Fixed a small bug in PTN parsing, other stuff

4 files changed, 144 insertions, 66 deletions

diff --git a/extract.sh b/extract.sh
index a48e5ab..77c0ddc 100755
--- a/extract.sh
+++ b/extract.sh
@@ -3,48 +3,47 @@
 db_file=games_anon.db
 
 query() {
-		echo "SELECT $1 FROM games WHERE (size == $2) \
-and ((result == 'R-0') or (result == '0-R'));"
-# and (result != '1-0') and (result != '0-1') and (result != '0-0')\
-# and (result != '1/2-1/2');"
+		query="(size == $1) and (result != '1-0') and (result != '0-1') and (result != '0-0')"
+		selct=""
+		# for player in rabbitboy84 fwwwwibib archvenison Simmon AaaarghBot; do
+		for player in; do
+				selct="$selct(player_black == '$player') or (player_white == '$player') or "
+		done;
+		if [ -n "$selct" ]; then
+				query="$query and (${selct:0:-4})"
+		fi
+		echo "SELECT $2 FROM games WHERE $query;"
 }
 
 extract() {
+		size="$1"
+		things="$2"
 		if [ ! -f "data/$db_file" ]; then
 				wget "https://www.playtak.com/games_anon.db" -O "data/$db_file"
 		fi
-		for size in 5; do
-				echo Extracing games of size "$size"...
-				sqlite3 "data/$db_file" "$(query $1 $size)" | shuf > "data/playtak-$size"
-		done
+		echo Extracing games of size "$size"...
+		sqlite3 "data/$db_file" "$(query $size $things)" | shuf > "data/playtak-$size"
 }
 
 process() {
-		num_games=$1
-		echo -e "Beginning to process data\n"
-		for i in 5; do
-				echo "Size $i..."
-				./pptdb "$i" "data/playtak-$i" > "data/check-$i"
-				tail -n21 "data/check-$i"
-				echo -e "\tStripping overflows and illegal games..."
-				grep -Fvxf "data/check-$i" "data/playtak-$i" > "data/good-playtak-$i"
-				echo -e "\tChoosing $num_games from what remains ..."
-				shuf -n $num_games "data/good-playtak-$i" > "data/smalltak-$i"
-				echo -en "\tGenerating training data... "
-				./pptdb "$i" "data/smalltak-$i" generate
-				echo -e "\tWrote $(wc -l data/training-$i.csv | cut -d\  -f1) samples. Shuffling these..."
-				tail -n+2 "data/training-$i.csv" | shuf > "data/shuf-$i.csv"
-				head -n 1 "data/training-$i.csv" | cat "data/shuf-$i.csv" > "data/smalltrain-$i.csv"
-				rm "data/shuf-$i.csv"
-				mv "data/smalltrain-5.csv" "data/training-5.csv "
-				echo -e "\n\tDone! Sample training data in data/training-$i.csv"
-		done
+		size=$1
+		num_games=$2
+		./pptdb "$size" "data/playtak-$size" > "data/check-$size"
+		tail -n22 "data/check-$size"
+		echo -e "\tStripping overflows and illegal games..."
+		grep -Fvxf "data/check-$size" "data/playtak-$size" > "data/good-playtak-$size"
+		echo -e "\tChoosing $num_games from what remains ..."
+		shuf -n $num_games "data/good-playtak-$size" > "data/smalltak-$size"
+		echo -en "\tGenerating training data... "
+		./pptdb "$size" "data/smalltak-$size" generate
+		echo -e "\n\tDone! Wrote $(wc -l data/training-$size.csv | cut -d\  -f1) samples to data/training-$size.csv."
 }
 
-if [ ! -f "data/playtak-5" ]; then
-		extract notation,result
-fi
 
 make pptdb
 
-process 20000
+for size in 5 6; do
+		extract $size notation,result
+		process $size 20000
+		echo
+done
diff --git a/include/tak.c b/include/tak.c
index 1d29694..bb06a16 100644
--- a/include/tak.c
+++ b/include/tak.c
@@ -319,10 +319,11 @@ check_win(void) {
 		} else {
 			return WIN_ROAD_BLACK;
 		}
-	} else {
+	} else if (rw == WIN_ROAD_WHITE) {
 		return rw;
 	}
 
+	// Do we do a flat count?
 	int8_t total = 0, board_full = 1;
 	for (uint8_t k = 0; k < NUM_SQUARES; k++) {
 		if (COUNT_AT(k) == 0) {
@@ -331,14 +332,14 @@ check_win(void) {
 			total += ((colours[k] & 1) == C_BLACK) ? +1 : -1 ;
 		}
 	}
-
-	// Do we do a flat count?
 	if (black_count == 0 || white_count == 0 || board_full) {
 		// Decide based on count
 		if (total > 0)  return WIN_FLAT_BLACK;
 		else if (total < 0) return WIN_FLAT_WHITE;
 		else return WIN_DRAW;
 	}
+
+	return 0xFF;
 }
 // ===================================================================
 // PTN place parser
@@ -424,7 +425,7 @@ parse_move(const uint8_t board_size, char *ptn,
 	// assume n = 1
 	ptn++;
 	if (*ptn == 0) {
-		*out_steps = picked_up;
+		*out_steps = 1;
 		out_drops[0] = picked_up;
 		return PTN_VALID;
 	}
@@ -556,7 +557,11 @@ do_ptn(char *ptn) {
 		// more conservative here :)
 		if (ply >= board_size) {
 			won = check_win();
-			if (won < 0xFF) return GAME_END;
+			if (won < 0xFF) {
+				// Winning move, but no need to update current colour
+				ply++;
+				return GAME_END;
+			}
 		}
 		// Only step if the game isn't over yet
 		next_ply();
diff --git a/src/ctaklm.c b/src/ctaklm.c
index 9d369f8..3d4310c 100644
--- a/src/ctaklm.c
+++ b/src/ctaklm.c
@@ -190,16 +190,17 @@ end_game(char *line, char *win) {
 	putchar('\n');
 }
 
-static void
+static int
 handle_turn(char *line) {
 	// Track win state
 	uint8_t new_win = (won == 0xFF);
 	switch (do_ptn(line)) {
 		// Errors
-		case PTN_INVALID:  { puts("Invalid PTN."); break; }
-		case ACT_ILLEGAL:  { puts("Illegal action."); break; }
+		case PTN_INVALID:  { puts("Invalid PTN."); return -1; break; }
+		case ACT_ILLEGAL:  { puts("Illegal action."); return -1; break; }
 		case ACT_OVERFLOW: {
 			puts("Move would cause internal overflow, select another.");
+			return -1;
 			break;
 		}
 			// Game has ended
@@ -216,6 +217,7 @@ handle_turn(char *line) {
 				}
 			}
 			puts("Game over, enter `new' to play again.");
+			if (! new_win) return -1;
 			break;
 		}
 			// Valid, append to game log
@@ -227,6 +229,8 @@ handle_turn(char *line) {
 			break;
 		}
 	}
+
+	return 0;
 }
 
 static void
@@ -238,6 +242,59 @@ new_game(uint8_t size) {
 	gamelog[0] = 0;
 }
 
+static int
+load_ptn(const char* fn) {
+	FILE *fh = NULL;
+
+	fh = fopen(fn, "r");
+	if (fh == NULL) return EXIT_FAILURE;
+
+	int space1, space2;
+	enum E_RESULT r;
+
+	ssize_t read;
+	size_t alloc_size;
+	char *line = NULL;
+	while ((read = getline(&line, &alloc_size, fh)) != -1) {
+		if (read && line[0] <= '9' && line[0] >= '0') {
+			// Trim trailing \n
+			line[read-1] = 0;
+			// Find first separator
+			space1 = 0;
+			while (space1 < read && line[space1++] != ' ');
+			// If still on line
+			if (space1 < read) {
+				// Find next separator or end of line, either way mark the split
+				space2 = space1;
+				while (space2 < read && line[space2] != ' ') space2++;
+				line[space2] = 0;
+				// Try the first piece we found
+				r = handle_turn(line+space1);
+				if (r) {
+					printf("Error on: %s\n", line+space1);
+					break;
+				}
+				// If there's a second piece, try it
+				if (space2 + 1 < read) {
+					r = handle_turn(line+space2+1);
+					if (r) {
+						printf("Error on: %s", line+space2+1);
+						break;
+					}
+				} else {
+					break;
+				}
+			}
+		}
+	}
+
+	if (line) free(line);
+	fclose(fh);
+
+
+	return EXIT_SUCCESS;
+}
+
 int
 main(int argc, char **argv) {
 	(void)(argc);
@@ -255,6 +312,12 @@ main(int argc, char **argv) {
 			print_info();
 		} else if (!strncmp(line,"log",3)) {
 			puts(gamelog);
+		} else if (!strncmp(line,"load",4)) {
+			if (strnlen(line,6) >= 6) {
+				load_ptn(line+5);
+			} else {
+				puts("Usage: load <file.ptn>.");
+			}
 		} else if (!strncmp(line,"auto",4)) {
 			if (!strncmp(line,"auto board",10)) {
 				auto_board = ~auto_board;
diff --git a/src/pptdb.c b/src/pptdb.c
index 92155b5..7179ab0 100644
--- a/src/pptdb.c
+++ b/src/pptdb.c
@@ -17,11 +17,11 @@ write_input(void) {
 	/* fprintf(training_fh,"%d,",current_colour == C_BLACK); */
 
 	// Two numbers for flats remaining
-	/* fprintf(training_fh,"%.6f,%.6f,", */
-	/*				(float)(white_count & 127)/max_flats, */
-	/*				(float)(black_count & 127)/max_flats); */
-	// Two data points for each square
-	/* float t; uint16_t mask; */
+	/*
+	 * fprintf(training_fh,"%.6f,%.6f,",
+	 *				(float)(white_count & 127)/max_flats,
+	 *				(float)(black_count & 127)/max_flats);
+	 */
 
 	float val;
 	uint16_t mask = 1;
@@ -48,21 +48,22 @@ write_input(void) {
 		mask <<= 1;
 	}
 
+
 	/*
-	float t;
-	for (int k = 0; k < board_size * board_size; k++) {
-		// stacks encoded as balanced ternary
-		const int h = COUNT_AT(k);
-		t = 0;
-		mask = 1<<h;
-		for (int j = 1; j < h; j++) {
-			t += (colours[k] & mask) ? +1 : -1;
-			t /= 3.0;
-			mask >>=1;
-		}
-		fprintf(training_fh,"%.8f,",2*t);
-	}
-	*/
+	 * float t;
+	 * for (int k = 0; k < board_size * board_size; k++) {
+	 *	// stacks encoded as balanced ternary
+	 *	const int h = COUNT_AT(k);
+	 *	t = 0;
+	 *	mask = 1<<h;
+	 *	for (int j = 1; j < h; j++) {
+	 *		t += (colours[k] & mask) ? +1 : -1;
+	 *		t /= 3.0;
+	 *		mask >>=1;
+	 *	}
+	 *	fprintf(training_fh,"%.8f,",2*t);
+	 * }
+	 */
 
 }
 
@@ -141,7 +142,8 @@ main(int argc, char **argv) {
 	enum E_RESULT r;
 	enum WIN_TYPE win;
 	uint32_t games = 0, overflow=0, illegal = 0;
-	uint32_t road_wins=0, flat_wins=0, road_turns=0, flat_turns=0;
+	uint32_t road_wins=0, flat_wins=0, road_turns=0, flat_turns=0,
+		white_wins = 0, black_wins = 0;
 
 	for (int k = 0; k < 16; k++) heights[k] = 0;
 
@@ -171,10 +173,10 @@ main(int argc, char **argv) {
 		}
 
 		/*
-		for (int k = 0; k < size*size; k++) {
-			fprintf(training_fh,"\"BT %d\",",k);
-		}
-		*/
+		 * for (int k = 0; k < size*size; k++) {
+		 *	fprintf(training_fh,"\"BT %d\",",k);
+		 * }
+		 */
 
 		fputs("\"White win\",\"Black win\"\n",training_fh);
 	} else generate=0;
@@ -206,6 +208,10 @@ main(int argc, char **argv) {
 					road_wins++;
 					road_turns += ply/2+1;
 				}
+				if (win == WIN_FLAT_BLACK || win == WIN_ROAD_BLACK)
+					black_wins++;
+				else if (win == WIN_FLAT_WHITE || win == WIN_ROAD_WHITE)
+					white_wins++;
 			}
 		}
 		games++;
@@ -219,9 +225,14 @@ main(int argc, char **argv) {
 	printf("Read %d games\n",games);
 
 	if (generate==0) {
-		printf("Illegals:  %d\nOverflows: %d\nRoad wins: %d\nFlat wins: %d\n\
-Average turns to road win: %.3f\nAverage turns to flat win: %f\n",
-					 illegal,overflow, road_wins, flat_wins,
+		printf("Illegals:  %d\nOverflows: %d\n\
+Black wins: %.3f%%\n\
+Road wins: %d\nFlat wins: %d\n\
+Average turns to road win: %.3f\n\
+Average turns to flat win: %f\n",
+					 illegal, overflow,
+					 (double)black_wins / (double)(black_wins+white_wins) * 100,
+					 road_wins, flat_wins,
 					 (double)(road_turns)/(double)(road_wins),
 					 (double)(flat_turns)/(double)(flat_wins));
 		for (int k = 2; k < 16; k++) {