du tournoi
	"Joseki",		// GAM_JOSEKI		Joseki utilis=E9 (Tsume, Hishi, Yokofudori ...)
	"Fen",			// GAM_FEN		Position initiale (pour une fin de partie, un=20=20
tsume ...)
	"Round",		// GAM_ROUND		Ronde de la partie
	"Result",		// GAM_RESULT		R=E9sultat final
	"Moves",		// GAM_MOVES		Nombre de coups
	"Handicap",		// GAM_HANDICAP		Nature du handicap
	"Comment",		// GAM_COMMENT		Commentateur

	// Fusionner
	"Venue",		// GAM_VENUE		Lieu de la partie
	"Site",			// GAM_SITE		Lieu de la partie
	"Country",		// GAM_COUNTRY		Pays du lieu de la partie

And for compatibilities with GSDB :

	// Ent=EAtes supprim=E9s
	"Source",		// GAM_SOURCE		Origine de la feuille de partie
	"Name",			// GAM_NAME		Nom du joueur aui a saisi la partie
	"Email",		// GAM_EMAIL		Email du joueur qui a saisi la partie
	"Proam",		// GAM_PROAM		Partie entre amateurs ou professionnels

	// Equivalences
	"Opening",		// GAM_OPENING		Joseki
	"Black",		// GAM_BLACK		Sente
	"White",		// GAM_WHITE		Gote
	"Black_grade",		// GAM_BLACKGRADE	SenteGrade
	"White_grade",		// GAM_WHITEGRADE	GoteGrade


In the previous list, Fen stands for the FEN notation.

Handicaps in the "Handicap" header :

	"Even",
	"Sente",
	"Lance",
	"Bishop",
	"Rook",
	"Rook+Lance",
	"Rook+Bishop",
	"Four piece",
	"Six Piece",
	"Eight Piece",


> * How should the time be added to the moves?

I add the time with brackets in minutes like this : {1} for one=20=20
minute of thinking. This sequence is handled like a special comment.=20=20
I have prefered to enclose the thinking time as simply as possible in=20=
=20
the scoresheet, without special tags, but other ideas are welcomed,

Else :

For the encoding, some players prefer to use the 7 bit ascii encoding=20=
=20
only instead of allowing exotic characters. I know this has some=20=20
drawbacks for some operating system but personnaly I find ennoying to=20=
=20
remove Japanese comments for this reason only. Ok, I don't read=20=20
Japanese at all but some other players do :-)) And this can motivate=20=20
some players to learn Japanese, or even motivate some Japanese to use=20=
=20
a western shogi program to improve in shogi !

For the moves encoding, I already have written about long move=20=20
notation, short move ... I don't know if this is interesting for you,=20=
=20
all have almost been written about it. The short notation allows=20=20
compacting large gamescores. For the complete collection, I save Mb=20=20
in the whole file. And this adds a step for the engine of importing=20=20
Japanese kifu files.

I find variations very useful as well : with parenthesis, any depth=20=20
allowed.

Cheers,

Eric Cheymol

----------------
Get the European shogi archives at http://eric.macshogi.com/
The first Mac OS X Shogi Database Program at http://www.macshogi.com/


--^----------------------------------------------------------------
This email was sent to: =
shogi-l <at> shogi <dot> net

EASY UNSUBSCRIBE click here: http://topica.com/u/?a2i6Ys.=
aCGIy8.=
c2hvZ2kt
Or send an email to: shogi-unsubscribe <at> topica <dot> com

For Topica's complete suite of email marketing solutions visit:
http://www.topica.com/?p=3DTEXFOOTER
--^----------------------------------------------------------------