mirror of
				https://github.com/ggml-org/llama.cpp.git
				synced 2025-10-28 08:31:25 +00:00 
			
		
		
		
	
		
			
				
	
	
		
			21 lines
		
	
	
		
			630 B
		
	
	
	
		
			C++
		
	
	
	
	
	
			
		
		
	
	
			21 lines
		
	
	
		
			630 B
		
	
	
	
		
			C++
		
	
	
	
	
	
| #pragma once
 | |
| 
 | |
| #include <cstdint>
 | |
| #include <vector>
 | |
| #include <unordered_map>
 | |
| #include <unordered_set>
 | |
| 
 | |
| struct range_nfd {
 | |
|     uint32_t first;
 | |
|     uint32_t last;
 | |
|     uint32_t nfd;
 | |
| };
 | |
| 
 | |
| static const uint32_t MAX_CODEPOINTS = 0x110000;
 | |
| 
 | |
| extern const std::initializer_list<std::pair<uint32_t, uint16_t>> unicode_ranges_flags;
 | |
| extern const std::unordered_set<uint32_t> unicode_set_whitespace;
 | |
| extern const std::initializer_list<std::pair<uint32_t, uint32_t>> unicode_map_lowercase;
 | |
| extern const std::initializer_list<std::pair<uint32_t, uint32_t>> unicode_map_uppercase;
 | |
| extern const std::initializer_list<range_nfd> unicode_ranges_nfd;
 | 
