1//===-- Support/DJB.cpp ---DJB Hash -----------------------------*- C++ -*-===//
2//
3// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4// See https://llvm.org/LICENSE.txt for license information.
5// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6//
7//===----------------------------------------------------------------------===//
8//
9// This file contains support for the DJ Bernstein hash function.
10//
11//===----------------------------------------------------------------------===//
12
13#include "llvm/Support/DJB.h"
14#include "llvm/ADT/ArrayRef.h"
15#include "llvm/Support/Compiler.h"
16#include "llvm/Support/ConvertUTF.h"
17#include "llvm/Support/Unicode.h"
18
19using namespace llvm;
20
21static UTF32 chopOneUTF32(StringRef &Buffer) {
22 UTF32 C;
23 const UTF8 *const Begin8Const =
24 reinterpret_cast<const UTF8 *>(Buffer.begin());
25 const UTF8 *Begin8 = Begin8Const;
26 UTF32 *Begin32 = &C;
27
28 // In lenient mode we will always end up with a "reasonable" value in C for
29 // non-empty input.
30 assert(!Buffer.empty());
31 ConvertUTF8toUTF32(sourceStart: &Begin8, sourceEnd: reinterpret_cast<const UTF8 *>(Buffer.end()),
32 targetStart: &Begin32, targetEnd: &C + 1, flags: lenientConversion);
33 Buffer = Buffer.drop_front(N: Begin8 - Begin8Const);
34 return C;
35}
36
37static StringRef toUTF8(UTF32 C, MutableArrayRef<UTF8> Storage) {
38 const UTF32 *Begin32 = &C;
39 UTF8 *Begin8 = Storage.begin();
40
41 // The case-folded output should always be a valid unicode character, so use
42 // strict mode here.
43 ConversionResult CR = ConvertUTF32toUTF8(sourceStart: &Begin32, sourceEnd: &C + 1, targetStart: &Begin8,
44 targetEnd: Storage.end(), flags: strictConversion);
45 assert(CR == conversionOK && "Case folding produced invalid char?");
46 (void)CR;
47 return StringRef(reinterpret_cast<char *>(Storage.begin()),
48 Begin8 - Storage.begin());
49}
50
51static UTF32 foldCharDwarf(UTF32 C) {
52 // DWARF v5 addition to the unicode folding rules.
53 // Fold "Latin Small Letter Dotless I" and "Latin Capital Letter I With Dot
54 // Above" into "i".
55 if (C == 0x130 || C == 0x131)
56 return 'i';
57 return sys::unicode::foldCharSimple(C);
58}
59
60static std::optional<uint32_t> fastCaseFoldingDjbHash(StringRef Buffer,
61 uint32_t H) {
62 bool AllASCII = true;
63 for (unsigned char C : Buffer) {
64 H = H * 33 + ('A' <= C && C <= 'Z' ? C - 'A' + 'a' : C);
65 AllASCII &= C <= 0x7f;
66 }
67 if (AllASCII)
68 return H;
69 return std::nullopt;
70}
71
72uint32_t llvm::caseFoldingDjbHash(StringRef Buffer, uint32_t H) {
73 if (std::optional<uint32_t> Result = fastCaseFoldingDjbHash(Buffer, H))
74 return *Result;
75
76 std::array<UTF8, UNI_MAX_UTF8_BYTES_PER_CODE_POINT> Storage;
77 while (!Buffer.empty()) {
78 UTF32 C = foldCharDwarf(C: chopOneUTF32(Buffer));
79 StringRef Folded = toUTF8(C, Storage);
80 H = djbHash(Buffer: Folded, H);
81 }
82 return H;
83}
84