xref: /openbsd/gnu/usr.bin/binutils/bfd/doc/hash.texi (revision 7b36286a)
1@section Hash Tables
2@cindex Hash tables
3BFD provides a simple set of hash table functions.  Routines
4are provided to initialize a hash table, to free a hash table,
5to look up a string in a hash table and optionally create an
6entry for it, and to traverse a hash table.  There is
7currently no routine to delete an string from a hash table.
8
9The basic hash table does not permit any data to be stored
10with a string.  However, a hash table is designed to present a
11base class from which other types of hash tables may be
12derived.  These derived types may store additional information
13with the string.  Hash tables were implemented in this way,
14rather than simply providing a data pointer in a hash table
15entry, because they were designed for use by the linker back
16ends.  The linker may create thousands of hash table entries,
17and the overhead of allocating private data and storing and
18following pointers becomes noticeable.
19
20The basic hash table code is in @code{hash.c}.
21
22@menu
23* Creating and Freeing a Hash Table::
24* Looking Up or Entering a String::
25* Traversing a Hash Table::
26* Deriving a New Hash Table Type::
27@end menu
28
29@node Creating and Freeing a Hash Table, Looking Up or Entering a String, Hash Tables, Hash Tables
30@subsection Creating and freeing a hash table
31@findex bfd_hash_table_init
32@findex bfd_hash_table_init_n
33To create a hash table, create an instance of a @code{struct
34bfd_hash_table} (defined in @code{bfd.h}) and call
35@code{bfd_hash_table_init} (if you know approximately how many
36entries you will need, the function @code{bfd_hash_table_init_n},
37which takes a @var{size} argument, may be used).
38@code{bfd_hash_table_init} returns @code{FALSE} if some sort of
39error occurs.
40
41@findex bfd_hash_newfunc
42The function @code{bfd_hash_table_init} take as an argument a
43function to use to create new entries.  For a basic hash
44table, use the function @code{bfd_hash_newfunc}.  @xref{Deriving
45a New Hash Table Type}, for why you would want to use a
46different value for this argument.
47
48@findex bfd_hash_allocate
49@code{bfd_hash_table_init} will create an objalloc which will be
50used to allocate new entries.  You may allocate memory on this
51objalloc using @code{bfd_hash_allocate}.
52
53@findex bfd_hash_table_free
54Use @code{bfd_hash_table_free} to free up all the memory that has
55been allocated for a hash table.  This will not free up the
56@code{struct bfd_hash_table} itself, which you must provide.
57
58@node Looking Up or Entering a String, Traversing a Hash Table, Creating and Freeing a Hash Table, Hash Tables
59@subsection Looking up or entering a string
60@findex bfd_hash_lookup
61The function @code{bfd_hash_lookup} is used both to look up a
62string in the hash table and to create a new entry.
63
64If the @var{create} argument is @code{FALSE}, @code{bfd_hash_lookup}
65will look up a string.  If the string is found, it will
66returns a pointer to a @code{struct bfd_hash_entry}.  If the
67string is not found in the table @code{bfd_hash_lookup} will
68return @code{NULL}.  You should not modify any of the fields in
69the returns @code{struct bfd_hash_entry}.
70
71If the @var{create} argument is @code{TRUE}, the string will be
72entered into the hash table if it is not already there.
73Either way a pointer to a @code{struct bfd_hash_entry} will be
74returned, either to the existing structure or to a newly
75created one.  In this case, a @code{NULL} return means that an
76error occurred.
77
78If the @var{create} argument is @code{TRUE}, and a new entry is
79created, the @var{copy} argument is used to decide whether to
80copy the string onto the hash table objalloc or not.  If
81@var{copy} is passed as @code{FALSE}, you must be careful not to
82deallocate or modify the string as long as the hash table
83exists.
84
85@node Traversing a Hash Table, Deriving a New Hash Table Type, Looking Up or Entering a String, Hash Tables
86@subsection Traversing a hash table
87@findex bfd_hash_traverse
88The function @code{bfd_hash_traverse} may be used to traverse a
89hash table, calling a function on each element.  The traversal
90is done in a random order.
91
92@code{bfd_hash_traverse} takes as arguments a function and a
93generic @code{void *} pointer.  The function is called with a
94hash table entry (a @code{struct bfd_hash_entry *}) and the
95generic pointer passed to @code{bfd_hash_traverse}.  The function
96must return a @code{boolean} value, which indicates whether to
97continue traversing the hash table.  If the function returns
98@code{FALSE}, @code{bfd_hash_traverse} will stop the traversal and
99return immediately.
100
101@node Deriving a New Hash Table Type, , Traversing a Hash Table, Hash Tables
102@subsection Deriving a new hash table type
103Many uses of hash tables want to store additional information
104which each entry in the hash table.  Some also find it
105convenient to store additional information with the hash table
106itself.  This may be done using a derived hash table.
107
108Since C is not an object oriented language, creating a derived
109hash table requires sticking together some boilerplate
110routines with a few differences specific to the type of hash
111table you want to create.
112
113An example of a derived hash table is the linker hash table.
114The structures for this are defined in @code{bfdlink.h}.  The
115functions are in @code{linker.c}.
116
117You may also derive a hash table from an already derived hash
118table.  For example, the a.out linker backend code uses a hash
119table derived from the linker hash table.
120
121@menu
122* Define the Derived Structures::
123* Write the Derived Creation Routine::
124* Write Other Derived Routines::
125@end menu
126
127@node Define the Derived Structures, Write the Derived Creation Routine, Deriving a New Hash Table Type, Deriving a New Hash Table Type
128@subsubsection Define the derived structures
129You must define a structure for an entry in the hash table,
130and a structure for the hash table itself.
131
132The first field in the structure for an entry in the hash
133table must be of the type used for an entry in the hash table
134you are deriving from.  If you are deriving from a basic hash
135table this is @code{struct bfd_hash_entry}, which is defined in
136@code{bfd.h}.  The first field in the structure for the hash
137table itself must be of the type of the hash table you are
138deriving from itself.  If you are deriving from a basic hash
139table, this is @code{struct bfd_hash_table}.
140
141For example, the linker hash table defines @code{struct
142bfd_link_hash_entry} (in @code{bfdlink.h}).  The first field,
143@code{root}, is of type @code{struct bfd_hash_entry}.  Similarly,
144the first field in @code{struct bfd_link_hash_table}, @code{table},
145is of type @code{struct bfd_hash_table}.
146
147@node Write the Derived Creation Routine, Write Other Derived Routines, Define the Derived Structures, Deriving a New Hash Table Type
148@subsubsection Write the derived creation routine
149You must write a routine which will create and initialize an
150entry in the hash table.  This routine is passed as the
151function argument to @code{bfd_hash_table_init}.
152
153In order to permit other hash tables to be derived from the
154hash table you are creating, this routine must be written in a
155standard way.
156
157The first argument to the creation routine is a pointer to a
158hash table entry.  This may be @code{NULL}, in which case the
159routine should allocate the right amount of space.  Otherwise
160the space has already been allocated by a hash table type
161derived from this one.
162
163After allocating space, the creation routine must call the
164creation routine of the hash table type it is derived from,
165passing in a pointer to the space it just allocated.  This
166will initialize any fields used by the base hash table.
167
168Finally the creation routine must initialize any local fields
169for the new hash table type.
170
171Here is a boilerplate example of a creation routine.
172@var{function_name} is the name of the routine.
173@var{entry_type} is the type of an entry in the hash table you
174are creating.  @var{base_newfunc} is the name of the creation
175routine of the hash table type your hash table is derived
176from.
177
178
179@example
180struct bfd_hash_entry *
181@var{function_name} (entry, table, string)
182     struct bfd_hash_entry *entry;
183     struct bfd_hash_table *table;
184     const char *string;
185@{
186  struct @var{entry_type} *ret = (@var{entry_type} *) entry;
187
188 /* Allocate the structure if it has not already been allocated by a
189    derived class.  */
190  if (ret == (@var{entry_type} *) NULL)
191    @{
192      ret = ((@var{entry_type} *)
193             bfd_hash_allocate (table, sizeof (@var{entry_type})));
194      if (ret == (@var{entry_type} *) NULL)
195        return NULL;
196    @}
197
198 /* Call the allocation method of the base class.  */
199  ret = ((@var{entry_type} *)
200        @var{base_newfunc} ((struct bfd_hash_entry *) ret, table, string));
201
202 /* Initialize the local fields here.  */
203
204  return (struct bfd_hash_entry *) ret;
205@}
206@end example
207@strong{Description}@*
208The creation routine for the linker hash table, which is in
209@code{linker.c}, looks just like this example.
210@var{function_name} is @code{_bfd_link_hash_newfunc}.
211@var{entry_type} is @code{struct bfd_link_hash_entry}.
212@var{base_newfunc} is @code{bfd_hash_newfunc}, the creation
213routine for a basic hash table.
214
215@code{_bfd_link_hash_newfunc} also initializes the local fields
216in a linker hash table entry: @code{type}, @code{written} and
217@code{next}.
218
219@node Write Other Derived Routines, , Write the Derived Creation Routine, Deriving a New Hash Table Type
220@subsubsection Write other derived routines
221You will want to write other routines for your new hash table,
222as well.
223
224You will want an initialization routine which calls the
225initialization routine of the hash table you are deriving from
226and initializes any other local fields.  For the linker hash
227table, this is @code{_bfd_link_hash_table_init} in @code{linker.c}.
228
229You will want a lookup routine which calls the lookup routine
230of the hash table you are deriving from and casts the result.
231The linker hash table uses @code{bfd_link_hash_lookup} in
232@code{linker.c} (this actually takes an additional argument which
233it uses to decide how to return the looked up value).
234
235You may want a traversal routine.  This should just call the
236traversal routine of the hash table you are deriving from with
237appropriate casts.  The linker hash table uses
238@code{bfd_link_hash_traverse} in @code{linker.c}.
239
240These routines may simply be defined as macros.  For example,
241the a.out backend linker hash table, which is derived from the
242linker hash table, uses macros for the lookup and traversal
243routines.  These are @code{aout_link_hash_lookup} and
244@code{aout_link_hash_traverse} in aoutx.h.
245
246