From 688a62c8263d00a2d650f5d2dfd124dd2e22be96 Mon Sep 17 00:00:00 2001
From: Kyryl Melekhin <k.melekhin@gmail.com>
Date: Sat, 23 Apr 2022 17:03:30 +0000
Subject: [PATCH] speed it up, test greediness

---
 pike.c  | 61 +++++++++++++++++++++++++++------------------------------
 test.sh | 39 ++++++++++++++++++++++++++++++++++++
 2 files changed, 68 insertions(+), 32 deletions(-)

diff --git a/pike.c b/pike.c
index bf0e32e..17e32d9 100644
--- a/pike.c
+++ b/pike.c
@@ -446,8 +446,7 @@ if (si) { \
 
 #define deccheck(nn) { decref(nsub) rec_check(nn) continue; } \
 
-#define onclist(nn)
-#define onnlist(nn) \
+#define onlist(nn) \
 if (sdense[spc] < sparsesz) \
 	if (sdense[sdense[spc] * 2] == (unsigned int)spc) \
 		deccheck(nn) \
@@ -467,27 +466,21 @@ subs[si++] = nsub; \
 goto next##nn; \
 
 #define saveclist() \
-newsub(memcpy(s1->sub, nsub->sub, osubp);, \
-memcpy(s1->sub, nsub->sub, osubp / 2);) \
-
-#define savenlist() \
-newsub(/*nop*/, /*nop*/) \
-memcpy(s1->sub, nsub->sub, osubp); \
-
-#define instclist(nn) \
-else if (spc == BOL) { \
-	if (_sp != s) { \
-		if (!si && !clistidx) \
-			return 0; \
-		deccheck(nn) \
-	} \
-	npc++; goto rec##nn; \
+if (npc[1] > nsubp / 2 && nsub->ref > 1) { \
+	nsub->ref--; \
+	newsub(memcpy(s1->sub, nsub->sub, osubp);, \
+	memcpy(s1->sub, nsub->sub, osubp / 2);) \
+	nsub = s1; \
+	nsub->ref = 1; \
 } \
 
-#define instnlist(nn) \
-else if (spc == JMP) { \
-	npc += 2 + npc[1]; \
-	goto rec##nn; \
+#define savenlist() \
+if (nsub->ref > 1) { \
+	nsub->ref--; \
+	newsub(/*nop*/, /*nop*/) \
+	memcpy(s1->sub, nsub->sub, osubp); \
+	nsub = s1; \
+	nsub->ref = 1; \
 } \
 
 #define clistmatch()
@@ -513,17 +506,12 @@ if ((unsigned int)spc < WBEG) { \
 } \
 next##nn: \
 if (spc > JMP) { \
-	on##list(nn) \
+	onlist(nn) \
 	npc += 2; \
 	pcs[si] = npc + npc[-1]; \
 	fastrec(nn, list, listidx) \
 } else if (spc == SAVE) { \
-	if (nsub->ref > 1) { \
-		nsub->ref--; \
-		save##list() \
-		nsub = s1; \
-		nsub->ref = 1; \
-	} \
+	save##list() \
 	nsub->sub[npc[1]] = _sp; \
 	npc += 2; \
 	goto rec##nn; \
@@ -534,7 +522,7 @@ if (spc > JMP) { \
 	npc++; goto rec##nn; \
 } else if (spc < 0) { \
 	spc = -spc; \
-	on##list(nn) \
+	onlist(nn) \
 	npc += 2; \
 	pcs[si] = npc; \
 	npc += npc[-1]; \
@@ -547,8 +535,17 @@ if (spc > JMP) { \
 	if (*_sp) \
 		deccheck(nn) \
 	npc++; goto rec##nn; \
-} inst##list(nn) \
-deccheck(nn) \
+} else if (spc == JMP) { \
+	npc += 2 + npc[1]; \
+	goto rec##nn; \
+} else { \
+	if (_sp != s) { \
+		if (!si && !clistidx) \
+			return 0; \
+		deccheck(nn) \
+	} \
+	npc++; goto rec##nn; \
+} \
 
 #define swaplist() \
 tmp = clist; \
@@ -567,7 +564,7 @@ int re_pikevm(rcode *prog, const char *s, const char **subp, int nsubp)
 	int *insts = prog->insts;
 	int *pcs[prog->splits];
 	rsub *subs[prog->splits];
-	unsigned int sdense[prog->sparsesz], sparsesz;
+	unsigned int sdense[prog->sparsesz], sparsesz = 0;
 	rsub *nsub, *s1, *matched = NULL, *freesub = NULL;
 	rthread _clist[prog->len], _nlist[prog->len];
 	rthread *clist = _clist, *nlist = _nlist, *tmp;
diff --git a/test.sh b/test.sh
index d3246ed..b4ff3d0 100755
--- a/test.sh
+++ b/test.sh
@@ -172,6 +172,19 @@ aaaaa(aa)aa(aa(a)a)?aa
 ([0-9])+.(.*)
 (([0-9])+)(.)(.*)
 (abc|sjd|qwq(hs|qw|oo)|(ty|xx|pp)we)
+(a?)?
+(a?)?
+(a?)??
+(a?)(a?)(a?a?)aaa(a?)|(b?)
+(a?)(a?)(a?a?)aaa(a?)|(b?)
+(a*)??
+(a*)|(b*)|(c*)
+(a?|a?)|(a*)?
+(a?|a?)a|(a*)?
+(a?|a?)d|(a*)?
+((a*b*c*)|(a*c*b*))+?
+((a*b*c*)|(a*c*b*))*?
+((a*b*c*)|(a*c*b*))+
 "
 input="\
 abcdef
@@ -345,6 +358,19 @@ h:98:   :3234utt;strokeliin:miter;stroke-mirlimit:10;stroke-dasharray:none;strok
 650-253-000123434-45551221
 650-253-000123434-455512213224hsaqer
 ppwe
+m
+a
+aa
+aa
+aaaaaaaaaaaaaa
+a
+bbbbbbbbb
+aaa
+aaa
+aaa
+bbb
+bbb
+bbb
 "
 expect="\
 (0,3)
@@ -518,6 +544,19 @@ expect="\
 (0,26)(2,3)(4,26)
 (0,36)(0,3)(2,3)(3,4)(4,36)
 (0,4)(0,4)(?,?)(0,2)
+(0,0)(0,0)
+(0,1)(0,1)
+(0,0)(?,?)
+(0,0)(?,?)(?,?)(?,?)(?,?)(0,0)
+(0,8)(0,1)(1,2)(2,4)(7,8)(?,?)
+(0,0)(?,?)
+(0,0)(0,0)(?,?)(?,?)
+(0,1)(0,1)(?,?)
+(0,2)(0,1)(?,?)
+(0,3)(?,?)(0,3)
+(0,3)(0,3)(0,3)(?,?)
+(0,0)(?,?)(?,?)(?,?)
+(0,3)(0,3)(0,3)(?,?)
 (0,0)
 "