1 // © 2018 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
3 //
4 // From the double-conversion library. Original license:
5 //
6 // Copyright 2012 the V8 project authors. All rights reserved.
7 // Redistribution and use in source and binary forms, with or without
8 // modification, are permitted provided that the following conditions are
9 // met:
10 //
11 //     * Redistributions of source code must retain the above copyright
12 //       notice, this list of conditions and the following disclaimer.
13 //     * Redistributions in binary form must reproduce the above
14 //       copyright notice, this list of conditions and the following
15 //       disclaimer in the documentation and/or other materials provided
16 //       with the distribution.
17 //     * Neither the name of Google Inc. nor the names of its
18 //       contributors may be used to endorse or promote products derived
19 //       from this software without specific prior written permission.
20 //
21 // THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
22 // "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
23 // LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
24 // A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
25 // OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
26 // SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
27 // LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
28 // DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
29 // THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
30 // (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
31 // OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
32 
33 // ICU PATCH: ifdef around UCONFIG_NO_FORMATTING
34 #include "unicode/utypes.h"
35 #if !UCONFIG_NO_FORMATTING
36 
37 #ifndef DOUBLE_CONVERSION_STRING_TO_DOUBLE_H_
38 #define DOUBLE_CONVERSION_STRING_TO_DOUBLE_H_
39 
40 // ICU PATCH: Customize header file paths for ICU.
41 
42 #include "double-conversion-utils.h"
43 
44 // ICU PATCH: Wrap in ICU namespace
45 U_NAMESPACE_BEGIN
46 
47 namespace double_conversion {
48 
49 class StringToDoubleConverter {
50  public:
51   // Enumeration for allowing octals and ignoring junk when converting
52   // strings to numbers.
53   enum Flags {
54     NO_FLAGS = 0,
55     ALLOW_HEX = 1,
56     ALLOW_OCTALS = 2,
57     ALLOW_TRAILING_JUNK = 4,
58     ALLOW_LEADING_SPACES = 8,
59     ALLOW_TRAILING_SPACES = 16,
60     ALLOW_SPACES_AFTER_SIGN = 32,
61     ALLOW_CASE_INSENSITIVITY = 64,
62     ALLOW_CASE_INSENSIBILITY = 64,  // Deprecated
63     ALLOW_HEX_FLOATS = 128,
64   };
65 
66   static const uc16 kNoSeparator = '\0';
67 
68   // Flags should be a bit-or combination of the possible Flags-enum.
69   //  - NO_FLAGS: no special flags.
70   //  - ALLOW_HEX: recognizes the prefix "0x". Hex numbers may only be integers.
71   //      Ex: StringToDouble("0x1234") -> 4660.0
72   //          In StringToDouble("0x1234.56") the characters ".56" are trailing
73   //          junk. The result of the call is hence dependent on
74   //          the ALLOW_TRAILING_JUNK flag and/or the junk value.
75   //      With this flag "0x" is a junk-string. Even with ALLOW_TRAILING_JUNK,
76   //      the string will not be parsed as "0" followed by junk.
77   //
78   //  - ALLOW_OCTALS: recognizes the prefix "0" for octals:
79   //      If a sequence of octal digits starts with '0', then the number is
80   //      read as octal integer. Octal numbers may only be integers.
81   //      Ex: StringToDouble("01234") -> 668.0
82   //          StringToDouble("012349") -> 12349.0  // Not a sequence of octal
83   //                                               // digits.
84   //          In StringToDouble("01234.56") the characters ".56" are trailing
85   //          junk. The result of the call is hence dependent on
86   //          the ALLOW_TRAILING_JUNK flag and/or the junk value.
87   //          In StringToDouble("01234e56") the characters "e56" are trailing
88   //          junk, too.
89   //  - ALLOW_TRAILING_JUNK: ignore trailing characters that are not part of
90   //      a double literal.
91   //  - ALLOW_LEADING_SPACES: skip over leading whitespace, including spaces,
92   //                          new-lines, and tabs.
93   //  - ALLOW_TRAILING_SPACES: ignore trailing whitespace.
94   //  - ALLOW_SPACES_AFTER_SIGN: ignore whitespace after the sign.
95   //       Ex: StringToDouble("-   123.2") -> -123.2.
96   //           StringToDouble("+   123.2") -> 123.2
97   //  - ALLOW_CASE_INSENSITIVITY: ignore case of characters for special values:
98   //      infinity and nan.
99   //  - ALLOW_HEX_FLOATS: allows hexadecimal float literals.
100   //      This *must* start with "0x" and separate the exponent with "p".
101   //      Examples: 0x1.2p3 == 9.0
102   //                0x10.1p0 == 16.0625
103   //      ALLOW_HEX and ALLOW_HEX_FLOATS are indendent.
104   //
105   // empty_string_value is returned when an empty string is given as input.
106   // If ALLOW_LEADING_SPACES or ALLOW_TRAILING_SPACES are set, then a string
107   // containing only spaces is converted to the 'empty_string_value', too.
108   //
109   // junk_string_value is returned when
110   //  a) ALLOW_TRAILING_JUNK is not set, and a junk character (a character not
111   //     part of a double-literal) is found.
112   //  b) ALLOW_TRAILING_JUNK is set, but the string does not start with a
113   //     double literal.
114   //
115   // infinity_symbol and nan_symbol are strings that are used to detect
116   // inputs that represent infinity and NaN. They can be null, in which case
117   // they are ignored.
118   // The conversion routine first reads any possible signs. Then it compares the
119   // following character of the input-string with the first character of
120   // the infinity, and nan-symbol. If either matches, the function assumes, that
121   // a match has been found, and expects the following input characters to match
122   // the remaining characters of the special-value symbol.
123   // This means that the following restrictions apply to special-value symbols:
124   //  - they must not start with signs ('+', or '-'),
125   //  - they must not have the same first character.
126   //  - they must not start with digits.
127   //
128   // If the separator character is not kNoSeparator, then that specific
129   // character is ignored when in between two valid digits of the significant.
130   // It is not allowed to appear in the exponent.
131   // It is not allowed to lead or trail the number.
132   // It is not allowed to appear twice next to each other.
133   //
134   // Examples:
135   //  flags = ALLOW_HEX | ALLOW_TRAILING_JUNK,
136   //  empty_string_value = 0.0,
137   //  junk_string_value = NaN,
138   //  infinity_symbol = "infinity",
139   //  nan_symbol = "nan":
140   //    StringToDouble("0x1234") -> 4660.0.
141   //    StringToDouble("0x1234K") -> 4660.0.
142   //    StringToDouble("") -> 0.0  // empty_string_value.
143   //    StringToDouble(" ") -> NaN  // junk_string_value.
144   //    StringToDouble(" 1") -> NaN  // junk_string_value.
145   //    StringToDouble("0x") -> NaN  // junk_string_value.
146   //    StringToDouble("-123.45") -> -123.45.
147   //    StringToDouble("--123.45") -> NaN  // junk_string_value.
148   //    StringToDouble("123e45") -> 123e45.
149   //    StringToDouble("123E45") -> 123e45.
150   //    StringToDouble("123e+45") -> 123e45.
151   //    StringToDouble("123E-45") -> 123e-45.
152   //    StringToDouble("123e") -> 123.0  // trailing junk ignored.
153   //    StringToDouble("123e-") -> 123.0  // trailing junk ignored.
154   //    StringToDouble("+NaN") -> NaN  // NaN string literal.
155   //    StringToDouble("-infinity") -> -inf.  // infinity literal.
156   //    StringToDouble("Infinity") -> NaN  // junk_string_value.
157   //
158   //  flags = ALLOW_OCTAL | ALLOW_LEADING_SPACES,
159   //  empty_string_value = 0.0,
160   //  junk_string_value = NaN,
161   //  infinity_symbol = NULL,
162   //  nan_symbol = NULL:
163   //    StringToDouble("0x1234") -> NaN  // junk_string_value.
164   //    StringToDouble("01234") -> 668.0.
165   //    StringToDouble("") -> 0.0  // empty_string_value.
166   //    StringToDouble(" ") -> 0.0  // empty_string_value.
167   //    StringToDouble(" 1") -> 1.0
168   //    StringToDouble("0x") -> NaN  // junk_string_value.
169   //    StringToDouble("0123e45") -> NaN  // junk_string_value.
170   //    StringToDouble("01239E45") -> 1239e45.
171   //    StringToDouble("-infinity") -> NaN  // junk_string_value.
172   //    StringToDouble("NaN") -> NaN  // junk_string_value.
173   //
174   //  flags = NO_FLAGS,
175   //  separator = ' ':
176   //    StringToDouble("1 2 3 4") -> 1234.0
177   //    StringToDouble("1  2") -> NaN // junk_string_value
178   //    StringToDouble("1 000 000.0") -> 1000000.0
179   //    StringToDouble("1.000 000") -> 1.0
180   //    StringToDouble("1.0e1 000") -> NaN // junk_string_value
181   StringToDoubleConverter(int flags,
182                           double empty_string_value,
183                           double junk_string_value,
184                           const char* infinity_symbol,
185                           const char* nan_symbol,
186                           uc16 separator = kNoSeparator)
flags_(flags)187       : flags_(flags),
188         empty_string_value_(empty_string_value),
189         junk_string_value_(junk_string_value),
190         infinity_symbol_(infinity_symbol),
191         nan_symbol_(nan_symbol),
192         separator_(separator) {
193   }
194 
195   // Performs the conversion.
196   // The output parameter 'processed_characters_count' is set to the number
197   // of characters that have been processed to read the number.
198   // Spaces than are processed with ALLOW_{LEADING|TRAILING}_SPACES are included
199   // in the 'processed_characters_count'. Trailing junk is never included.
200   double StringToDouble(const char* buffer,
201                         int length,
202                         int* processed_characters_count) const;
203 
204   // Same as StringToDouble above but for 16 bit characters.
205   double StringToDouble(const uc16* buffer,
206                         int length,
207                         int* processed_characters_count) const;
208 
209   // Same as StringToDouble but reads a float.
210   // Note that this is not equivalent to static_cast<float>(StringToDouble(...))
211   // due to potential double-rounding.
212   float StringToFloat(const char* buffer,
213                       int length,
214                       int* processed_characters_count) const;
215 
216   // Same as StringToFloat above but for 16 bit characters.
217   float StringToFloat(const uc16* buffer,
218                       int length,
219                       int* processed_characters_count) const;
220 
221  private:
222   const int flags_;
223   const double empty_string_value_;
224   const double junk_string_value_;
225   const char* const infinity_symbol_;
226   const char* const nan_symbol_;
227   const uc16 separator_;
228 
229   template <class Iterator>
230   double StringToIeee(Iterator start_pointer,
231                       int length,
232                       bool read_as_double,
233                       int* processed_characters_count) const;
234 
235   DOUBLE_CONVERSION_DISALLOW_IMPLICIT_CONSTRUCTORS(StringToDoubleConverter);
236 };
237 
238 }  // namespace double_conversion
239 
240 // ICU PATCH: Close ICU namespace
241 U_NAMESPACE_END
242 
243 #endif  // DOUBLE_CONVERSION_STRING_TO_DOUBLE_H_
244 #endif // ICU PATCH: close #if !UCONFIG_NO_FORMATTING
245