perldelta abda9fe0fe75ae824723761c1c98af958f17a41c

[perl5.git] / pp_pack.c

diff --git a/pp_pack.c b/pp_pack.c
index 7e6dc4d..7033649 100644 (file)
--- a/pp_pack.c
+++ b/pp_pack.c
@@ -112,21 +112,17 @@ typedef union {
 /* CROSSCOMPILE and MULTIARCH are going to affect pp_pack() and pp_unpack().
    --jhi Feb 1999 */
 
-#if U16SIZE > SIZE16 || U32SIZE > SIZE32
-#  if BYTEORDER == 0x1234 || BYTEORDER == 0x12345678    /* little-endian */
-#    define OFF16(p)   ((char*)(p))
-#    define OFF32(p)   ((char*)(p))
-#  else
-#    if BYTEORDER == 0x4321 || BYTEORDER == 0x87654321  /* big-endian */
-#      define OFF16(p) ((char*)(p) + (sizeof(U16) - SIZE16))
-#      define OFF32(p) ((char*)(p) + (sizeof(U32) - SIZE32))
-#    else
-       ++++ bad cray byte order
-#    endif
-#  endif
-#else
+#if U16SIZE <= SIZE16 && U32SIZE <= SIZE32
 #  define OFF16(p)     ((char *) (p))
 #  define OFF32(p)     ((char *) (p))
+#elif BYTEORDER == 0x1234 || BYTEORDER == 0x12345678    /* little-endian */
+#  define OFF16(p)     ((char*)(p))
+#  define OFF32(p)     ((char*)(p))
+#elif BYTEORDER == 0x4321 || BYTEORDER == 0x87654321  /* big-endian */
+#  define OFF16(p)     ((char*)(p) + (sizeof(U16) - SIZE16))
+#  define OFF32(p)     ((char*)(p) + (sizeof(U32) - SIZE32))
+#else
+#  error "bad cray byte order"
 #endif
 
 #define PUSH16(utf8, cur, p, needs_swap)                        \
@@ -195,7 +191,7 @@ S_mul128(pTHX_ SV *sv, U8 m)
 
   PERL_ARGS_ASSERT_MUL128;
 
-  if (!strnEQ(s, "0000", 4)) {  /* need to grow sv */
+  if (! memBEGINs(s, len, "0000")) {  /* need to grow sv */
     SV * const tmpNew = newSVpvs("0000000000");
 
     sv_catsv(tmpNew, sv);
@@ -251,12 +247,15 @@ STATIC U8
 utf8_to_byte(pTHX_ const char **s, const char *end, I32 datumtype)
 {
     STRLEN retlen;
-    UV val = utf8n_to_uvchr((U8 *) *s, end-*s, &retlen,
+    UV val;
+
+    if (*s >= end) {
+       goto croak;
+    }
+    val = utf8n_to_uvchr((U8 *) *s, end-*s, &retlen,
                         ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY);
-    /* We try to process malformed UTF-8 as much as possible (preferably with
-       warnings), but these two mean we make no progress in the string and
-       might enter an infinite loop */
-    if (retlen == (STRLEN) -1 || retlen == 0)
+    if (retlen == (STRLEN) -1)
+      croak:
        Perl_croak(aTHX_ "Malformed UTF-8 string in '%c' format in unpack",
                   (int) TYPE_NO_MODIFIERS(datumtype));
     if (val >= 0x100) {
@@ -290,7 +289,7 @@ S_utf8_to_bytes(pTHX_ const char **s, const char *end, const char *buf, int buf_
     for (;buf_len > 0; buf_len--) {
        if (from >= end) return FALSE;
        val = utf8n_to_uvchr((U8 *) from, end-from, &retlen, flags);
-       if (retlen == (STRLEN) -1 || retlen == 0) {
+       if (retlen == (STRLEN) -1) {
            from += UTF8SKIP(from);
            bad |= 1;
        } else from += retlen;
@@ -396,7 +395,7 @@ STMT_START {                                                        \
     STRLEN retlen;                                             \
     if (str >= end) break;                                     \
     val = utf8n_to_uvchr((U8 *) str, end-str, &retlen, utf8_flags);    \
-    if (retlen == (STRLEN) -1 || retlen == 0) {                        \
+    if (retlen == (STRLEN) -1) {                               \
        *cur = '\0';                                            \
        Perl_croak(aTHX_ "Malformed UTF-8 string in pack");     \
     }                                                          \
@@ -1073,9 +1072,14 @@ S_unpack_rec(pTHX_ tempsym_t* symptr, const char *s, const char *strbeg, const c
                /* 'A' strips both nulls and spaces */
                const char *ptr;
                if (utf8 && (symptr->flags & FLAG_WAS_UTF8)) {
-                   for (ptr = s+len-1; ptr >= s; ptr--)
-                       if (*ptr != 0 && !UTF8_IS_CONTINUATION(*ptr) &&
-                           !isSPACE_utf8(ptr)) break;
+                    for (ptr = s+len-1; ptr >= s; ptr--) {
+                        if (   *ptr != 0
+                            && !UTF8_IS_CONTINUATION(*ptr)
+                            && !isSPACE_utf8_safe(ptr, strend))
+                        {
+                            break;
+                        }
+                    }
                    if (ptr >= s) ptr += UTF8SKIP(ptr);
                    else ptr++;
                    if (ptr > s+len)
@@ -1220,7 +1224,7 @@ S_unpack_rec(pTHX_ tempsym_t* symptr, const char *s, const char *strbeg, const c
                    STRLEN retlen;
                    aint = utf8n_to_uvchr((U8 *) s, strend-s, &retlen,
                                 ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY);
-                   if (retlen == (STRLEN) -1 || retlen == 0)
+                   if (retlen == (STRLEN) -1)
                        Perl_croak(aTHX_ "Malformed UTF-8 string in unpack");
                    s += retlen;
                  }
@@ -1243,7 +1247,7 @@ S_unpack_rec(pTHX_ tempsym_t* symptr, const char *s, const char *strbeg, const c
                    STRLEN retlen;
                    const UV val = utf8n_to_uvchr((U8 *) s, strend-s, &retlen,
                                         ckWARN(WARN_UTF8) ? 0 : UTF8_ALLOW_ANY);
-                   if (retlen == (STRLEN) -1 || retlen == 0)
+                   if (retlen == (STRLEN) -1)
                        Perl_croak(aTHX_ "Malformed UTF-8 string in unpack");
                    s += retlen;
                    if (!checksum)
@@ -1305,7 +1309,7 @@ S_unpack_rec(pTHX_ tempsym_t* symptr, const char *s, const char *strbeg, const c
                                                        strend - s,
                                                        &retlen,
                                                        UTF8_ALLOW_DEFAULT));
-                   if (retlen == (STRLEN) -1 || retlen == 0)
+                   if (retlen == (STRLEN) -1)
                        Perl_croak(aTHX_ "Malformed UTF-8 string in unpack");
                    s += retlen;
                }
@@ -1585,7 +1589,8 @@ S_unpack_rec(pTHX_ tempsym_t* symptr, const char *s, const char *strbeg, const c
                    if (++bytes >= sizeof(UV)) {        /* promote to string */
                        const char *t;
 
-                       sv = Perl_newSVpvf(aTHX_ "%.*"UVuf, (int)TYPE_DIGITS(UV), auv);
+                       sv = Perl_newSVpvf(aTHX_ "%.*" UVuf,
+                                                 (int)TYPE_DIGITS(UV), auv);
                        while (s < strend) {
                            ch = SHIFT_BYTE(utf8, s, strend, datumtype);
                            sv = mul128(sv, (U8)(ch & 0x7f));
@@ -2063,9 +2068,9 @@ S_sv_check_infnan(pTHX_ SV *sv, I32 datumtype)
        const I32 c = TYPE_NO_MODIFIERS(datumtype);
        const NV nv = SvNV_nomg(sv);
        if (c == 'w')
-           Perl_croak(aTHX_ "Cannot compress %"NVgf" in pack", nv);
+           Perl_croak(aTHX_ "Cannot compress %" NVgf " in pack", nv);
        else
-           Perl_croak(aTHX_ "Cannot pack %"NVgf" with '%c'", nv, (int) c);
+           Perl_croak(aTHX_ "Cannot pack %" NVgf " with '%c'", nv, (int) c);
     }
     return sv;
 }
@@ -2355,7 +2360,7 @@ S_pack_rec(pTHX_ SV *cat, tempsym_t* symptr, SV **beglist, SV **endlist )
                if (!S_utf8_to_bytes(aTHX_ &aptr, end, cur, fromlen,
                                  datumtype | TYPE_IS_PACK))
                    Perl_croak(aTHX_ "panic: predicted utf8 length not available, "
-                              "for '%c', aptr=%p end=%p cur=%p, fromlen=%"UVuf,
+                              "for '%c', aptr=%p end=%p cur=%p, fromlen=%" UVuf,
                               (int)datumtype, aptr, end, cur, (UV)fromlen);
                cur += fromlen;
                len -= fromlen;
@@ -2581,17 +2586,14 @@ S_pack_rec(pTHX_ SV *cat, tempsym_t* symptr, SV **beglist, SV **endlist )
                if (in_bytes) auv = auv % 0x100;
                if (utf8) {
                  W_utf8:
-                   if (cur > end) {
+                   if (cur >= end) {
                        *cur = '\0';
                        SvCUR_set(cat, cur - start);
 
                        GROWING(0, cat, start, cur, len+UTF8_MAXLEN);
                        end = start+SvLEN(cat)-UTF8_MAXLEN;
                    }
-                   cur = (char *) uvchr_to_utf8_flags((U8 *) cur,
-                                                      auv,
-                                                      warn_utf8 ?
-                                                      0 : UNICODE_ALLOW_ANY);
+                   cur = (char *) uvchr_to_utf8_flags((U8 *) cur, auv, 0);
                } else {
                    if (auv >= 0x100) {
                        if (!SvUTF8(cat)) {
@@ -2642,9 +2644,7 @@ S_pack_rec(pTHX_ SV *cat, tempsym_t* symptr, SV **beglist, SV **endlist )
                auv = SvUV_no_inf(fromstr, datumtype);
                if (utf8) {
                    U8 buffer[UTF8_MAXLEN], *endb;
-                   endb = uvchr_to_utf8_flags(buffer, UNI_TO_NATIVE(auv),
-                                              warn_utf8 ?
-                                              0 : UNICODE_ALLOW_ANY);
+                   endb = uvchr_to_utf8_flags(buffer, UNI_TO_NATIVE(auv), 0);
                    if (cur+(endb-buffer)*UTF8_EXPAND >= end) {
                        *cur = '\0';
                        SvCUR_set(cat, cur - start);
@@ -2660,9 +2660,9 @@ S_pack_rec(pTHX_ SV *cat, tempsym_t* symptr, SV **beglist, SV **endlist )
                        GROWING(0, cat, start, cur, len+UTF8_MAXLEN);
                        end = start+SvLEN(cat)-UTF8_MAXLEN;
                    }
-                   cur = (char *) uvchr_to_utf8_flags((U8 *) cur, UNI_TO_NATIVE(auv),
-                                                      warn_utf8 ?
-                                                      0 : UNICODE_ALLOW_ANY);
+                   cur = (char *) uvchr_to_utf8_flags((U8 *) cur,
+                                                       UNI_TO_NATIVE(auv),
+                                                      0);
                }
            }
            break;