Commit | Line | Data |
---|---|---|
b8d4c8d0 GM |
1 | @c -*-texinfo-*- |
2 | @c This is part of the GNU Emacs Lisp Reference Manual. | |
ba318903 | 3 | @c Copyright (C) 1999, 2001-2014 Free Software Foundation, Inc. |
b8d4c8d0 | 4 | @c See the file elisp.texi for copying conditions. |
ecc6530d | 5 | @node Hash Tables |
b8d4c8d0 GM |
6 | @chapter Hash Tables |
7 | @cindex hash tables | |
8 | @cindex lookup tables | |
9 | ||
10 | A hash table is a very fast kind of lookup table, somewhat like an | |
11 | alist (@pxref{Association Lists}) in that it maps keys to | |
12 | corresponding values. It differs from an alist in these ways: | |
13 | ||
14 | @itemize @bullet | |
15 | @item | |
16 | Lookup in a hash table is extremely fast for large tables---in fact, the | |
17 | time required is essentially @emph{independent} of how many elements are | |
18 | stored in the table. For smaller tables (a few tens of elements) | |
19 | alists may still be faster because hash tables have a more-or-less | |
20 | constant overhead. | |
21 | ||
22 | @item | |
23 | The correspondences in a hash table are in no particular order. | |
24 | ||
25 | @item | |
26 | There is no way to share structure between two hash tables, | |
27 | the way two alists can share a common tail. | |
28 | @end itemize | |
29 | ||
30 | Emacs Lisp provides a general-purpose hash table data type, along | |
16d1ff5f CY |
31 | with a series of functions for operating on them. Hash tables have a |
32 | special printed representation, which consists of @samp{#s} followed | |
33 | by a list specifying the hash table properties and contents. | |
34 | @xref{Creating Hash}. (Note that the term ``hash notation'', which | |
35 | refers to the initial @samp{#} character used in the printed | |
36 | representations of objects with no read representation, has nothing to | |
37 | do with the term ``hash table''. @xref{Printed Representation}.) | |
b8d4c8d0 GM |
38 | |
39 | Obarrays are also a kind of hash table, but they are a different type | |
40 | of object and are used only for recording interned symbols | |
41 | (@pxref{Creating Symbols}). | |
42 | ||
43 | @menu | |
44 | * Creating Hash:: Functions to create hash tables. | |
45 | * Hash Access:: Reading and writing the hash table contents. | |
c3cbbb40 | 46 | * Defining Hash:: Defining new comparison methods. |
b8d4c8d0 GM |
47 | * Other Hash:: Miscellaneous. |
48 | @end menu | |
49 | ||
50 | @node Creating Hash | |
51 | @section Creating Hash Tables | |
52 | @cindex creating hash tables | |
53 | ||
54 | The principal function for creating a hash table is | |
55 | @code{make-hash-table}. | |
56 | ||
57 | @defun make-hash-table &rest keyword-args | |
58 | This function creates a new hash table according to the specified | |
59 | arguments. The arguments should consist of alternating keywords | |
60 | (particular symbols recognized specially) and values corresponding to | |
61 | them. | |
62 | ||
63 | Several keywords make sense in @code{make-hash-table}, but the only two | |
64 | that you really need to know about are @code{:test} and @code{:weakness}. | |
65 | ||
66 | @table @code | |
67 | @item :test @var{test} | |
68 | This specifies the method of key lookup for this hash table. The | |
69 | default is @code{eql}; @code{eq} and @code{equal} are other | |
70 | alternatives: | |
71 | ||
72 | @table @code | |
73 | @item eql | |
74 | Keys which are numbers are ``the same'' if they are @code{equal}, that | |
75 | is, if they are equal in value and either both are integers or both | |
76 | are floating point numbers; otherwise, two distinct objects are never | |
16152b76 | 77 | ``the same''. |
b8d4c8d0 GM |
78 | |
79 | @item eq | |
80 | Any two distinct Lisp objects are ``different'' as keys. | |
81 | ||
82 | @item equal | |
16152b76 | 83 | Two Lisp objects are ``the same'', as keys, if they are equal |
b8d4c8d0 GM |
84 | according to @code{equal}. |
85 | @end table | |
86 | ||
87 | You can use @code{define-hash-table-test} (@pxref{Defining Hash}) to | |
88 | define additional possibilities for @var{test}. | |
89 | ||
90 | @item :weakness @var{weak} | |
91 | The weakness of a hash table specifies whether the presence of a key or | |
92 | value in the hash table preserves it from garbage collection. | |
93 | ||
94 | The value, @var{weak}, must be one of @code{nil}, @code{key}, | |
95 | @code{value}, @code{key-or-value}, @code{key-and-value}, or @code{t} | |
96 | which is an alias for @code{key-and-value}. If @var{weak} is @code{key} | |
97 | then the hash table does not prevent its keys from being collected as | |
98 | garbage (if they are not referenced anywhere else); if a particular key | |
99 | does get collected, the corresponding association is removed from the | |
100 | hash table. | |
101 | ||
102 | If @var{weak} is @code{value}, then the hash table does not prevent | |
103 | values from being collected as garbage (if they are not referenced | |
104 | anywhere else); if a particular value does get collected, the | |
105 | corresponding association is removed from the hash table. | |
106 | ||
107 | If @var{weak} is @code{key-and-value} or @code{t}, both the key and | |
108 | the value must be live in order to preserve the association. Thus, | |
109 | the hash table does not protect either keys or values from garbage | |
110 | collection; if either one is collected as garbage, that removes the | |
111 | association. | |
112 | ||
113 | If @var{weak} is @code{key-or-value}, either the key or | |
114 | the value can preserve the association. Thus, associations are | |
115 | removed from the hash table when both their key and value would be | |
116 | collected as garbage (if not for references from weak hash tables). | |
117 | ||
118 | The default for @var{weak} is @code{nil}, so that all keys and values | |
119 | referenced in the hash table are preserved from garbage collection. | |
120 | ||
121 | @item :size @var{size} | |
122 | This specifies a hint for how many associations you plan to store in the | |
123 | hash table. If you know the approximate number, you can make things a | |
124 | little more efficient by specifying it this way. If you specify too | |
125 | small a size, the hash table will grow automatically when necessary, but | |
126 | doing that takes some extra time. | |
127 | ||
128 | The default size is 65. | |
129 | ||
130 | @item :rehash-size @var{rehash-size} | |
16152b76 | 131 | When you add an association to a hash table and the table is ``full'', |
b8d4c8d0 GM |
132 | it grows automatically. This value specifies how to make the hash table |
133 | larger, at that time. | |
134 | ||
135 | If @var{rehash-size} is an integer, it should be positive, and the hash | |
136 | table grows by adding that much to the nominal size. If | |
137 | @var{rehash-size} is a floating point number, it had better be greater | |
138 | than 1, and the hash table grows by multiplying the old size by that | |
139 | number. | |
140 | ||
141 | The default value is 1.5. | |
142 | ||
143 | @item :rehash-threshold @var{threshold} | |
144 | This specifies the criterion for when the hash table is ``full'' (so | |
145 | it should be made larger). The value, @var{threshold}, should be a | |
146 | positive floating point number, no greater than 1. The hash table is | |
147 | ``full'' whenever the actual number of entries exceeds this fraction | |
148 | of the nominal size. The default for @var{threshold} is 0.8. | |
149 | @end table | |
150 | @end defun | |
151 | ||
152 | @defun makehash &optional test | |
153 | This is equivalent to @code{make-hash-table}, but with a different style | |
154 | argument list. The argument @var{test} specifies the method | |
155 | of key lookup. | |
156 | ||
157 | This function is obsolete. Use @code{make-hash-table} instead. | |
158 | @end defun | |
159 | ||
16d1ff5f CY |
160 | You can also create a new hash table using the printed representation |
161 | for hash tables. The Lisp reader can read this printed | |
162 | representation, provided each element in the specified hash table has | |
163 | a valid read syntax (@pxref{Printed Representation}). For instance, | |
164 | the following specifies a new hash table containing the keys | |
165 | @code{key1} and @code{key2} (both symbols) associated with @code{val1} | |
166 | (a symbol) and @code{300} (a number) respectively. | |
167 | ||
168 | @example | |
169 | #s(hash-table size 30 data (key1 val1 key2 300)) | |
170 | @end example | |
171 | ||
172 | @noindent | |
173 | The printed representation for a hash table consists of @samp{#s} | |
174 | followed by a list beginning with @samp{hash-table}. The rest of the | |
175 | list should consist of zero or more property-value pairs specifying | |
176 | the hash table's properties and initial contents. The properties and | |
177 | values are read literally. Valid property names are @code{size}, | |
178 | @code{test}, @code{weakness}, @code{rehash-size}, | |
179 | @code{rehash-threshold}, and @code{data}. The @code{data} property | |
180 | should be a list of key-value pairs for the initial contents; the | |
181 | other properties have the same meanings as the matching | |
182 | @code{make-hash-table} keywords (@code{:size}, @code{:test}, etc.), | |
183 | described above. | |
184 | ||
185 | Note that you cannot specify a hash table whose initial contents | |
186 | include objects that have no read syntax, such as buffers and frames. | |
187 | Such objects may be added to the hash table after it is created. | |
188 | ||
b8d4c8d0 GM |
189 | @node Hash Access |
190 | @section Hash Table Access | |
191 | ||
192 | This section describes the functions for accessing and storing | |
193 | associations in a hash table. In general, any Lisp object can be used | |
194 | as a hash key, unless the comparison method imposes limits. Any Lisp | |
195 | object can also be used as the value. | |
196 | ||
197 | @defun gethash key table &optional default | |
198 | This function looks up @var{key} in @var{table}, and returns its | |
199 | associated @var{value}---or @var{default}, if @var{key} has no | |
200 | association in @var{table}. | |
201 | @end defun | |
202 | ||
203 | @defun puthash key value table | |
204 | This function enters an association for @var{key} in @var{table}, with | |
205 | value @var{value}. If @var{key} already has an association in | |
206 | @var{table}, @var{value} replaces the old associated value. | |
207 | @end defun | |
208 | ||
209 | @defun remhash key table | |
210 | This function removes the association for @var{key} from @var{table}, if | |
211 | there is one. If @var{key} has no association, @code{remhash} does | |
212 | nothing. | |
213 | ||
214 | @b{Common Lisp note:} In Common Lisp, @code{remhash} returns | |
215 | non-@code{nil} if it actually removed an association and @code{nil} | |
216 | otherwise. In Emacs Lisp, @code{remhash} always returns @code{nil}. | |
217 | @end defun | |
218 | ||
219 | @defun clrhash table | |
220 | This function removes all the associations from hash table @var{table}, | |
221 | so that it becomes empty. This is also called @dfn{clearing} the hash | |
222 | table. | |
223 | ||
224 | @b{Common Lisp note:} In Common Lisp, @code{clrhash} returns the empty | |
225 | @var{table}. In Emacs Lisp, it returns @code{nil}. | |
226 | @end defun | |
227 | ||
228 | @defun maphash function table | |
229 | @anchor{Definition of maphash} | |
230 | This function calls @var{function} once for each of the associations in | |
231 | @var{table}. The function @var{function} should accept two | |
232 | arguments---a @var{key} listed in @var{table}, and its associated | |
233 | @var{value}. @code{maphash} returns @code{nil}. | |
234 | @end defun | |
235 | ||
236 | @node Defining Hash | |
237 | @section Defining Hash Comparisons | |
238 | @cindex hash code | |
239 | @cindex define hash comparisons | |
240 | ||
241 | You can define new methods of key lookup by means of | |
242 | @code{define-hash-table-test}. In order to use this feature, you need | |
243 | to understand how hash tables work, and what a @dfn{hash code} means. | |
244 | ||
245 | You can think of a hash table conceptually as a large array of many | |
246 | slots, each capable of holding one association. To look up a key, | |
247 | @code{gethash} first computes an integer, the hash code, from the key. | |
248 | It reduces this integer modulo the length of the array, to produce an | |
249 | index in the array. Then it looks in that slot, and if necessary in | |
250 | other nearby slots, to see if it has found the key being sought. | |
251 | ||
252 | Thus, to define a new method of key lookup, you need to specify both a | |
253 | function to compute the hash code from a key, and a function to compare | |
254 | two keys directly. | |
255 | ||
256 | @defun define-hash-table-test name test-fn hash-fn | |
257 | This function defines a new hash table test, named @var{name}. | |
258 | ||
259 | After defining @var{name} in this way, you can use it as the @var{test} | |
260 | argument in @code{make-hash-table}. When you do that, the hash table | |
261 | will use @var{test-fn} to compare key values, and @var{hash-fn} to compute | |
262 | a ``hash code'' from a key value. | |
263 | ||
264 | The function @var{test-fn} should accept two arguments, two keys, and | |
16152b76 | 265 | return non-@code{nil} if they are considered ``the same''. |
b8d4c8d0 GM |
266 | |
267 | The function @var{hash-fn} should accept one argument, a key, and return | |
268 | an integer that is the ``hash code'' of that key. For good results, the | |
269 | function should use the whole range of integer values for hash codes, | |
270 | including negative integers. | |
271 | ||
272 | The specified functions are stored in the property list of @var{name} | |
273 | under the property @code{hash-table-test}; the property value's form is | |
274 | @code{(@var{test-fn} @var{hash-fn})}. | |
275 | @end defun | |
276 | ||
277 | @defun sxhash obj | |
278 | This function returns a hash code for Lisp object @var{obj}. | |
279 | This is an integer which reflects the contents of @var{obj} | |
280 | and the other Lisp objects it points to. | |
281 | ||
282 | If two objects @var{obj1} and @var{obj2} are equal, then @code{(sxhash | |
283 | @var{obj1})} and @code{(sxhash @var{obj2})} are the same integer. | |
284 | ||
285 | If the two objects are not equal, the values returned by @code{sxhash} | |
286 | are usually different, but not always; once in a rare while, by luck, | |
287 | you will encounter two distinct-looking objects that give the same | |
288 | result from @code{sxhash}. | |
289 | @end defun | |
290 | ||
291 | This example creates a hash table whose keys are strings that are | |
292 | compared case-insensitively. | |
293 | ||
294 | @example | |
295 | (defun case-fold-string= (a b) | |
4217dc29 | 296 | (eq t (compare-strings a nil nil b nil nil t))) |
b8d4c8d0 GM |
297 | (defun case-fold-string-hash (a) |
298 | (sxhash (upcase a))) | |
299 | ||
300 | (define-hash-table-test 'case-fold | |
301 | 'case-fold-string= 'case-fold-string-hash) | |
302 | ||
303 | (make-hash-table :test 'case-fold) | |
304 | @end example | |
305 | ||
306 | Here is how you could define a hash table test equivalent to the | |
307 | predefined test value @code{equal}. The keys can be any Lisp object, | |
308 | and equal-looking objects are considered the same key. | |
309 | ||
310 | @example | |
311 | (define-hash-table-test 'contents-hash 'equal 'sxhash) | |
312 | ||
313 | (make-hash-table :test 'contents-hash) | |
314 | @end example | |
315 | ||
316 | @node Other Hash | |
317 | @section Other Hash Table Functions | |
318 | ||
319 | Here are some other functions for working with hash tables. | |
320 | ||
321 | @defun hash-table-p table | |
322 | This returns non-@code{nil} if @var{table} is a hash table object. | |
323 | @end defun | |
324 | ||
325 | @defun copy-hash-table table | |
326 | This function creates and returns a copy of @var{table}. Only the table | |
327 | itself is copied---the keys and values are shared. | |
328 | @end defun | |
329 | ||
330 | @defun hash-table-count table | |
331 | This function returns the actual number of entries in @var{table}. | |
332 | @end defun | |
333 | ||
334 | @defun hash-table-test table | |
335 | This returns the @var{test} value that was given when @var{table} was | |
336 | created, to specify how to hash and compare keys. See | |
337 | @code{make-hash-table} (@pxref{Creating Hash}). | |
338 | @end defun | |
339 | ||
340 | @defun hash-table-weakness table | |
341 | This function returns the @var{weak} value that was specified for hash | |
342 | table @var{table}. | |
343 | @end defun | |
344 | ||
345 | @defun hash-table-rehash-size table | |
346 | This returns the rehash size of @var{table}. | |
347 | @end defun | |
348 | ||
349 | @defun hash-table-rehash-threshold table | |
350 | This returns the rehash threshold of @var{table}. | |
351 | @end defun | |
352 | ||
353 | @defun hash-table-size table | |
354 | This returns the current nominal size of @var{table}. | |
355 | @end defun | |
bb962da2 XF |
356 | |
357 | The following two functions are provided by the @file{subr-x} library. | |
358 | To use them, you need to load this library first. | |
359 | ||
360 | @defun hash-table-keys hash-table | |
361 | This returns a list of keys in @var{hash-table}. | |
362 | @end defun | |
363 | ||
364 | @defun hash-table-values hash-table | |
365 | This returns a list of values in @var{hash-table}. | |
366 | @end defun |