]> git.proxmox.com Git - ceph.git/blob - ceph/src/boost/libs/vmd/doc/vmd_identifier_subtyping.qbk
add subtree-ish sources for 12.0.3
[ceph.git] / ceph / src / boost / libs / vmd / doc / vmd_identifier_subtyping.qbk
1 [/
2 (C) Copyright Edward Diener 2011-2015
3 Distributed under the Boost Software License, Version 1.0.
4 (See accompanying file LICENSE_1_0.txt or copy at
5 http://www.boost.org/LICENSE_1_0.txt).
6 ]
7
8 [section:vmd_identifier_subtype Identifier subtypes]
9
10 Identifiers are the low-level data types which macro programmers
11 use to pass preprocessing data most often. As we have seen VMD
12 has a system for registering and detecting identifiers so that
13 they can be parsed as part of preprocessor data. This system also
14 includes comparing identifiers for equality or inequality using
15 BOOST_VMD_EQUAL/BOOST_VMD_NOT_EQUAL and matching identifiers
16 using identifier modifiers in BOOST_VMD_IS_IDENTIFIER and
17 BOOST_VMD_ELEM. Together these facilities provide a rich set
18 of functionality for handling identifiers in macros.
19
20 Both numbers and v-types are subtypes of identifiers, and can
21 both be individually recognized as data types of their own or
22 worked with as identifiers using the identifier facilities
23 already mentioned. Numbers, in particular, also have a rich set
24 of functionality within the Boost PP library. As subtypes numbers
25 and v-types can be used as filter modifiers and can be returned
26 as specific types either when invoking BOOST_VMD_GET_TYPE
27 or when using return type modifiers. Furthermore VMD recognizes
28 their individual v-types, BOOST_VMD_TYPE_NUMBER and
29 BOOST_VMD_TYPE_TYPE, as VMD data when parsing sequences.
30
31 It is possible for the end-user to define his own identifier subtype.
32 This is called a "user-defined subtype". Once a user-define subtype is
33 created all the generic type facilities of VMD which subtypes such as
34 a number or a v-type possess is automatically available for that
35 user-defined subtype.
36
37 [heading Defining a subtype]
38
39 In order to define a user-defined subtype a number of steps need to be followed.
40 These steps will be explained in detail further below:
41
42 # Register and pre-detect all identifiers of that subtype.
43 # Register and pre-detect a v-type name for that subtype.
44 # Subset register all identifiers of the subtype.
45 # Subset register the v-type name for the subtype.
46
47 When we do the above, it is best to put all the macros in a single
48 header file and always include that header file when we work generically
49 with our user-defined subtype.
50
51 [heading Register and pre-detect all identifiers of that subtype]
52
53 Registering and pre-detecting all of the identifiers of that subtype
54 is exactly the same as registering and pre-detecting any identifier.
55
56 Let's create some identifiers based for use in the mythical "udef" library.
57 We will put all our macros in the header file udef_vmd_macros.hpp.
58
59 We will need distinct names for the identifiers in our library, so we will
60 append UDEF_ to our identifier names to make them unique. Our udef library
61 deals in geometrical shapes so we will create a user-defined subtype which
62 consists of identifiers for the various shapes our udef library can
63 manipulate in their macros. So our identifier registrations and pre-detections
64 placed in our header file will be:
65
66 #define BOOST_VMD_REGISTER_UDEF_CIRCLE (UDEF_CIRCLE)
67 #define BOOST_VMD_REGISTER_UDEF_SQUARE (UDEF_SQUARE)
68 #define BOOST_VMD_REGISTER_UDEF_TRIANGLE (UDEF_TRIANGLE)
69 #define BOOST_VMD_REGISTER_UDEF_HEXAGON (UDEF_HEXAGON)
70
71 #define BOOST_VMD_DETECT_UDEF_CIRCLE_UDEF_CIRCLE
72 #define BOOST_VMD_DETECT_UDEF_SQUARE_UDEF_SQUARE
73 #define BOOST_VMD_DETECT_UDEF_TRIANGLE_UDEF_TRIANGLE
74 #define BOOST_VMD_DETECT_UDEF_HEXAGON_UDEF_HEXAGON
75
76 [heading Register and pre-detect a v-type name for that subtype]
77
78 We need to create a unique v-type name for our user-defined subtype.
79 The name does not have to begin with BOOST_VMD_TYPE_ but it should be
80 unique. Since BOOST_VMD_TYPE_ is the common beginning of all v-types
81 we will use it for consistency but will append to it UDEF_SHAPES to
82 give it a uniqueness which should not be duplicated:
83
84 #define BOOST_VMD_REGISTER_BOOST_VMD_TYPE_UDEF_SHAPES (BOOST_VMD_TYPE_UDEF_SHAPES)
85
86 #define BOOST_VMD_DETECT_BOOST_VMD_TYPE_UDEF_SHAPES_BOOST_VMD_TYPE_UDEF_SHAPES
87
88 [heading Subtype register all identifiers of the subtype]
89
90 The macro to register an identifier subset starts with BOOST_VMD_SUBTYPE_REGISTER_
91 and you append to it each identifier in the subset. This is very much like the
92 way you use the BOOST_VMD_REGISTER_ macro. The difference is that unlike the
93 BOOST_VMD_REGISTER_ macro, which expands to a tuple whose single element is the
94 identifier, the BOOST_VMD_SUBTYPE_REGISTER_ expands to a tuple of two elements
95 where the first element is the subtype v-type and the second element is the identifier.
96
97 For our udef user-defined subtype this would be:
98
99 #define BOOST_VMD_SUBTYPE_REGISTER_UDEF_CIRCLE (BOOST_VMD_TYPE_UDEF_SHAPES,UDEF_CIRCLE)
100 #define BOOST_VMD_SUBTYPE_REGISTER_UDEF_SQUARE (BOOST_VMD_TYPE_UDEF_SHAPES,UDEF_SQUARE)
101 #define BOOST_VMD_SUBTYPE_REGISTER_UDEF_TRIANGLE (BOOST_VMD_TYPE_UDEF_SHAPES,UDEF_TRIANGLE)
102 #define BOOST_VMD_SUBTYPE_REGISTER_UDEF_HEXAGON (BOOST_VMD_TYPE_UDEF_SHAPES,UDEF_HEXAGON)
103
104 [heading Subtype register the v-type name for the subtype]
105
106 Doing a subset register of the actual udef v-type is fairly easy once we understand
107 how to register an identifier subset. The only particular thing to realize is
108 that the type of any v-type is the v-type BOOST_VMD_TYPE_TYPE. So our subset
109 register of our new v-type BOOST_VMD_TYPE_UDEF_SHAPES is:
110
111 #define BOOST_VMD_SUBTYPE_REGISTER_BOOST_VMD_TYPE_UDEF_SHAPES (BOOST_VMD_TYPE_TYPE,BOOST_VMD_TYPE_UDEF_SHAPES)
112
113 [heading Using our identifier subset]
114
115 Once we have added all of the above object-like macros for defining our user-defined
116 subtype to the udef_vmd_macros.hpp header file we have a new data type which we can
117 use generically just like we can use numbers or v-types generically. It is important
118 to include the header udef_vmd_macros.hpp in some translation unit whenever we need
119 the VMD functionality for our new data type. So in our examples we will assume that
120 an '#include udef_vmd_macros.hpp' precedes each example.
121
122 #include <boost/vmd/get_type.hpp>
123
124 #define A_SEQUENCE UDEF_SQUARE
125 #define A_SEQUENCE2 217
126 #define A_SEQUENCE3 BOOST_VMD_TYPE_UDEF_SHAPES
127 #define A_SEQUENCE4 BOOST_VMD_TYPE_NUMBER
128
129 BOOST_VMD_GET_TYPE(A_SEQUENCE) will return 'BOOST_VMD_TYPE_UDEF_SHAPES'
130 BOOST_VMD_GET_TYPE(A_SEQUENCE2) will return 'BOOST_VMD_TYPE_NUMBER'
131 BOOST_VMD_GET_TYPE(A_SEQUENCE3) will return 'BOOST_VMD_TYPE_TYPE'
132 BOOST_VMD_GET_TYPE(A_SEQUENCE4) will return 'BOOST_VMD_TYPE_TYPE'
133
134 Here we see that when we use our BOOST_VMD_GET_TYPE macro on a single-element
135 sequence which is one of our user-defined subtype values we correctly get back
136 our user-defined subtype's v-type, just like we do when we ask for the type of a number. Also
137 when we use our BOOST_VMD_GET_TYPE macro on our user-defined subtype's v-type itself we correctly
138 get back the type of all v-types, which is BOOST_VMD_TYPE_TYPE, just like we do
139 when we ask for the type of the v-type of a number.
140
141 #include <boost/vmd/elem.hpp>
142
143 #define A_SEQUENCE5 (1,2) UDEF_TRIANGLE
144
145 BOOST_VMD_ELEM(1,A_SEQUENCE5,BOOST_VMD_RETURN_TYPE) will return '(BOOST_VMD_TYPE_UDEF_SHAPES,UDEF_TRIANGLE)'
146 BOOST_VMD_ELEM(0,A_SEQUENCE5,BOOST_VMD_RETURN_TYPE) will return '(BOOST_VMD_TYPE_TUPLE,(1,2))'
147
148 Here we see that we can use the return type modifier to get back both the type
149 and the value in a two-element tuple for our user-defined subtype just as we so
150 for any other type.
151
152 #include <boost/vmd/equal.hpp>
153
154 #define A_SEQUENCE6 UDEF_CIRCLE
155 #define A_SEQUENCE7 168
156
157 BOOST_VMD_EQUAL(A_SEQUENCE6,UDEF_CIRCLE,BOOST_VMD_TYPE_UDEF_SHAPES) will return '1'
158 BOOST_VMD_EQUAL(A_SEQUENCE6,UDEF_CIRCLE,BOOST_VMD_TYPE_LIST) will return '0'
159 BOOST_VMD_EQUAL(A_SEQUENCE7,168,BOOST_VMD_TYPE_NUMBER) will return '1'
160 BOOST_VMD_EQUAL(A_SEQUENCE7,168,BOOST_VMD_TYPE_SEQ) will return '0'
161
162 Here we can see that we can use the filter modifier with our user-defined subtype's v-type
163 just as we can do with any other v-type, such as the number v-type.
164
165 In all respects once we define our subtype and provide those definitions in
166 a header file, our user-defined subtype acts like any other v-type in our system.
167 Since VMD functionality is largely based on being able to recognize the type of
168 data in macro input being able to define another 'type', as an identifier subtype,
169 which VMD understands has value for the macro programmer.
170
171 [heading Uniqueness of identifier subtype values and v-type]
172
173 When we define a new identifier subtype we need to be careful that
174 the values of that subtype and its actual v-type are unique identifiers
175 within any translation unit. This is the main difference between just
176 defining identifiers and defining an identifier subtype.
177
178 Recall that when we just register and pre-detect identifiers we will have
179 no problems if the same identifiers already have been registered and pre-detected
180 within the same translation unit. This is because we are just redefining the
181 exact same macro if this is the case.
182
183 But with identifier subtypes, when we use the BOOST_VMD_SUBTYPE_REGISTER_ macro
184 to associate our subtype's v-type with our subtype identifiers, we will have
185 problems if someone else has also defined an identifier subtype using the same
186 identifiers as we use since we will be redefining the same object-like macro name
187 with a different expansion. Even if someone else has registered/pre-detected an
188 identifier we are using for out subtype without defining a subtype based on that
189 identifier we will be causing a problem defining our subtype because VMD macros which
190 generically return the type of a sequence or sequence element will return our
191 subtype as the type rather than just BOOST_VMD_TYPE_IDENTIFIER which some programmer
192 might expect.
193
194 The gist of this is that if we define a user-defined subtype its identifiers need
195 to be unique within a given translation unit, and yet unique names make it harder
196 for an end-user to use macros more naturally. In our given example with the mythical
197 udef library we used identifiers such as 'UDEF_CIRCLE' etc. instead of the more natural
198 sounding CIRCLE. So with user-defined identifier subtypes we have a tradeoff; we need
199 unique identifier names both for our subtype identifiers and the v-type for our
200 subtype identifiers so as not to conflict with others who might be using identifier
201 subtypes, but those unique names might make using macros less "natural" On the other
202 hand, just registering/pre-detecting identifiers has no such problem. This is an
203 issue of which any user, looking to create his own data type using VMD by defining
204 user-defined subtypes, should be aware.
205
206 [endsect]