From 27c0562cbbe5355d3f5136ea79d6922df3d47b27 Mon Sep 17 00:00:00 2001
From: Ulya Trofimovich <skvadrik@gmail.com>
Date: Sun, 27 Jan 2019 10:23:23 +0000
Subject: [PATCH] libre2c_posix: extended regex_t structure to hold more
 submatch data.

Added fields:
    - re_nsub: total number of submatch groups, required by POSIX standard

    - pmatch: buffer for submatch results, usually supplied by the user to
      regexec(), but we allow to do the allocation and storage in regex_t.
      This is convenient for users that have hard time managing memory, e.g.
      java bindings to libre2c_posix.

    - regs: buffer for internal use by regexec(), strored in regex_t to
      avoid repeated memory allocation on each call to regexec() with the
      same regex.
---
 re2c/libre2c_posix/regcomp.cc   | 7 +++++++
 re2c/libre2c_posix/regex-impl.h | 5 ++++-
 re2c/libre2c_posix/regexec.cc   | 3 +--
 re2c/libre2c_posix/regfree.cc   | 4 +++-
 re2c/libre2c_posix/test.cpp     | 3 +++
 5 files changed, 18 insertions(+), 4 deletions(-)

diff --git a/re2c/libre2c_posix/regcomp.cc b/re2c/libre2c_posix/regcomp.cc
index 5bd53ff0..356354e4 100644
--- a/re2c/libre2c_posix/regcomp.cc
+++ b/re2c/libre2c_posix/regcomp.cc
@@ -35,6 +35,7 @@ int regcomp(regex_t *preg, const char *pattern, int /* cflags */)
     arv.push_back(ar);
     RESpec re(arv, opt, warn, *preg->rmgr);
 
+    preg->char2class = new size_t[256];
     split_charset(re);
     for (uint32_t i = 1, j = 0; i < re.charset.size(); ++i) {
         for (; j < re.charset[i]; ++j) {
@@ -50,6 +51,12 @@ int regcomp(regex_t *preg, const char *pattern, int /* cflags */)
 
     dfa_t *dfa = new dfa_t(*nfa, opt, "", warn);
 
+//    compact_and_optimize_tags(opt, *dfa);
+
+    DASSERT(dfa->rules.size() == 1);
+    preg->re_nsub = dfa->rules[0].ncap + 1;
+    preg->pmatch = new regmatch_t[preg->re_nsub];
+    preg->regs = new regoff_t[dfa->maxtagver + 1];
     preg->nfa = nfa;
     preg->dfa = dfa;
 
diff --git a/re2c/libre2c_posix/regex-impl.h b/re2c/libre2c_posix/regex-impl.h
index b2b0520c..7fac2352 100644
--- a/re2c/libre2c_posix/regex-impl.h
+++ b/re2c/libre2c_posix/regex-impl.h
@@ -17,10 +17,13 @@ struct RangeMgr;
 
 struct regex_t
 {
+    size_t re_nsub;
     re2c::RangeMgr *rmgr;
     const re2c::nfa_t *nfa;
     const re2c::dfa_t *dfa;
-    size_t char2class[256];
+    regmatch_t *pmatch;
+    regoff_t *regs;
+    size_t *char2class;
 };
 
 #endif // _RE2C_LIB_REGEX_IMPL_
diff --git a/re2c/libre2c_posix/regexec.cc b/re2c/libre2c_posix/regexec.cc
index 6e3f46f9..a20855fa 100644
--- a/re2c/libre2c_posix/regexec.cc
+++ b/re2c/libre2c_posix/regexec.cc
@@ -33,7 +33,7 @@ int regexec(const regex_t *preg, const char *string, size_t nmatch,
 {
     const dfa_t *dfa = preg->dfa;
     int result = REG_NOMATCH;
-    regoff_t *regs = new regoff_t[dfa->maxtagver + 1];
+    regoff_t *regs = preg->regs;
     size_t i = 0;
     const char *p = string, *q = p;
     const dfa_state_t *s, *x = NULL;
@@ -93,6 +93,5 @@ int regexec(const regex_t *preg, const char *string, size_t nmatch,
         }
     }
 
-    delete[] regs;
     return result;
 }
diff --git a/re2c/libre2c_posix/regfree.cc b/re2c/libre2c_posix/regfree.cc
index c4aeb579..2b85c161 100644
--- a/re2c/libre2c_posix/regfree.cc
+++ b/re2c/libre2c_posix/regfree.cc
@@ -9,8 +9,10 @@ using namespace re2c;
 void regfree(regex_t *preg)
 {
     delete preg->rmgr;
-
     delete preg->nfa;
+    delete[] preg->pmatch;
+    delete[] preg->regs;
+    delete[] preg->char2class;
 
     const dfa_t *dfa = preg->dfa;
     delete &dfa->charset;
diff --git a/re2c/libre2c_posix/test.cpp b/re2c/libre2c_posix/test.cpp
index 4f4dbdce..7e29ade3 100644
--- a/re2c/libre2c_posix/test.cpp
+++ b/re2c/libre2c_posix/test.cpp
@@ -1,3 +1,4 @@
+#include <assert.h>
 #include <stdio.h>
 #include <stdlib.h>
 
@@ -39,6 +40,8 @@ static int test(const char *pattern, const char *string
         goto end;
     }
 
+    assert(nmatch == 0 || nmatch == re.re_nsub);
+
     for (uint32_t i = 0; i < nmatch; ++i) {
         regoff_t so = submatch[2 * i];
         regoff_t eo = submatch[2 * i + 1];
-- 
2.40.0