1 // Copyright 2011 the V8 project authors. All rights reserved.
2 // Use of this source code is governed by a BSD-style license that can be
3 // found in the LICENSE file.
12 // Return the shortest correct representation.
13 // For example the output of 0.299999999999999988897 is (the less accurate but
16 // Return a fixed number of digits after the decimal point.
17 // For instance fixed(0.1, 4) becomes 0.1000
18 // If the input number is big, the output will be big.
20 // Return a fixed number of digits, no matter what the exponent is.
24 // The maximal length of digits a double can have in base 10.
25 // Note that DoubleToAscii null-terminates its input. So the given buffer should
26 // be at least kBase10MaximalLength + 1 characters long.
27 const int kBase10MaximalLength = 17;
29 // Converts the given double 'v' to ASCII.
30 // The result should be interpreted as buffer * 10^(point-length).
32 // The output depends on the given mode:
33 // - SHORTEST: produce the least amount of digits for which the internal
34 // identity requirement is still satisfied. If the digits are printed
35 // (together with the correct exponent) then reading this number will give
36 // 'v' again. The buffer will choose the representation that is closest to
37 // 'v'. If there are two at the same distance, than the one farther away
38 // from 0 is chosen (halfway cases - ending with 5 - are rounded up).
39 // In this mode the 'requested_digits' parameter is ignored.
40 // - FIXED: produces digits necessary to print a given number with
41 // 'requested_digits' digits after the decimal point. The produced digits
42 // might be too short in which case the caller has to fill the gaps with '0's.
43 // Example: toFixed(0.001, 5) is allowed to return buffer="1", point=-2.
44 // Halfway cases are rounded towards +/-Infinity (away from 0). The call
45 // toFixed(0.15, 2) thus returns buffer="2", point=0.
46 // The returned buffer may contain digits that would be truncated from the
47 // shortest representation of the input.
48 // - PRECISION: produces 'requested_digits' where the first digit is not '0'.
49 // Even though the length of produced digits usually equals
50 // 'requested_digits', the function is allowed to return fewer digits, in
51 // which case the caller has to fill the missing digits with '0's.
52 // Halfway cases are again rounded away from 0.
53 // 'DoubleToAscii' expects the given buffer to be big enough to hold all digits
54 // and a terminating null-character. In SHORTEST-mode it expects a buffer of
55 // at least kBase10MaximalLength + 1. Otherwise, the size of the output is
56 // limited to requested_digits digits plus the null terminator.
57 void DoubleToAscii(double v, DtoaMode mode, int requested_digits,
58 Vector<char> buffer, int* sign, int* length, int* point);
60 } } // namespace v8::internal