1 //===--- clang/Basic/CharInfo.h - Classifying ASCII Characters ------------===//
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 #ifndef LLVM_CLANG_BASIC_CHARINFO_H
11 #define LLVM_CLANG_BASIC_CHARINFO_H
12 
13 #include "clang/Basic/LLVM.h"
14 #include "llvm/ADT/StringRef.h"
15 #include "llvm/Support/Compiler.h"
16 #include "llvm/Support/DataTypes.h"
17 
18 namespace clang {
19 namespace charinfo {
20   extern const uint16_t InfoTable[256];
21 
22   enum {
23     CHAR_HORZ_WS  = 0x0001,  // '\t', '\f', '\v'.  Note, no '\0'
24     CHAR_VERT_WS  = 0x0002,  // '\r', '\n'
25     CHAR_SPACE    = 0x0004,  // ' '
26     CHAR_DIGIT    = 0x0008,  // 0-9
27     CHAR_XLETTER  = 0x0010,  // a-f,A-F
28     CHAR_UPPER    = 0x0020,  // A-Z
29     CHAR_LOWER    = 0x0040,  // a-z
30     CHAR_UNDER    = 0x0080,  // _
31     CHAR_PERIOD   = 0x0100,  // .
32     CHAR_RAWDEL   = 0x0200,  // {}[]#<>%:;?*+-/^&|~!=,"'
33     CHAR_PUNCT    = 0x0400   // `$@()
34   };
35 
36   enum {
37     CHAR_XUPPER = CHAR_XLETTER | CHAR_UPPER,
38     CHAR_XLOWER = CHAR_XLETTER | CHAR_LOWER
39   };
40 } // end namespace charinfo
41 
42 /// Returns true if this is an ASCII character.
isASCII(char c)43 LLVM_READNONE static inline bool isASCII(char c) {
44   return static_cast<unsigned char>(c) <= 127;
45 }
46 
47 /// Returns true if this is a valid first character of a C identifier,
48 /// which is [a-zA-Z_].
49 LLVM_READONLY static inline bool isIdentifierHead(unsigned char c,
50                                                   bool AllowDollar = false) {
51   using namespace charinfo;
52   if (InfoTable[c] & (CHAR_UPPER|CHAR_LOWER|CHAR_UNDER))
53     return true;
54   return AllowDollar && c == '$';
55 }
56 
57 /// Returns true if this is a body character of a C identifier,
58 /// which is [a-zA-Z0-9_].
59 LLVM_READONLY static inline bool isIdentifierBody(unsigned char c,
60                                                   bool AllowDollar = false) {
61   using namespace charinfo;
62   if (InfoTable[c] & (CHAR_UPPER|CHAR_LOWER|CHAR_DIGIT|CHAR_UNDER))
63     return true;
64   return AllowDollar && c == '$';
65 }
66 
67 /// Returns true if this character is horizontal ASCII whitespace:
68 /// ' ', '\\t', '\\f', '\\v'.
69 ///
70 /// Note that this returns false for '\\0'.
isHorizontalWhitespace(unsigned char c)71 LLVM_READONLY static inline bool isHorizontalWhitespace(unsigned char c) {
72   using namespace charinfo;
73   return (InfoTable[c] & (CHAR_HORZ_WS|CHAR_SPACE)) != 0;
74 }
75 
76 /// Returns true if this character is vertical ASCII whitespace: '\\n', '\\r'.
77 ///
78 /// Note that this returns false for '\\0'.
isVerticalWhitespace(unsigned char c)79 LLVM_READONLY static inline bool isVerticalWhitespace(unsigned char c) {
80   using namespace charinfo;
81   return (InfoTable[c] & CHAR_VERT_WS) != 0;
82 }
83 
84 /// Return true if this character is horizontal or vertical ASCII whitespace:
85 /// ' ', '\\t', '\\f', '\\v', '\\n', '\\r'.
86 ///
87 /// Note that this returns false for '\\0'.
isWhitespace(unsigned char c)88 LLVM_READONLY static inline bool isWhitespace(unsigned char c) {
89   using namespace charinfo;
90   return (InfoTable[c] & (CHAR_HORZ_WS|CHAR_VERT_WS|CHAR_SPACE)) != 0;
91 }
92 
93 /// Return true if this character is an ASCII digit: [0-9]
isDigit(unsigned char c)94 LLVM_READONLY static inline bool isDigit(unsigned char c) {
95   using namespace charinfo;
96   return (InfoTable[c] & CHAR_DIGIT) != 0;
97 }
98 
99 /// Return true if this character is a lowercase ASCII letter: [a-z]
isLowercase(unsigned char c)100 LLVM_READONLY static inline bool isLowercase(unsigned char c) {
101   using namespace charinfo;
102   return (InfoTable[c] & CHAR_LOWER) != 0;
103 }
104 
105 /// Return true if this character is an uppercase ASCII letter: [A-Z]
isUppercase(unsigned char c)106 LLVM_READONLY static inline bool isUppercase(unsigned char c) {
107   using namespace charinfo;
108   return (InfoTable[c] & CHAR_UPPER) != 0;
109 }
110 
111 /// Return true if this character is an ASCII letter: [a-zA-Z]
isLetter(unsigned char c)112 LLVM_READONLY static inline bool isLetter(unsigned char c) {
113   using namespace charinfo;
114   return (InfoTable[c] & (CHAR_UPPER|CHAR_LOWER)) != 0;
115 }
116 
117 /// Return true if this character is an ASCII letter or digit: [a-zA-Z0-9]
isAlphanumeric(unsigned char c)118 LLVM_READONLY static inline bool isAlphanumeric(unsigned char c) {
119   using namespace charinfo;
120   return (InfoTable[c] & (CHAR_DIGIT|CHAR_UPPER|CHAR_LOWER)) != 0;
121 }
122 
123 /// Return true if this character is an ASCII hex digit: [0-9a-fA-F]
isHexDigit(unsigned char c)124 LLVM_READONLY static inline bool isHexDigit(unsigned char c) {
125   using namespace charinfo;
126   return (InfoTable[c] & (CHAR_DIGIT|CHAR_XLETTER)) != 0;
127 }
128 
129 /// Return true if this character is an ASCII punctuation character.
130 ///
131 /// Note that '_' is both a punctuation character and an identifier character!
isPunctuation(unsigned char c)132 LLVM_READONLY static inline bool isPunctuation(unsigned char c) {
133   using namespace charinfo;
134   return (InfoTable[c] & (CHAR_UNDER|CHAR_PERIOD|CHAR_RAWDEL|CHAR_PUNCT)) != 0;
135 }
136 
137 /// Return true if this character is an ASCII printable character; that is, a
138 /// character that should take exactly one column to print in a fixed-width
139 /// terminal.
isPrintable(unsigned char c)140 LLVM_READONLY static inline bool isPrintable(unsigned char c) {
141   using namespace charinfo;
142   return (InfoTable[c] & (CHAR_UPPER|CHAR_LOWER|CHAR_PERIOD|CHAR_PUNCT|
143                           CHAR_DIGIT|CHAR_UNDER|CHAR_RAWDEL|CHAR_SPACE)) != 0;
144 }
145 
146 /// Return true if this is the body character of a C preprocessing number,
147 /// which is [a-zA-Z0-9_.].
isPreprocessingNumberBody(unsigned char c)148 LLVM_READONLY static inline bool isPreprocessingNumberBody(unsigned char c) {
149   using namespace charinfo;
150   return (InfoTable[c] &
151           (CHAR_UPPER|CHAR_LOWER|CHAR_DIGIT|CHAR_UNDER|CHAR_PERIOD)) != 0;
152 }
153 
154 /// Return true if this is the body character of a C++ raw string delimiter.
isRawStringDelimBody(unsigned char c)155 LLVM_READONLY static inline bool isRawStringDelimBody(unsigned char c) {
156   using namespace charinfo;
157   return (InfoTable[c] & (CHAR_UPPER|CHAR_LOWER|CHAR_PERIOD|
158                           CHAR_DIGIT|CHAR_UNDER|CHAR_RAWDEL)) != 0;
159 }
160 
161 
162 /// Converts the given ASCII character to its lowercase equivalent.
163 ///
164 /// If the character is not an uppercase character, it is returned as is.
toLowercase(char c)165 LLVM_READONLY static inline char toLowercase(char c) {
166   if (isUppercase(c))
167     return c + 'a' - 'A';
168   return c;
169 }
170 
171 /// Converts the given ASCII character to its uppercase equivalent.
172 ///
173 /// If the character is not a lowercase character, it is returned as is.
toUppercase(char c)174 LLVM_READONLY static inline char toUppercase(char c) {
175   if (isLowercase(c))
176     return c + 'A' - 'a';
177   return c;
178 }
179 
180 
181 /// Return true if this is a valid ASCII identifier.
182 ///
183 /// Note that this is a very simple check; it does not accept '$' or UCNs as
184 /// valid identifier characters.
isValidIdentifier(StringRef S)185 LLVM_READONLY static inline bool isValidIdentifier(StringRef S) {
186   if (S.empty() || !isIdentifierHead(S[0]))
187     return false;
188 
189   for (StringRef::iterator I = S.begin(), E = S.end(); I != E; ++I)
190     if (!isIdentifierBody(*I))
191       return false;
192 
193   return true;
194 }
195 
196 } // end namespace clang
197 
198 #endif
199