add windows compat with PCRE2_BUILD_FROM_SOURCE=1 control (#10)

* add windows compat with PCRE2_BUILD_FROM_SOURCE=1 control

* update docs

* fix compile

* fix compile

* fix version

* fix windows linking

* fix windows compile

* fix windows compile2

* compile for utf16

* revert to use pcre2-8 utf8

Author: Qubitium

README.md

@@ -57,9 +57,28 @@ exposing PCRE2’s extended flag set through the Pythonic `Flag` enum
 - `pcre.escape()` delegates directly to `re.escape` for byte and text
   patterns so escaping semantics remain identical.
 
+### `regex` package compatibility
+
+The [`regex`](https://pypi.org/project/regex/) package interprets both
+`\uXXXX`/`\u{...}` and `\UXXXXXXXX` escapes as UTF-8 code points, while
+PCRE2 expects hexadecimal escapes to use the `\x{...}` form. Enable
+`Flag.COMPAT_REGEX` to translate those escapes automatically when compiling
+patterns:
+
+```python
+from pcre import compile, Flag
+
+pattern = compile(r"\\u{1F600}", flags=Flag.COMPAT_REGEX)
+assert pattern.pattern == r"\\x{1F600}"
+```
+
+Set the default behaviour globally with `pcre.configure(compat_regex=True)`
+so that subsequent calls to `compile()` and the module-level helpers apply
+the conversion without repeating the flag.
+
 ### Automatic pattern caching
 
-`pcre.compile()` caches the final `Pattern` wrapper for up to 2048
+`pcre.compile()` caches the final `Pattern` wrapper for up to 128
 unique `(pattern, flags)` pairs when the pattern object is hashable. This
 keeps repeated calls to top-level helpers efficient without any extra work
 from the caller. Adjust the capacity with `pcre.set_cache_limit(n)`—pass
@@ -168,7 +187,7 @@ location.
 # Notes
 
 ## Pattern cache
-- `pcre.compile()` caches hashable `(pattern, flags)` pairs, keeping up to 2048 entries.
+- `pcre.compile()` caches hashable `(pattern, flags)` pairs, keeping up to 128 entries.
 - Use `pcre.clear_cache()` when you need to free the cache proactively.
 - Non-hashable pattern objects skip the cache and are compiled each time.
 
@@ -179,12 +198,12 @@ location.
 
 ## Additional usage notes
 - All top-level helpers (`match`, `search`, `fullmatch`, `finditer`, `findall`) defer to the cached compiler.
-- Compiled `Pattern` objects expose `.pattern`, `.pattern_bytes`, `.flags`, and `.groupindex` for introspection.
+- Compiled `Pattern` objects expose `.pattern`, `.flags`, `.jit`, and `.groupindex` for introspection.
 - Execution helpers accept `pos`, `endpos`, and `options`, allowing you to thread PCRE2 execution flags per call.
 
 ## Memory allocation
 - The extension selects the fastest available allocator at import time: it
-  prefers tcmalloc, then jemalloc, and finally falls back to the platform
+  prefers jemalloc, then tcmalloc, and finally falls back to the platform
   `malloc`. Optional allocators are loaded via `dlopen`, so no additional
   link flags are required when they are absent.
 - All internal buffers (match data wrappers, JIT stack cache entries, error

pcre_ext/pcre2.c

@@ -10,9 +10,31 @@
 #include <immintrin.h>
 #endif
 
+#define STRINGIFY_DETAIL(value) #value
+#define STRINGIFY(value) STRINGIFY_DETAIL(value)
+
+static const char *
+resolve_pcre2_prerelease(void)
+{
+    const char *raw = STRINGIFY(Z PCRE2_PRERELEASE);
+
+    if (raw[1] == '\0') {
+        return "";
+    }
+
+    raw += 1;
+    while (*raw == ' ') {
+        raw++;
+    }
+
+    return raw;
+}
+
 static int default_jit_enabled = 1;
 static PyThread_type_lock default_jit_lock = NULL;
 static PyThread_type_lock cpu_feature_lock = NULL;
+static char pcre2_library_version[64] = "unknown";
+static int pcre2_version_initialized = 0;
 #if defined(PCRE2_USE_OFFSET_LIMIT)
 static int offset_limit_support = -1;
 #endif
@@ -1871,6 +1893,8 @@ module_configure(PyObject *Py_UNUSED(module), PyObject *args, PyObject *kwargs)
 
 static PyObject *module_cpu_ascii_vector_mode(PyObject *Py_UNUSED(module), PyObject *Py_UNUSED(args));
 static PyObject *module_memory_allocator(PyObject *Py_UNUSED(module), PyObject *Py_UNUSED(args));
+static PyObject *module_get_pcre2_version(PyObject *Py_UNUSED(module), PyObject *Py_UNUSED(args));
+static void initialize_pcre2_version(void);
 
 static PyMethodDef module_methods[] = {
     {"compile", (PyCFunction)module_compile, METH_VARARGS | METH_KEYWORDS, PyDoc_STR("Compile a pattern into a PCRE2 Pattern object." )},
@@ -1888,6 +1912,7 @@ static PyMethodDef module_methods[] = {
     {"get_jit_stack_cache_count", (PyCFunction)module_get_jit_stack_cache_count, METH_NOARGS, PyDoc_STR("Return the number of cached JIT stacks currently stored." )},
     {"get_jit_stack_limits", (PyCFunction)module_get_jit_stack_limits, METH_NOARGS, PyDoc_STR("Return the configured (start, max) JIT stack sizes." )},
     {"set_jit_stack_limits", (PyCFunction)module_set_jit_stack_limits, METH_VARARGS, PyDoc_STR("Set the (start, max) sizes for newly created JIT stacks." )},
+    {"get_library_version", (PyCFunction)module_get_pcre2_version, METH_NOARGS, PyDoc_STR("Return the PCRE2 library version string." )},
     {"get_allocator", (PyCFunction)module_memory_allocator, METH_NOARGS, PyDoc_STR("Return the name of the active heap allocator (tcmalloc/jemalloc/malloc)." )},
     {"_cpu_ascii_vector_mode", (PyCFunction)module_cpu_ascii_vector_mode, METH_NOARGS, PyDoc_STR("Return the active ASCII vector width (0=scalar,1=SSE2,2=AVX2,3=AVX512)." )},
     {NULL, NULL, 0, NULL},
@@ -2023,6 +2048,12 @@ PyInit_cpcre2(void)
         goto error;
     }
 
+    initialize_pcre2_version();
+
+    if (PyModule_AddStringConstant(module, "PCRE2_VERSION", pcre2_library_version) < 0) {
+        goto error;
+    }
+
     if (PyModule_AddStringConstant(module, "__version__", "0.1.0") < 0) {
         goto error;
     }
@@ -2071,3 +2102,45 @@ module_memory_allocator(PyObject *Py_UNUSED(module), PyObject *Py_UNUSED(args))
     const char *name = pcre_memory_allocator_name();
     return PyUnicode_FromString(name);
 }
+
+static PyObject *
+module_get_pcre2_version(PyObject *Py_UNUSED(module), PyObject *Py_UNUSED(args))
+{
+    initialize_pcre2_version();
+    return PyUnicode_FromString(pcre2_library_version);
+}
+
+static void
+initialize_pcre2_version(void)
+{
+    if (pcre2_version_initialized) {
+        return;
+    }
+
+    char buffer[sizeof(pcre2_library_version)] = {0};
+    if (pcre2_config(PCRE2_CONFIG_VERSION, buffer) == 0 && buffer[0] != '\0') {
+        strncpy(pcre2_library_version, buffer, sizeof(pcre2_library_version) - 1);
+        pcre2_library_version[sizeof(pcre2_library_version) - 1] = '\0';
+    } else {
+        const char *pre_release = resolve_pcre2_prerelease();
+        if (pre_release[0] != '\0') {
+            (void)snprintf(
+                pcre2_library_version,
+                sizeof(pcre2_library_version),
+                "%d.%d-%s",
+                PCRE2_MAJOR,
+                PCRE2_MINOR,
+                pre_release
+            );
+        } else {
+            (void)snprintf(
+                pcre2_library_version,
+                sizeof(pcre2_library_version),
+                "%d.%d",
+                PCRE2_MAJOR,
+                PCRE2_MINOR
+            );
+        }
+    }
+    pcre2_version_initialized = 1;
+}

pcre_ext/pcre2.h

@@ -42,8 +42,8 @@ POSSIBILITY OF SUCH DAMAGE.
 /* The current PCRE version information. */
 
 #define PCRE2_MAJOR           10
-#define PCRE2_MINOR           47
-#define PCRE2_PRERELEASE      -DEV
+#define PCRE2_MINOR           46
+#define PCRE2_PRERELEASE      
 #define PCRE2_DATE            2025-08-27
 
 /* When an application links to a PCRE DLL in Windows, the symbols that are
@@ -343,10 +343,6 @@ pcre2_pattern_convert(). */
 #define PCRE2_ERROR_PERL_ECLASS_EMPTY_EXPR         214
 #define PCRE2_ERROR_PERL_ECLASS_MISSING_CLOSE      215
 #define PCRE2_ERROR_PERL_ECLASS_UNEXPECTED_CHAR    216
-#define PCRE2_ERROR_EXPECTED_CAPTURE_GROUP         217
-#define PCRE2_ERROR_MISSING_OPENING_PARENTHESIS    218
-#define PCRE2_ERROR_MISSING_NUMBER_TERMINATOR      219
-#define PCRE2_ERROR_NULL_ERROROFFSET               220
 
 /* "Expected" matching error codes: no match and partial match. */
 
@@ -436,10 +432,6 @@ released, the numbers must not be changed. */
 #define PCRE2_ERROR_JIT_UNSUPPORTED   (-68)
 #define PCRE2_ERROR_REPLACECASE       (-69)
 #define PCRE2_ERROR_TOOLARGEREPLACE   (-70)
-#define PCRE2_ERROR_DIFFSUBSPATTERN   (-71)
-#define PCRE2_ERROR_DIFFSUBSSUBJECT   (-72)
-#define PCRE2_ERROR_DIFFSUBSOFFSET    (-73)
-#define PCRE2_ERROR_DIFFSUBSOPTIONS   (-74)
 
 
 /* Request types for pcre2_pattern_info() */
@@ -492,7 +484,6 @@ released, the numbers must not be changed. */
 #define PCRE2_CONFIG_NEVER_BACKSLASH_C      13
 #define PCRE2_CONFIG_COMPILED_WIDTHS        14
 #define PCRE2_CONFIG_TABLES_LENGTH          15
-#define PCRE2_CONFIG_EFFECTIVE_LINKSIZE     16
 
 /* Optimization directives for pcre2_set_optimize().
 For binary compatibility, only add to this list; do not renumber. */
@@ -752,14 +743,14 @@ PCRE2_EXP_DECL pcre2_match_data *PCRE2_CALL_CONVENTION \
 PCRE2_EXP_DECL pcre2_match_data *PCRE2_CALL_CONVENTION \
   pcre2_match_data_create_from_pattern(const pcre2_code *, \
     pcre2_general_context *); \
-PCRE2_EXP_DECL void PCRE2_CALL_CONVENTION \
-  pcre2_match_data_free(pcre2_match_data *); \
 PCRE2_EXP_DECL int PCRE2_CALL_CONVENTION \
   pcre2_dfa_match(const pcre2_code *, PCRE2_SPTR, PCRE2_SIZE, PCRE2_SIZE, \
     uint32_t, pcre2_match_data *, pcre2_match_context *, int *, PCRE2_SIZE); \
 PCRE2_EXP_DECL int PCRE2_CALL_CONVENTION \
   pcre2_match(const pcre2_code *, PCRE2_SPTR, PCRE2_SIZE, PCRE2_SIZE, \
     uint32_t, pcre2_match_data *, pcre2_match_context *); \
+PCRE2_EXP_DECL void PCRE2_CALL_CONVENTION \
+  pcre2_match_data_free(pcre2_match_data *); \
 PCRE2_EXP_DECL PCRE2_SPTR PCRE2_CALL_CONVENTION \
   pcre2_get_mark(pcre2_match_data *); \
 PCRE2_EXP_DECL PCRE2_SIZE PCRE2_CALL_CONVENTION \
@@ -771,9 +762,7 @@ PCRE2_EXP_DECL uint32_t PCRE2_CALL_CONVENTION \
 PCRE2_EXP_DECL PCRE2_SIZE *PCRE2_CALL_CONVENTION \
   pcre2_get_ovector_pointer(pcre2_match_data *); \
 PCRE2_EXP_DECL PCRE2_SIZE PCRE2_CALL_CONVENTION \
-  pcre2_get_startchar(pcre2_match_data *); \
-PCRE2_EXP_DECL int PCRE2_CALL_CONVENTION \
-  pcre2_next_match(pcre2_match_data *, PCRE2_SIZE *, uint32_t *);
+  pcre2_get_startchar(pcre2_match_data *);
 
 
 /* Convenience functions for handling matched substrings. */
@@ -953,7 +942,6 @@ pcre2_compile are called by application code. */
 #define pcre2_match_data_create               PCRE2_SUFFIX(pcre2_match_data_create_)
 #define pcre2_match_data_create_from_pattern  PCRE2_SUFFIX(pcre2_match_data_create_from_pattern_)
 #define pcre2_match_data_free                 PCRE2_SUFFIX(pcre2_match_data_free_)
-#define pcre2_next_match                      PCRE2_SUFFIX(pcre2_next_match_)
 #define pcre2_pattern_convert                 PCRE2_SUFFIX(pcre2_pattern_convert_)
 #define pcre2_pattern_info                    PCRE2_SUFFIX(pcre2_pattern_info_)
 #define pcre2_serialize_decode                PCRE2_SUFFIX(pcre2_serialize_decode_)

pcre_ext/util.c

@@ -5,6 +5,24 @@
 
 #include "pcre2_module.h"
 #include <string.h>
+#include <stdint.h>
+
+#if defined(_MSC_VER)
+static inline unsigned int
+popcountll(uint64_t value)
+{
+    value -= (value >> 1) & 0x5555555555555555ULL;
+    value = (value & 0x3333333333333333ULL) + ((value >> 2) & 0x3333333333333333ULL);
+    value = (value + (value >> 4)) & 0x0F0F0F0F0F0F0F0FULL;
+    return (unsigned int)((value * 0x0101010101010101ULL) >> 56);
+}
+#else
+static inline unsigned int
+popcountll(uint64_t value)
+{
+    return (unsigned int)__builtin_popcountll(value);
+}
+#endif
 
 PyObject *
 bytes_from_text(PyObject *obj)
@@ -75,7 +93,7 @@ utf8_index_to_offset(PyObject *unicode_obj, Py_ssize_t index, Py_ssize_t *offset
         for (Py_ssize_t i = 0; i < fast_chunks; ++i) {
             uint64_t block;
             memcpy(&block, ptr, sizeof(uint64_t));
-            non_ascii += __builtin_popcountll(block & high_bit_mask);
+            non_ascii += popcountll(block & high_bit_mask);
             ptr += chunk;
         }