b6b9d458 |
1 | .\" -*-nroff-*- |
2 | .de VS |
3 | .sp 1 |
4 | .RS 5 |
5 | .nf |
6 | .ft B |
7 | .. |
8 | .de VE |
9 | .ft R |
10 | .fi |
11 | .RE |
12 | .sp 1 |
13 | .. |
14 | .TH sym 3mLib "8 May 1999" mLib |
15 | .SH NAME |
16 | sym \- symbol table manager |
17 | .SH SYNOPSIS |
18 | .nf |
19 | .B "#include <mLib/sym.h>" |
20 | |
21 | .BI "void sym_create(sym_table *" t ); |
22 | .BI "void sym_destroy(sym_table *" t ); |
23 | |
24 | .BI "char *SYM_NAME(void *" p ); |
25 | .BI "void *sym_find(sym_table *" t , |
26 | .BI " const char *" n ", long " l , |
27 | .BI " size_t " sz ", unsigned *" f ); |
28 | .BI "void sym_remove(sym_table *" t ", void *" b ); |
29 | |
30 | .BI "void sym_mkiter(sym_iter *" i ", sym_table *" t ); |
31 | .BI "void *sym_next(sym_iter *" i ); |
32 | .fi |
33 | .SH "HOW IT WORKS" |
34 | The |
35 | .B sym |
36 | functions implement a data structure often described as a dictionary, a |
37 | finite map, an associative array, or a symbol table. It associates |
38 | .I values |
39 | with |
40 | .I keys |
41 | such that the value corresponding to a given key can be found quickly. |
42 | Additionally, all stored associations can be enumerated. |
43 | .PP |
44 | The interface provides an |
45 | .I intrusive |
46 | symbol table. The data objects stored in the table must include a small |
47 | header used by the symbol table manager. This reduces the amount of |
48 | pointer fiddling that needs to be done, and in practice doesn't seem to |
49 | be much of a problem. It's also fairly easy to construct a |
50 | non-intrusive interface if you really want one. |
51 | .PP |
52 | There are three main data structures involved in the interface: |
53 | .TP |
54 | .B sym_table |
55 | Keeps track of the information associated with a particular table. |
56 | .TP |
57 | .B sym_base |
58 | The header which must be attached to the front of all the value |
59 | objects. |
60 | .TP |
61 | .B sym_iter |
62 | An iterator object, used for enumerating all of the associations stored |
63 | in a symbol table. |
64 | .PP |
65 | All of the above data structures should be considered |
66 | .IR opaque : |
67 | don't try looking inside. Representations have changed in the past, and |
68 | they may change again in the future. |
69 | .SH "CREATION AND DESTRUCTION" |
70 | The |
71 | .B sym_table |
72 | object itself needs to be allocated by the caller. It is initialized by |
73 | passing it to the function |
74 | .BR sym_create . |
75 | After initialization, the table contains no entries. |
76 | .PP |
77 | Initializing a symbol table involves allocating some memory. If this |
78 | allocation failes, an |
79 | .B EXC_NOMEM |
80 | exception is raised. |
81 | .PP |
82 | When a symbol table is no longer needed, the memory occupied by the |
83 | values and other maintenance structures can be reclaimed by calling |
84 | .BR sym_destroy . |
85 | Any bits of user data attached to the symbol table values should have |
86 | previously been destroyed. |
87 | .SH "ADDING, SEARCHING AND REMOVING" |
88 | Most of the actual work is done by the function |
89 | .BR sym_find . |
90 | It does both lookup and creation, depending on its arguments. To do its |
91 | job, it needs to know the following bits of information: |
92 | .TP |
93 | .I t |
94 | A pointer to a symbol table to manipulate. |
95 | .TP |
96 | .I n |
97 | The address of the |
98 | .I key |
99 | to look up or create. Usually this will be a simple text string, |
100 | although it can actually be any arbitrary binary data. |
101 | .TP |
102 | .I l |
103 | The length of the key. If this is \-1, |
104 | .B sym_find |
105 | assumes that the key is a null-terminated string, and calculates its |
106 | length itself. |
107 | .TP |
108 | .I sz |
109 | The size of the value block to allocate if the key could not be found. |
110 | If this is zero, no value is allocated, and a null pointer is returned |
111 | to indicate an unsuccessful lookup. |
112 | .TP |
113 | .I f |
114 | The address of a `found' flag to set. This is an output parameter. On |
115 | exit, |
116 | .B sym_find |
117 | will set the value of |
118 | .BI * f |
119 | to zero if the key could not be found, or nonzero if it was found. This |
120 | can be used to tell whether the value returned has been newly allocated, |
121 | or whether it was already in the table. |
122 | .PP |
123 | The macro |
124 | .B SYM_NAME |
125 | will return the key associated with a given value block. You must use |
126 | this macro rather than fiddling inside the |
127 | .B sym_base |
128 | structure. |
129 | .PP |
130 | A symbol can be removed from the table by calling |
131 | .BR sym_remove , |
132 | passing the symbol table itself, and the value block that needs |
133 | removing. |
134 | .SH ENUMERATION |
135 | Enumerating the values in a symbol table is fairly simple. Allocate a |
136 | .B sym_iter |
137 | object from somewhere. Attach it to a symbol table by calling |
138 | .BR sym_mkiter , |
139 | and passing in the addresses of the iterator and the symbol table. |
140 | Then, each call to |
141 | .B sym_next |
142 | will return a different value from the symbol table, until all of them |
143 | have been enumerated, at which point, |
144 | .B sym_next |
145 | returns a null pointer. |
146 | .PP |
147 | It's safe to remove the symbol you've just been returned by |
148 | .BR sym_next . |
149 | However, it's not safe to remove any other symbol. So don't do that. |
150 | .PP |
151 | When you've finished with an iterator, it's safe to just throw it away. |
152 | You don't need to call any functions beforehand. |
153 | .SH "USE IN PRACTICE" |
154 | In normal use, the keys are simple strings (usually identifiers from |
155 | some language), and the values are nontrivial structures providing |
156 | information about types and values. |
157 | .PP |
158 | In this case, you'd define something like the following structure for |
159 | your values: |
160 | .VS |
161 | typedef struct val { |
162 | sym_base _base; /* Symbol header */ |
163 | unsigned type; /* Type of this symbol */ |
164 | int dispoff; /* Which display variable is in */ |
165 | size_t frameoff; /* Offset of variable in frame */ |
166 | } val; |
167 | .VE |
168 | Given a pointer |
169 | .I v |
170 | to a |
171 | .BR val , |
172 | you can find the variable's name by calling |
173 | .BI SYM_NAME( v )\fR. |
174 | .PP |
175 | You can look up a name in the table by saying something like: |
176 | .VS |
177 | val *v = sym_find(t, name, -1, 0, 0); |
178 | if (!v) |
179 | error("unknown variable `%s'", name); |
180 | .VE |
181 | You can add in a new variable by saying something like |
182 | .VS |
183 | unsigned f; |
184 | val *v = sym_find(t, name, -1, sizeof(val), &f); |
185 | if (f) |
186 | error("variable `%s' already exists", name); |
187 | /* fill in v */ |
188 | .VE |
189 | You can examine all the variables in your symbol table by saying |
190 | something like: |
191 | .VS |
192 | sym_iter i; |
193 | val *v; |
194 | |
195 | for (sym_mkiter(&i, t); (v = sym_next(&i)) != 0; ) { |
196 | /* ... */ |
197 | } |
198 | .VE |
199 | That ought to be enough examples to be getting on with. |
200 | .SH CAVEATS |
201 | The symbol table manager requires the suballocator (see |
202 | .BR sub (3mLib) |
203 | for details). You must ensure that |
204 | .B sub_init |
205 | has been called before using any symbol tables in your program. |
206 | .SH IMPLEMENTATION |
207 | The symbol table is an extensible hashtable, using a 32-bit CRC as the |
208 | hash function. The hash chains are kept very short (probably too short, |
209 | actually). Every time a symbol is found, its block is promoted to the |
210 | front of its bin chain so it gets found faster next time. |
211 | .SH SEE ALSO |
212 | .BR sub (3mLib). |
213 | .SH AUTHOR |
214 | Mark Wooding, <mdw@nsict.org> |