[gnugo-devel] Hoshi Josekis

[Evan Berggren Daniel]

In a game on nngs today, ggIII (another copy of Gnu Go, I believe running
3.3.11), got in trouble dealing with a two space high pincer after a low
approach on a hoshi stone.  The following patch adds several joseki
variations for the two space high pincer.  Also attached is a test case
and the sgf.  Do we want to start make regression tests for josekis?  I
think creating a new file joseki.tst and collecting tests about correct
joseki play and choices would be a good idea.

Evan Daniel

loadsgf games/nngs/ggiii-bconwil-200302251140.sgf 10
10 reg_genmove white
#? [Q2]



(;
GM[1]US[Brought to you by No Name Go Server]
CoPyright[
  This game was played on the No Name Go Server
  Permission to reproduce this game is given.]
GN[ggIII-bconwil(B) NNGS]
EV[None]
RE[B+12.5]
PW[ggIII]WR[10k*]
PB[bconwil]BR[10k*]
PC[NNGS Worldwide: nngs.cosmic.org 9696]
DT[2003-02-25]
SZ[19]TM[600]KM[5.5]

;B[pd];W[dp];B[pp];W[dd];B[pj];W[nq];B[kp];W[no];B[on];W[oo];B[po];W[pr]
;B[qq];W[qr];B[hq];W[nd];B[pf];W[jd];B[cn];W[pc];B[qc];W[oc];B[eq];W[dq]
;B[ep];W[do];B[ck];W[eo];B[fo];W[qb];B[rc];W[em];B[ch];W[fn];B[go];W[hm]
;B[bp];W[rb];B[qe];W[rq];B[rp];W[nf];B[mj];W[fc];B[cc];W[cd];B[bd];W[be]
;B[bc];W[db];B[cb];W[dc];B[ce];W[cf];B[de];W[ee];B[bf];W[df];B[ae];W[bg]
;B[ce];W[cg];B[ab];W[dh];B[bh];W[mh];B[di];W[oh];B[qh];W[eh];B[ei];W[kr]
;B[jr];W[kq];B[jq];W[nn];B[om];W[ml];B[km];W[lk];B[ig];W[jl];B[fh];W[hk]
;B[ef];W[jj];B[hd];W[hc];B[id];W[ic];B[je];W[kd];B[ke];W[nk];B[ok];W[nj]
;B[eg];W[rr];B[mq];W[mr];B[nr];W[mp];B[ks];W[oq];B[lq];W[lr];B[lp];W[or]
;B[dr];W[cr];B[er];W[bq];B[cp];W[ap];B[bn];W[bo];B[co];W[cq];B[ao];W[sc]
;B[sd];W[sb];B[re];W[sp];B[ro];W[so];B[sn];W[sq];B[rn];W[fe];B[gf];W[ca]
;B[ba];W[da];B[ag];W[kh];B[ld];W[le];B[me];W[lc];B[lf];W[gj];B[fj];W[fk]
;B[ek];W[el];B[dl];W[lo];B[ko];W[ln];B[kn];W[aq];B[in];W[md];B[le];W[gd]
;B[gi];W[hi];B[hh];W[ii];B[ne];W[oe];B[of];W[ng];B[od];W[mc];B[dg];W[de]
;B[ge];W[oi];B[ph];W[im];B[hn];W[nm];B[gm];W[gn];B[jm];W[kl];B[lm];W[mm]C[
 evand started observation.
]
;B[kg];W[jh];B[lh];W[li];B[lg];W[jg];B[jf];W[ih];B[hf];W[dm];B[cm];W[pq]
;B[qp];W[pi];B[qi];W[oj];B[pk];W[ds];B[es];W[cs];B[og];W[dn];B[be];W[op]
;B[ol];W[ff];B[fg];W[mg];B[nl];W[ll];B[mf];W[mi];B[hl];W[il];B[fm];W[en]
;B[gl];W[ns];B[ls];W[ms];B[js];W[tt];B[tt]
;)



Index: patterns/hoshi.sgf
===================================================================
RCS file: /cvsroot/gnugo/gnugo/patterns/hoshi.sgf,v
retrieving revision 1.17
diff -u -d -r1.17 hoshi.sgf
--- patterns/hoshi.sgf  9 Feb 2003 15:29:10 -0000       1.17
+++ patterns/hoshi.sgf  25 Feb 2003 17:54:14 -0000
@@ -506,7 +506,40 @@
 )
 )

-(;B[jd]C[0];W[tt];B[qh]MA[ij]C[S];W[qc]C[S]MA[hj];B[pc]MA[ij]C[S])
+(;B[jd]C[0];W[tt]
+(;B[qh]MA[ij]C[S];W[qc]C[S]MA[hj];B[pc]MA[ij]C[S])
+
+(;B[pi]MA[ij];W[of]MA[ij]
+(;B[nd]MA[ij];W[rd]MA[ij];B[qc]MA[ij]
+(;W[ri]MA[ij]
+(;B[mf]C[j
+]MA[ij])
+
+(;B[tt];W[mf]C[j
+]MA[ij])
+)
+
+(;W[tt]C[#this is why w must play S11
+];B[pg]MA[ij];W[pf]MA[ij];B[qg]
+MA[ij];W[rg]MA[ij];B[rh]MA[ij];W[rf]MA[ij])
+)
+
+(;B[ne]C[0
+#gg has trouble here; I assume this is reasonable  /evan
+];
+W[rd]MA[ij];B[qc]MA[ij];W[ri]MA[ij])
+)
+)
+
+(;B[pi]MA[ij]
+(;W[qc]MA[ij];B[qd]MA[ij];W[pc]MA[ij];B[od]MA[ij];W[rd]MA[ij];B[re]
+MA[ij];W[rc]MA[ij];B[qe]MA[ij];W[nc]MA[ij];B[pf]C[j
+])
+
+(;W[of]MA[ij];B[nd]MA[ij];W[rd]MA[ij];B[qc]MA[ij];W[ri]MA[ij];B[jc]
+MA[ij];W[qi]MA[ij]C[j
+])
+)
 )

 (;W[pf];B[nd]MA[lh]C[S]