BurntSushi · darkyzhou · Nov 2, 2024
diff --git a/pcre2-sys/src/bindings.rs b/pcre2-sys/src/bindings.rs
@@ -1,8 +1,8 @@
-/* automatically generated by rust-bindgen 0.69.4 */
+/* automatically generated by rust-bindgen 0.70.1 */
 
 pub const PCRE2_MAJOR: u32 = 10;
-pub const PCRE2_MINOR: u32 = 43;
-pub const PCRE2_DATE: u32 = 2006;
+pub const PCRE2_MINOR: u32 = 44;
+pub const PCRE2_DATE: u32 = 2011;
 pub const PCRE2_ANCHORED: u32 = 2147483648;
 pub const PCRE2_NO_UTF_CHECK: u32 = 1073741824;
 pub const PCRE2_ENDANCHORED: u32 = 536870912;
@@ -419,6 +419,12 @@ extern "C" {
         arg2: usize,
     ) -> ::libc::c_int;
 }
+extern "C" {
+    pub fn pcre2_set_max_pattern_compiled_length_8(
+        arg1: *mut pcre2_compile_context_8,
+        arg2: usize,
+    ) -> ::libc::c_int;
+}
 extern "C" {
     pub fn pcre2_set_max_varlookbehind_8(
         arg1: *mut pcre2_compile_context_8,
@@ -916,6 +922,12 @@ extern "C" {
         arg2: usize,
     ) -> ::libc::c_int;
 }
+extern "C" {
+    pub fn pcre2_set_max_pattern_compiled_length_16(
+        arg1: *mut pcre2_compile_context_16,
+        arg2: usize,
+    ) -> ::libc::c_int;
+}
 extern "C" {
     pub fn pcre2_set_max_varlookbehind_16(
         arg1: *mut pcre2_compile_context_16,
@@ -1419,6 +1431,12 @@ extern "C" {
         arg2: usize,
     ) -> ::libc::c_int;
 }
+extern "C" {
+    pub fn pcre2_set_max_pattern_compiled_length_32(
+        arg1: *mut pcre2_compile_context_32,
+        arg2: usize,
+    ) -> ::libc::c_int;
+}
 extern "C" {
     pub fn pcre2_set_max_varlookbehind_32(
         arg1: *mut pcre2_compile_context_32,

diff --git a/pcre2-sys/update-pcre2 b/pcre2-sys/update-pcre2
@@ -11,7 +11,7 @@ pcre2_sys_dir="$(dirname "$0")"
 cd "$pcre2_sys_dir"
 
 dest="upstream"
-version="10.43"
+version="10.44"
 dir="pcre2-$version"
 archive="$dir.tar.gz"
 url="https://github.com/PCRE2Project/pcre2/releases/download/pcre2-$version/$archive"

diff --git a/pcre2-sys/upstream/include/config.h b/pcre2-sys/upstream/include/config.h
@@ -215,7 +215,7 @@ sure both macros are undefined; an emulation function will then be used. */
    Care must be taken if it is increased, because it guards against integer
    overflow caused by enormously large patterns. */
 #ifndef MAX_NAME_SIZE
-#define MAX_NAME_SIZE 32
+#define MAX_NAME_SIZE 128
 #endif
 
 /* The value of MAX_VARLOOKBEHIND specifies the default maximum length, in
@@ -245,7 +245,7 @@ sure both macros are undefined; an emulation function will then be used. */
 #define PACKAGE_NAME "PCRE2"
 
 /* Define to the full name and version of this package. */
-#define PACKAGE_STRING "PCRE2 10.43"
+#define PACKAGE_STRING "PCRE2 10.44"
 
 /* Define to the one symbol short name of this package. */
 #define PACKAGE_TARNAME "pcre2"
@@ -254,7 +254,7 @@ sure both macros are undefined; an emulation function will then be used. */
 #define PACKAGE_URL ""
 
 /* Define to the version of this package. */
-#define PACKAGE_VERSION "10.43"
+#define PACKAGE_VERSION "10.44"
 
 /* The value of PARENS_NEST_LIMIT specifies the maximum depth of nested
    parentheses (of any kind) in a pattern. This limits the amount of system
@@ -458,7 +458,7 @@ sure both macros are undefined; an emulation function will then be used. */
 #endif
 
 /* Version number of package */
-#define VERSION "10.43"
+#define VERSION "10.44"
 
 /* Number of bits in a file offset, on hosts where this is settable. */
 /* #undef _FILE_OFFSET_BITS */

diff --git a/pcre2-sys/upstream/include/pcre2.h b/pcre2-sys/upstream/include/pcre2.h
@@ -42,9 +42,9 @@ POSSIBILITY OF SUCH DAMAGE.
 /* The current PCRE version information. */
 
 #define PCRE2_MAJOR           10
-#define PCRE2_MINOR           43
+#define PCRE2_MINOR           44
 #define PCRE2_PRERELEASE      
-#define PCRE2_DATE            2024-02-16
+#define PCRE2_DATE            2024-06-07
 
 /* When an application links to a PCRE DLL in Windows, the symbols that are
 imported have to be identified as such. When building PCRE2, the appropriate
@@ -603,6 +603,8 @@ PCRE2_EXP_DECL int PCRE2_CALL_CONVENTION \
   pcre2_set_compile_extra_options(pcre2_compile_context *, uint32_t); \
 PCRE2_EXP_DECL int PCRE2_CALL_CONVENTION \
   pcre2_set_max_pattern_length(pcre2_compile_context *, PCRE2_SIZE); \
+PCRE2_EXP_DECL int PCRE2_CALL_CONVENTION \
+  pcre2_set_max_pattern_compiled_length(pcre2_compile_context *, PCRE2_SIZE); \
 PCRE2_EXP_DECL int PCRE2_CALL_CONVENTION \
   pcre2_set_max_varlookbehind(pcre2_compile_context *, uint32_t); \
 PCRE2_EXP_DECL int PCRE2_CALL_CONVENTION \
@@ -901,6 +903,7 @@ pcre2_compile are called by application code. */
 #define pcre2_set_match_limit                 PCRE2_SUFFIX(pcre2_set_match_limit_)
 #define pcre2_set_max_varlookbehind           PCRE2_SUFFIX(pcre2_set_max_varlookbehind_)
 #define pcre2_set_max_pattern_length          PCRE2_SUFFIX(pcre2_set_max_pattern_length_)
+#define pcre2_set_max_pattern_compiled_length PCRE2_SUFFIX(pcre2_set_max_pattern_compiled_length_)
 #define pcre2_set_newline                     PCRE2_SUFFIX(pcre2_set_newline_)
 #define pcre2_set_parens_nest_limit           PCRE2_SUFFIX(pcre2_set_parens_nest_limit_)
 #define pcre2_set_offset_limit                PCRE2_SUFFIX(pcre2_set_offset_limit_)

diff --git a/pcre2-sys/upstream/src/pcre2_compile.c b/pcre2-sys/upstream/src/pcre2_compile.c
@@ -7,7 +7,7 @@ and semantics are as close as possible to those of the Perl 5 language.
 
                        Written by Philip Hazel
      Original API code Copyright (c) 1997-2012 University of Cambridge
-          New API code Copyright (c) 2016-2023 University of Cambridge
+          New API code Copyright (c) 2016-2024 University of Cambridge
 
 -----------------------------------------------------------------------------
 Redistribution and use in source and binary forms, with or without
@@ -808,7 +808,8 @@ enum { ERR0 = COMPILE_ERROR_BASE,
        ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69, ERR70,
        ERR71, ERR72, ERR73, ERR74, ERR75, ERR76, ERR77, ERR78, ERR79, ERR80,
        ERR81, ERR82, ERR83, ERR84, ERR85, ERR86, ERR87, ERR88, ERR89, ERR90,
-       ERR91, ERR92, ERR93, ERR94, ERR95, ERR96, ERR97, ERR98, ERR99, ERR100 };
+       ERR91, ERR92, ERR93, ERR94, ERR95, ERR96, ERR97, ERR98, ERR99, ERR100,
+       ERR101 };
 
 /* This is a table of start-of-pattern options such as (*UTF) and settings such
 as (*LIMIT_MATCH=nnnn) and (*CRLF). For completeness and backward
@@ -7549,7 +7550,8 @@ for (;; pptr++)
             if (lengthptr != NULL)
               {
               PCRE2_SIZE delta;
-              if (PRIV(ckd_smul)(&delta, repeat_min - 1, length_prevgroup) ||
+              if (PRIV(ckd_smul)(&delta, repeat_min - 1,
+                                 (int)length_prevgroup) ||
                   OFLOW_MAX - *lengthptr < delta)
                 {
                 *errorcodeptr = ERR20;
@@ -7599,7 +7601,7 @@ for (;; pptr++)
             {
             PCRE2_SIZE delta;
             if (PRIV(ckd_smul)(&delta, repeat_max,
-                               length_prevgroup + 1 + 2 + 2*LINK_SIZE) ||
+                               (int)length_prevgroup + 1 + 2 + 2*LINK_SIZE) ||
                 OFLOW_MAX + (2 + 2*LINK_SIZE) - *lengthptr < delta)
               {
               *errorcodeptr = ERR20;
@@ -9908,7 +9910,7 @@ do
   *bptr |= branchlength;  /* branchlength never more than 65535 */
   bptr = *pptrptr;
   }
-while (*bptr == META_ALT);
+while (META_CODE(*bptr) == META_ALT);
 
 /* If any branch is of variable length, the whole lookbehind is of variable
 length. If the maximum length of any branch exceeds the maximum for variable
@@ -10601,14 +10603,21 @@ if (length > MAX_PATTERN_SIZE)
   goto HAD_CB_ERROR;
   }
 
-/* Compute the size of, and then get and initialize, the data block for storing
-the compiled pattern and names table. Integer overflow should no longer be
-possible because nowadays we limit the maximum value of cb.names_found and
-cb.name_entry_size. */
+/* Compute the size of, then, if not too large, get and initialize the data
+block for storing the compiled pattern and names table. Integer overflow should
+no longer be possible because nowadays we limit the maximum value of
+cb.names_found and cb.name_entry_size. */
 
 re_blocksize = sizeof(pcre2_real_code) +
   CU2BYTES(length +
   (PCRE2_SIZE)cb.names_found * (PCRE2_SIZE)cb.name_entry_size);
+
+if (re_blocksize > ccontext->max_pattern_compiled_length)
+  {
+  errorcode = ERR101;
+  goto HAD_CB_ERROR;
+  }
+
 re = (pcre2_real_code *)
   ccontext->memctl.malloc(re_blocksize, ccontext->memctl.memory_data);
 if (re == NULL)

diff --git a/pcre2-sys/upstream/src/pcre2_context.c b/pcre2-sys/upstream/src/pcre2_context.c
@@ -7,7 +7,7 @@ and semantics are as close as possible to those of the Perl 5 language.
 
                        Written by Philip Hazel
      Original API code Copyright (c) 1997-2012 University of Cambridge
-          New API code Copyright (c) 2016-2023 University of Cambridge
+          New API code Copyright (c) 2016-2024 University of Cambridge
 
 -----------------------------------------------------------------------------
 Redistribution and use in source and binary forms, with or without
@@ -136,6 +136,7 @@ const pcre2_compile_context PRIV(default_compile_context) = {
   NULL,                                      /* Stack guard data */
   PRIV(default_tables),                      /* Character tables */
   PCRE2_UNSET,                               /* Max pattern length */
+  PCRE2_UNSET,                               /* Max pattern compiled length */
   BSR_DEFAULT,                               /* Backslash R default */
   NEWLINE_DEFAULT,                           /* Newline convention */
   PARENS_NEST_LIMIT,                         /* As it says */
@@ -352,6 +353,13 @@ ccontext->max_pattern_length = length;
 return 0;
 }
 
+PCRE2_EXP_DEFN int PCRE2_CALL_CONVENTION
+pcre2_set_max_pattern_compiled_length(pcre2_compile_context *ccontext, PCRE2_SIZE length)
+{
+ccontext->max_pattern_compiled_length = length;
+return 0;
+}
+
 PCRE2_EXP_DEFN int PCRE2_CALL_CONVENTION
 pcre2_set_newline(pcre2_compile_context *ccontext, uint32_t newline)
 {

diff --git a/pcre2-sys/upstream/src/pcre2_error.c b/pcre2-sys/upstream/src/pcre2_error.c
@@ -7,7 +7,7 @@ and semantics are as close as possible to those of the Perl 5 language.
 
                        Written by Philip Hazel
      Original API code Copyright (c) 1997-2012 University of Cambridge
-          New API code Copyright (c) 2016-2023 University of Cambridge
+          New API code Copyright (c) 2016-2024 University of Cambridge
 
 -----------------------------------------------------------------------------
 Redistribution and use in source and binary forms, with or without
@@ -189,6 +189,7 @@ static const unsigned char compile_error_texts[] =
   "\\K is not allowed in lookarounds (but see PCRE2_EXTRA_ALLOW_LOOKAROUND_BSK)\0"
   /* 100 */
   "branch too long in variable-length lookbehind assertion\0"
+  "compiled pattern would be longer than the limit set by the application\0"
   ;
 
 /* Match-time and UTF error texts are in the same format. */

diff --git a/pcre2-sys/upstream/src/pcre2_extuni.c b/pcre2-sys/upstream/src/pcre2_extuni.c
@@ -7,7 +7,7 @@ and semantics are as close as possible to those of the Perl 5 language.
 
                        Written by Philip Hazel
      Original API code Copyright (c) 1997-2012 University of Cambridge
-          New API code Copyright (c) 2016-2021 University of Cambridge
+          New API code Copyright (c) 2016-2024 University of Cambridge
 
 -----------------------------------------------------------------------------
 Redistribution and use in source and binary forms, with or without
@@ -75,7 +75,11 @@ return NULL;
 *      Match an extended grapheme sequence       *
 *************************************************/
 
-/*
+/* NOTE: The logic contained in this function is replicated in three special-
+purpose functions in the pcre2_jit_compile.c module. If the logic below is
+changed, they must be kept in step so that the interpreter and the JIT have the
+same behaviour.
+
 Arguments:
   c              the first character
   eptr           pointer to next character
@@ -92,6 +96,7 @@ PCRE2_SPTR
 PRIV(extuni)(uint32_t c, PCRE2_SPTR eptr, PCRE2_SPTR start_subject,
   PCRE2_SPTR end_subject, BOOL utf, int *xcount)
 {
+BOOL was_ep_ZWJ = FALSE;
 int lgb = UCD_GRAPHBREAK(c);
 
 while (eptr < end_subject)
@@ -102,6 +107,12 @@ while (eptr < end_subject)
   rgb = UCD_GRAPHBREAK(c);
   if ((PRIV(ucp_gbtable)[lgb] & (1u << rgb)) == 0) break;
 
+  /* ZWJ followed by Extended Pictographic is allowed only if the ZWJ was
+  preceded by Extended Pictographic. */
+
+  if (lgb == ucp_gbZWJ && rgb == ucp_gbExtended_Pictographic && !was_ep_ZWJ)
+    break;
+
   /* Not breaking between Regional Indicators is allowed only if there
   are an even number of preceding RIs. */
 
@@ -129,12 +140,15 @@ while (eptr < end_subject)
     if ((ricount & 1) != 0) break;  /* Grapheme break required */
     }
 
-  /* If Extend or ZWJ follows Extended_Pictographic, do not update lgb; this
-  allows any number of them before a following Extended_Pictographic. */
+  /* Set a flag when ZWJ follows Extended Pictographic (with optional Extend in
+  between; see next statement). */
+
+  was_ep_ZWJ = (lgb == ucp_gbExtended_Pictographic && rgb == ucp_gbZWJ);
+
+  /* If Extend follows Extended_Pictographic, do not update lgb; this allows
+  any number of them before a following ZWJ. */
 
-  if ((rgb != ucp_gbExtend && rgb != ucp_gbZWJ) ||
-       lgb != ucp_gbExtended_Pictographic)
-    lgb = rgb;
+  if (rgb != ucp_gbExtend || lgb != ucp_gbExtended_Pictographic) lgb = rgb;
 
   eptr += len;
   if (xcount != NULL) *xcount += 1;

diff --git a/pcre2-sys/upstream/src/pcre2_intmodedep.h b/pcre2-sys/upstream/src/pcre2_intmodedep.h
@@ -7,7 +7,7 @@ and semantics are as close as possible to those of the Perl 5 language.
 
                        Written by Philip Hazel
      Original API code Copyright (c) 1997-2012 University of Cambridge
-          New API code Copyright (c) 2016-2023 University of Cambridge
+          New API code Copyright (c) 2016-2024 University of Cambridge
 
 -----------------------------------------------------------------------------
 Redistribution and use in source and binary forms, with or without
@@ -568,6 +568,7 @@ typedef struct pcre2_real_compile_context {
   void *stack_guard_data;
   const uint8_t *tables;
   PCRE2_SIZE max_pattern_length;
+  PCRE2_SIZE max_pattern_compiled_length;
   uint16_t bsr_convention;
   uint16_t newline_convention;
   uint32_t parens_nest_limit;