1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
|
[/==============================================================================
Copyright (C) 2001-2008 Joel de Guzman
Copyright (C) 2001-2008 Hartmut Kaiser
Distributed under the Boost Software License, Version 1.0. (See accompanying
file LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)
===============================================================================/]
[section:basics Parser Basics]
[heading Lazy Argument]
Some parsers (e.g. primitives and non-terminals) may take in additional
attributes. Such parsers take the form:
p(a1, a2,..., aN)
where `p` is a parser. Each of the arguments (a1 ... aN) can either be an
immediate value, or a function, `f`, with signature:
T f(Unused, Context)
where `T`, the function's return value, is compatible with the argument
type expected and `Context` is the parser's __context__ type (The first
argument is __unused__ to make the `Context` the second argument. This
is done for uniformity with __actions__).
[heading Character Encoding Namespace]
Some parsers need to know which character set a `char` or `wchar_t` is
operating on. For example, the `alnum` parser works differently with
ISO8859.1 and ASCII encodings. Where necessary, Spirit encodes (tags)
the parser with the character set.
We have a namespace for each character set Spirit will be supporting.
That includes `ascii`, `iso8859_1`, `standard` and `standard_wide` (and
in the future, `unicode`). In each of the character encoding namespaces,
we place tagged versions of parsers such as `alnum`, `space` etc.
Example:
using boost::spirit::ascii::space; // use the ASCII space parser
Namespaces:
* boost::spirit::ascii
* boost::spirit::iso8859_1
* boost::spirit::standard
* boost::spirit::standard_wide
For ease of use, the components in this namespaces are also brought into
the qi sub-namespaces with the same names:
* boost::spirit::qi::ascii
* boost::spirit::qi::iso8859_1
* boost::spirit::qi::standard
* boost::spirit::qi::standard_wide
[heading Examples]
All sections in the reference present some real world examples. The
examples use a common test harness to keep the example code as minimal
and direct to the point as possible. The test harness is presented
below.
Some includes:
[reference_includes]
Our test functions:
These functions test the parsers without attributes.
[reference_test]
These functions test the parsers with user supplied attributes.
[reference_test_attr]
The `print_info` utility function prints information contained in the
__info__ class.
[reference_print_info]
[heading String]
[heading Header]
// forwards to <boost/spirit/home/support/string_traits.hpp>
#include <boost/spirit/support_string_traits.hpp>
A string can be any object `s`, of type, `S`, that satisfies the
following expression traits:
[table
[[Expression] [Semantics]]
[[`boost::spirit::traits::is_string<S>`] [Metafunction that evaluates to `mpl::true_` if
a certain type, `S` is a string, `mpl::false_`
otherwise (See __mpl_boolean_constant__).]]
[[`boost::spirit::traits::char_type_of<S>`] [Metafunction that returns the underlying
char type of a string type, `S`.]]
[[`boost::spirit::traits::get_c_string(s)`] [Function that returns the underlying
raw C-string from `s`.]]
[[`boost::spirit::traits::get_begin(s)`] [Function that returns an __stl__ iterator from `s`
that points to the beginning the string.]]
[[`boost::spirit::traits::get_end(s)`] [Function that returns an __stl__ iterator from `s`
that points to the end of the string.]]
]
[heading Models]
Predefined models include:
* any literal string, e.g. "Hello, World",
* a pointer/reference to a null-terminated array of characters
* a `std::basic_string<Char>`
The namespace `boost::spirit::traits` is open for users to provide their
own specializations. The customization points implemented by __qi__ usable
to customize the behavior of parsers are described in the section
__sec_customization_points__.
[endsect]
|
