1 //===-- llvm/ADT/StringExtras.h - Useful string functions -------*- C++ -*-===//
2 //
3 // The LLVM Compiler Infrastructure
4 //
5 // This file is distributed under the University of Illinois Open Source
6 // License. See LICENSE.TXT for details.
7 //
8 //===----------------------------------------------------------------------===//
9 //
10 // This file contains some functions that are useful when dealing with strings.
11 //
12 //===----------------------------------------------------------------------===//
13
14 #ifndef LLVM_ADT_STRINGEXTRAS_H
15 #define LLVM_ADT_STRINGEXTRAS_H
16
17 #include "llvm/Support/DataTypes.h"
18 #include "llvm/ADT/APFloat.h"
19 #include "llvm/ADT/DenseMapInfo.h"
20 #include "llvm/ADT/StringRef.h"
21 #include <cctype>
22 #include <cstdio>
23 #include <string>
24
25 namespace llvm {
26 template<typename T> class SmallVectorImpl;
27
28 /// hexdigit - Return the hexadecimal character for the
29 /// given number \arg X (which should be less than 16).
30 static inline char hexdigit(unsigned X, bool LowerCase = false) {
31 const char HexChar = LowerCase ? 'a' : 'A';
32 return X < 10 ? '0' + X : HexChar + X - 10;
33 }
34
35 /// utohex_buffer - Emit the specified number into the buffer specified by
36 /// BufferEnd, returning a pointer to the start of the string. This can be used
37 /// like this: (note that the buffer must be large enough to handle any number):
38 /// char Buffer[40];
39 /// printf("0x%s", utohex_buffer(X, Buffer+40));
40 ///
41 /// This should only be used with unsigned types.
42 ///
43 template<typename IntTy>
utohex_buffer(IntTy X,char * BufferEnd)44 static inline char *utohex_buffer(IntTy X, char *BufferEnd) {
45 char *BufPtr = BufferEnd;
46 *--BufPtr = 0; // Null terminate buffer.
47 if (X == 0) {
48 *--BufPtr = '0'; // Handle special case.
49 return BufPtr;
50 }
51
52 while (X) {
53 unsigned char Mod = static_cast<unsigned char>(X) & 15;
54 *--BufPtr = hexdigit(Mod);
55 X >>= 4;
56 }
57 return BufPtr;
58 }
59
utohexstr(uint64_t X)60 static inline std::string utohexstr(uint64_t X) {
61 char Buffer[17];
62 return utohex_buffer(X, Buffer+17);
63 }
64
65 static inline std::string utostr_32(uint32_t X, bool isNeg = false) {
66 char Buffer[11];
67 char *BufPtr = Buffer+11;
68
69 if (X == 0) *--BufPtr = '0'; // Handle special case...
70
71 while (X) {
72 *--BufPtr = '0' + char(X % 10);
73 X /= 10;
74 }
75
76 if (isNeg) *--BufPtr = '-'; // Add negative sign...
77
78 return std::string(BufPtr, Buffer+11);
79 }
80
81 static inline std::string utostr(uint64_t X, bool isNeg = false) {
82 char Buffer[21];
83 char *BufPtr = Buffer+21;
84
85 if (X == 0) *--BufPtr = '0'; // Handle special case...
86
87 while (X) {
88 *--BufPtr = '0' + char(X % 10);
89 X /= 10;
90 }
91
92 if (isNeg) *--BufPtr = '-'; // Add negative sign...
93 return std::string(BufPtr, Buffer+21);
94 }
95
96
itostr(int64_t X)97 static inline std::string itostr(int64_t X) {
98 if (X < 0)
99 return utostr(static_cast<uint64_t>(-X), true);
100 else
101 return utostr(static_cast<uint64_t>(X));
102 }
103
ftostr(double V)104 static inline std::string ftostr(double V) {
105 char Buffer[200];
106 sprintf(Buffer, "%20.6e", V);
107 char *B = Buffer;
108 while (*B == ' ') ++B;
109 return B;
110 }
111
ftostr(const APFloat & V)112 static inline std::string ftostr(const APFloat& V) {
113 if (&V.getSemantics() == &APFloat::IEEEdouble)
114 return ftostr(V.convertToDouble());
115 else if (&V.getSemantics() == &APFloat::IEEEsingle)
116 return ftostr((double)V.convertToFloat());
117 return "<unknown format in ftostr>"; // error
118 }
119
LowercaseString(const std::string & S)120 static inline std::string LowercaseString(const std::string &S) {
121 std::string result(S);
122 for (unsigned i = 0; i < S.length(); ++i)
123 if (isupper(result[i]))
124 result[i] = char(tolower(result[i]));
125 return result;
126 }
127
UppercaseString(const std::string & S)128 static inline std::string UppercaseString(const std::string &S) {
129 std::string result(S);
130 for (unsigned i = 0; i < S.length(); ++i)
131 if (islower(result[i]))
132 result[i] = char(toupper(result[i]));
133 return result;
134 }
135
136 /// StrInStrNoCase - Portable version of strcasestr. Locates the first
137 /// occurrence of string 's1' in string 's2', ignoring case. Returns
138 /// the offset of s2 in s1 or npos if s2 cannot be found.
139 StringRef::size_type StrInStrNoCase(StringRef s1, StringRef s2);
140
141 /// getToken - This function extracts one token from source, ignoring any
142 /// leading characters that appear in the Delimiters string, and ending the
143 /// token at any of the characters that appear in the Delimiters string. If
144 /// there are no tokens in the source string, an empty string is returned.
145 /// The function returns a pair containing the extracted token and the
146 /// remaining tail string.
147 std::pair<StringRef, StringRef> getToken(StringRef Source,
148 StringRef Delimiters = " \t\n\v\f\r");
149
150 /// SplitString - Split up the specified string according to the specified
151 /// delimiters, appending the result fragments to the output list.
152 void SplitString(StringRef Source,
153 SmallVectorImpl<StringRef> &OutFragments,
154 StringRef Delimiters = " \t\n\v\f\r");
155
156 /// HashString - Hash function for strings.
157 ///
158 /// This is the Bernstein hash function.
159 //
160 // FIXME: Investigate whether a modified bernstein hash function performs
161 // better: http://eternallyconfuzzled.com/tuts/algorithms/jsw_tut_hashing.aspx
162 // X*33+c -> X*33^c
163 static inline unsigned HashString(StringRef Str, unsigned Result = 0) {
164 for (unsigned i = 0, e = Str.size(); i != e; ++i)
165 Result = Result * 33 + Str[i];
166 return Result;
167 }
168
169 } // End llvm namespace
170
171 #endif
172