From b43272afa29a64dcb8bcf26a96a05bac40792b92 Mon Sep 17 00:00:00 2001
From: jaime-m-p <167997752+jaime-m-p@users.noreply.github.com>
Date: Sat, 18 May 2024 01:09:13 +0200
Subject: Unicode codepoint flags for custom regexs (#7245)

* Replace CODEPOINT_TYPE_* with codepoint_flags
* Update and bugfix brute force random test
* Deterministic brute force random test
* Unicode normalization NFD
* Get rid of BOM
---
 unicode.h | 56 ++++++++++++++++++++++++++++++++++++++++++++------------
 1 file changed, 44 insertions(+), 12 deletions(-)

(limited to 'unicode.h')

diff --git a/unicode.h b/unicode.h
index d6a14d47..7513be4a 100644
--- a/unicode.h
+++ b/unicode.h
@@ -4,24 +4,56 @@
 #include <string>
 #include <vector>
 
-#define CODEPOINT_TYPE_UNIDENTIFIED 0
-#define CODEPOINT_TYPE_NUMBER       1
-#define CODEPOINT_TYPE_LETTER       2
-#define CODEPOINT_TYPE_SEPARATOR    3
-#define CODEPOINT_TYPE_ACCENT_MARK  4
-#define CODEPOINT_TYPE_PUNCTUATION  5
-#define CODEPOINT_TYPE_SYMBOL       6
-#define CODEPOINT_TYPE_CONTROL      7
+struct codepoint_flags {
+    enum {
+        UNDEFINED       = 0x0001,
+        NUMBER          = 0x0002,  // regex: \p{N}
+        LETTER          = 0x0004,  // regex: \p{L}
+        SEPARATOR       = 0x0008,  // regex: \p{Z}
+        ACCENT_MARK     = 0x0010,  // regex: \p{M}
+        PUNCTUATION     = 0x0020,  // regex: \p{P}
+        SYMBOL          = 0x0040,  // regex: \p{S}
+        CONTROL         = 0x0080,  // regex: \p{C}
+        MASK_CATEGORIES = 0x00FF,
+    };
+
+    // codepoint type
+    uint16_t is_undefined   : 1;
+    uint16_t is_number      : 1;  // regex: \p{N}
+    uint16_t is_letter      : 1;  // regex: \p{L}
+    uint16_t is_separator   : 1;  // regex: \p{Z}
+    uint16_t is_accent_mark : 1;  // regex: \p{M}
+    uint16_t is_punctuation : 1;  // regex: \p{P}
+    uint16_t is_symbol      : 1;  // regex: \p{S}
+    uint16_t is_control     : 1;  // regex: \p{C}
+    // helper flags
+    uint16_t is_whitespace  : 1;  // regex: \s
+    uint16_t is_lowercase   : 1;
+    uint16_t is_uppercase   : 1;
+    uint16_t is_nfd         : 1;
+
+    // decode from uint16
+    inline codepoint_flags(const uint16_t flags=0) {
+        *reinterpret_cast<uint16_t*>(this) = flags;
+    }
+
+    inline uint16_t as_uint() const {
+        return *reinterpret_cast<const uint16_t*>(this);
+    }
+
+    inline uint16_t category_flag() const {
+        return this->as_uint() & MASK_CATEGORIES;
+    }
+};
+
 
 std::string unicode_cpt_to_utf8(uint32_t cp);
 std::vector<uint32_t> unicode_cpts_from_utf8(const std::string & utf8);
 
 std::vector<uint32_t> unicode_cpts_normalize_nfd(const std::vector<uint32_t> & cpts);
 
-int unicode_cpt_type(uint32_t cp);
-int unicode_cpt_type(const std::string & utf8);
-
-bool unicode_cpt_is_whitespace(uint32_t cp);
+codepoint_flags unicode_cpt_flags(const uint32_t cp);
+codepoint_flags unicode_cpt_flags(const std::string & utf8);
 
 std::string unicode_byte_to_utf8(uint8_t byte);
 uint8_t unicode_utf8_to_byte(const std::string & utf8);
-- 
cgit v1.2.3