Mercurial > emacs
annotate lispref/compile.texi @ 23096:e4419c63d4d7
(message-send-mail-with-sendmail): Bind
coding-system-for-write by the return value of
select-message-coding-system.
(message-send-mail-with-qmail): Likewise.
author | Kenichi Handa <handa@m17n.org> |
---|---|
date | Tue, 25 Aug 1998 10:25:03 +0000 |
parents | d4ac295a98b3 |
children | 05a836654c88 |
rev | line source |
---|---|
5945 | 1 @c -*-texinfo-*- |
2 @c This is part of the GNU Emacs Lisp Reference Manual. | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
3 @c Copyright (C) 1990, 1991, 1992, 1993, 1994 Free Software Foundation, Inc. |
5945 | 4 @c See the file elisp.texi for copying conditions. |
5 @setfilename ../info/compile | |
22138
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
6 @node Byte Compilation, Advising Functions, Loading, Top |
5945 | 7 @chapter Byte Compilation |
8 @cindex byte-code | |
9 @cindex compilation | |
10 | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
11 Emacs Lisp has a @dfn{compiler} that translates functions written |
5945 | 12 in Lisp into a special representation called @dfn{byte-code} that can be |
13 executed more efficiently. The compiler replaces Lisp function | |
14 definitions with byte-code. When a byte-code function is called, its | |
15 definition is evaluated by the @dfn{byte-code interpreter}. | |
16 | |
17 Because the byte-compiled code is evaluated by the byte-code | |
18 interpreter, instead of being executed directly by the machine's | |
19 hardware (as true compiled code is), byte-code is completely | |
20 transportable from machine to machine without recompilation. It is not, | |
21 however, as fast as true compiled code. | |
22 | |
22138
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
23 Compiling a Lisp file with the Emacs byte compiler always reads the |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
24 file as multibyte text, even if Emacs was started with @samp{--unibyte}, |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
25 unless the file specifies otherwise. This is so that compilation gives |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
26 results compatible with running the same file without compilation. |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
27 @xref{Loading Non-ASCII}. |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
28 |
5945 | 29 In general, any version of Emacs can run byte-compiled code produced |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
30 by recent earlier versions of Emacs, but the reverse is not true. A |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
31 major incompatible change was introduced in Emacs version 19.29, and |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
32 files compiled with versions since that one will definitely not run |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
33 in earlier versions unless you specify a special option. |
12098 | 34 @iftex |
35 @xref{Docs and Compilation}. | |
36 @end iftex | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
37 In addition, the modifier bits in keyboard characters were renumbered in |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
38 Emacs 19.29; as a result, files compiled in versions before 19.29 will |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
39 not work in subsequent versions if they contain character constants with |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
40 modifier bits. |
5945 | 41 |
42 @xref{Compilation Errors}, for how to investigate errors occurring in | |
43 byte compilation. | |
44 | |
45 @menu | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
46 * Speed of Byte-Code:: An example of speedup from byte compilation. |
5945 | 47 * Compilation Functions:: Byte compilation functions. |
12067 | 48 * Docs and Compilation:: Dynamic loading of documentation strings. |
49 * Dynamic Loading:: Dynamic loading of individual functions. | |
5945 | 50 * Eval During Compile:: Code to be evaluated when you compile. |
51 * Byte-Code Objects:: The data type used for byte-compiled functions. | |
52 * Disassembly:: Disassembling byte-code; how to read byte-code. | |
53 @end menu | |
54 | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
55 @node Speed of Byte-Code |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
56 @section Performance of Byte-Compiled Code |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
57 |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
58 A byte-compiled function is not as efficient as a primitive function |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
59 written in C, but runs much faster than the version written in Lisp. |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
60 Here is an example: |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
61 |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
62 @example |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
63 @group |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
64 (defun silly-loop (n) |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
65 "Return time before and after N iterations of a loop." |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
66 (let ((t1 (current-time-string))) |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
67 (while (> (setq n (1- n)) |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
68 0)) |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
69 (list t1 (current-time-string)))) |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
70 @result{} silly-loop |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
71 @end group |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
72 |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
73 @group |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
74 (silly-loop 100000) |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
75 @result{} ("Fri Mar 18 17:25:57 1994" |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
76 "Fri Mar 18 17:26:28 1994") ; @r{31 seconds} |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
77 @end group |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
78 |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
79 @group |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
80 (byte-compile 'silly-loop) |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
81 @result{} @r{[Compiled code not shown]} |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
82 @end group |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
83 |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
84 @group |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
85 (silly-loop 100000) |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
86 @result{} ("Fri Mar 18 17:26:52 1994" |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
87 "Fri Mar 18 17:26:58 1994") ; @r{6 seconds} |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
88 @end group |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
89 @end example |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
90 |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
91 In this example, the interpreted code required 31 seconds to run, |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
92 whereas the byte-compiled code required 6 seconds. These results are |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
93 representative, but actual results will vary greatly. |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
94 |
5945 | 95 @node Compilation Functions |
96 @comment node-name, next, previous, up | |
97 @section The Compilation Functions | |
98 @cindex compilation functions | |
99 | |
100 You can byte-compile an individual function or macro definition with | |
101 the @code{byte-compile} function. You can compile a whole file with | |
102 @code{byte-compile-file}, or several files with | |
103 @code{byte-recompile-directory} or @code{batch-byte-compile}. | |
104 | |
16736
981e116b4ac6
Minor cleanups for overfull hboxes.
Richard M. Stallman <rms@gnu.org>
parents:
12128
diff
changeset
|
105 The byte compiler produces error messages and warnings about each file |
981e116b4ac6
Minor cleanups for overfull hboxes.
Richard M. Stallman <rms@gnu.org>
parents:
12128
diff
changeset
|
106 in a buffer called @samp{*Compile-Log*}. These report things in your |
981e116b4ac6
Minor cleanups for overfull hboxes.
Richard M. Stallman <rms@gnu.org>
parents:
12128
diff
changeset
|
107 program that suggest a problem but are not necessarily erroneous. |
5945 | 108 |
109 @cindex macro compilation | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
110 Be careful when writing macro calls in files that you may someday |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
111 byte-compile. Macro calls are expanded when they are compiled, so the |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
112 macros must already be defined for proper compilation. For more |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
113 details, see @ref{Compiling Macros}. |
5945 | 114 |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
115 Normally, compiling a file does not evaluate the file's contents or |
12098 | 116 load the file. But it does execute any @code{require} calls at top |
117 level in the file. One way to ensure that necessary macro definitions | |
118 are available during compilation is to require the file that defines | |
119 them (@pxref{Named Features}). To avoid loading the macro definition files | |
120 when someone @emph{runs} the compiled program, write | |
121 @code{eval-when-compile} around the @code{require} calls (@pxref{Eval | |
122 During Compile}). | |
5945 | 123 |
124 @defun byte-compile symbol | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
125 This function byte-compiles the function definition of @var{symbol}, |
5945 | 126 replacing the previous definition with the compiled one. The function |
127 definition of @var{symbol} must be the actual code for the function; | |
128 i.e., the compiler does not follow indirection to another symbol. | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
129 @code{byte-compile} returns the new, compiled definition of |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
130 @var{symbol}. |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
131 |
12067 | 132 If @var{symbol}'s definition is a byte-code function object, |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
133 @code{byte-compile} does nothing and returns @code{nil}. Lisp records |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
134 only one function definition for any symbol, and if that is already |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
135 compiled, non-compiled code is not available anywhere. So there is no |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
136 way to ``compile the same definition again.'' |
5945 | 137 |
138 @example | |
139 @group | |
140 (defun factorial (integer) | |
141 "Compute factorial of INTEGER." | |
142 (if (= 1 integer) 1 | |
143 (* integer (factorial (1- integer))))) | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
144 @result{} factorial |
5945 | 145 @end group |
146 | |
147 @group | |
148 (byte-compile 'factorial) | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
149 @result{} |
5945 | 150 #[(integer) |
151 "^H\301U\203^H^@@\301\207\302^H\303^HS!\"\207" | |
152 [integer 1 * factorial] | |
153 4 "Compute factorial of INTEGER."] | |
154 @end group | |
155 @end example | |
156 | |
157 @noindent | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
158 The result is a byte-code function object. The string it contains is |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
159 the actual byte-code; each character in it is an instruction or an |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
160 operand of an instruction. The vector contains all the constants, |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
161 variable names and function names used by the function, except for |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
162 certain primitives that are coded as special instructions. |
5945 | 163 @end defun |
164 | |
165 @deffn Command compile-defun | |
166 This command reads the defun containing point, compiles it, and | |
167 evaluates the result. If you use this on a defun that is actually a | |
168 function definition, the effect is to install a compiled version of that | |
169 function. | |
170 @end deffn | |
171 | |
172 @deffn Command byte-compile-file filename | |
22138
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
173 This function compiles a file of Lisp code named @var{filename} into a |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
174 file of byte-code. The output file's name is made by changing the |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
175 @samp{.el} suffix into @samp{.elc}; if @var{filename} does not end in |
d4ac295a98b3
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21682
diff
changeset
|
176 @samp{.el}, it adds @samp{.elc} to the end of @var{filename}. |
5945 | 177 |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
178 Compilation works by reading the input file one form at a time. If it |
5945 | 179 is a definition of a function or macro, the compiled function or macro |
180 definition is written out. Other forms are batched together, then each | |
181 batch is compiled, and written so that its compiled code will be | |
182 executed when the file is read. All comments are discarded when the | |
183 input file is read. | |
184 | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
185 This command returns @code{t}. When called interactively, it prompts |
5945 | 186 for the file name. |
187 | |
188 @example | |
189 @group | |
190 % ls -l push* | |
191 -rw-r--r-- 1 lewis 791 Oct 5 20:31 push.el | |
192 @end group | |
193 | |
194 @group | |
195 (byte-compile-file "~/emacs/push.el") | |
196 @result{} t | |
197 @end group | |
198 | |
199 @group | |
200 % ls -l push* | |
201 -rw-r--r-- 1 lewis 791 Oct 5 20:31 push.el | |
202 -rw-rw-rw- 1 lewis 638 Oct 8 20:25 push.elc | |
203 @end group | |
204 @end example | |
205 @end deffn | |
206 | |
207 @deffn Command byte-recompile-directory directory flag | |
208 @cindex library compilation | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
209 This function recompiles every @samp{.el} file in @var{directory} that |
5945 | 210 needs recompilation. A file needs recompilation if a @samp{.elc} file |
211 exists but is older than the @samp{.el} file. | |
212 | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
213 When a @samp{.el} file has no corresponding @samp{.elc} file, @var{flag} |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
214 says what to do. If it is @code{nil}, these files are ignored. If it |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
215 is non-@code{nil}, the user is asked whether to compile each such file. |
5945 | 216 |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
217 The returned value of this command is unpredictable. |
5945 | 218 @end deffn |
219 | |
220 @defun batch-byte-compile | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
221 This function runs @code{byte-compile-file} on files specified on the |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
222 command line. This function must be used only in a batch execution of |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
223 Emacs, as it kills Emacs on completion. An error in one file does not |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16736
diff
changeset
|
224 prevent processing of subsequent files, but no output file will be |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16736
diff
changeset
|
225 generated for it, and the Emacs process will terminate with a nonzero |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16736
diff
changeset
|
226 status code. |
5945 | 227 |
228 @example | |
229 % emacs -batch -f batch-byte-compile *.el | |
230 @end example | |
231 @end defun | |
232 | |
233 @defun byte-code code-string data-vector max-stack | |
234 @cindex byte-code interpreter | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
235 This function actually interprets byte-code. A byte-compiled function |
5945 | 236 is actually defined with a body that calls @code{byte-code}. Don't call |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
237 this function yourself---only the byte compiler knows how to generate |
5945 | 238 valid calls to this function. |
239 | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
240 In Emacs version 18, byte-code was always executed by way of a call to |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
241 the function @code{byte-code}. Nowadays, byte-code is usually executed |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
242 as part of a byte-code function object, and only rarely through an |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
243 explicit call to @code{byte-code}. |
5945 | 244 @end defun |
245 | |
12067 | 246 @node Docs and Compilation |
247 @section Documentation Strings and Compilation | |
248 @cindex dynamic loading of documentation | |
249 | |
250 Functions and variables loaded from a byte-compiled file access their | |
251 documentation strings dynamically from the file whenever needed. This | |
12128
27144f55d1c6
fixed errors that appeared during update to 19.29.
Melissa Weisshaus <melissa@gnu.org>
parents:
12124
diff
changeset
|
252 saves space within Emacs, and makes loading faster because the |
12067 | 253 documentation strings themselves need not be processed while loading the |
254 file. Actual access to the documentation strings becomes slower as a | |
255 result, but this normally is not enough to bother users. | |
256 | |
257 Dynamic access to documentation strings does have drawbacks: | |
258 | |
259 @itemize @bullet | |
260 @item | |
261 If you delete or move the compiled file after loading it, Emacs can no | |
262 longer access the documentation strings for the functions and variables | |
263 in the file. | |
264 | |
265 @item | |
266 If you alter the compiled file (such as by compiling a new version), | |
267 then further access to documentation strings in this file will give | |
268 nonsense results. | |
269 @end itemize | |
270 | |
271 If your site installs Emacs following the usual procedures, these | |
272 problems will never normally occur. Installing a new version uses a new | |
273 directory with a different name; as long as the old version remains | |
274 installed, its files will remain unmodified in the places where they are | |
275 expected to be. | |
276 | |
12124 | 277 However, if you have built Emacs yourself and use it from the |
12067 | 278 directory where you built it, you will experience this problem |
279 occasionally if you edit and recompile Lisp files. When it happens, you | |
280 can cure the problem by reloading the file after recompiling it. | |
281 | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
282 Byte-compiled files made with recent versions of Emacs (since 19.29) |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
283 will not load into older versions because the older versions don't |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
284 support this feature. You can turn off this feature at compile time by |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
285 setting @code{byte-compile-dynamic-docstrings} to @code{nil}; then you |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
286 can compile files that will load into older Emacs versions. You can do |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
287 this globally, or for one source file by specifying a file-local binding |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
288 for the variable. One way to do that is by adding this string to the |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
289 file's first line: |
12067 | 290 |
291 @example | |
292 -*-byte-compile-dynamic-docstrings: nil;-*- | |
293 @end example | |
294 | |
295 @defvar byte-compile-dynamic-docstrings | |
296 If this is non-@code{nil}, the byte compiler generates compiled files | |
297 that are set up for dynamic loading of documentation strings. | |
298 @end defvar | |
299 | |
300 @cindex @samp{#@@@var{count}} | |
301 @cindex @samp{#$} | |
302 The dynamic documentation string feature writes compiled files that | |
303 use a special Lisp reader construct, @samp{#@@@var{count}}. This | |
304 construct skips the next @var{count} characters. It also uses the | |
305 @samp{#$} construct, which stands for ``the name of this file, as a | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
306 string.'' It is usually best not to use these constructs in Lisp source |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
307 files, since they are not designed to be clear to humans reading the |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
308 file. |
12067 | 309 |
310 @node Dynamic Loading | |
311 @section Dynamic Loading of Individual Functions | |
312 | |
313 @cindex dynamic loading of functions | |
314 @cindex lazy loading | |
315 When you compile a file, you can optionally enable the @dfn{dynamic | |
316 function loading} feature (also known as @dfn{lazy loading}). With | |
317 dynamic function loading, loading the file doesn't fully read the | |
318 function definitions in the file. Instead, each function definition | |
319 contains a place-holder which refers to the file. The first time each | |
320 function is called, it reads the full definition from the file, to | |
321 replace the place-holder. | |
322 | |
323 The advantage of dynamic function loading is that loading the file | |
324 becomes much faster. This is a good thing for a file which contains | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
325 many separate user-callable functions, if using one of them does not |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
326 imply you will probably also use the rest. A specialized mode which |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
327 provides many keyboard commands often has that usage pattern: a user may |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
328 invoke the mode, but use only a few of the commands it provides. |
12067 | 329 |
330 The dynamic loading feature has certain disadvantages: | |
331 | |
332 @itemize @bullet | |
333 @item | |
334 If you delete or move the compiled file after loading it, Emacs can no | |
335 longer load the remaining function definitions not already loaded. | |
336 | |
337 @item | |
338 If you alter the compiled file (such as by compiling a new version), | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
339 then trying to load any function not already loaded will yield nonsense |
12067 | 340 results. |
341 @end itemize | |
342 | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
343 These problems will never happen in normal circumstances with |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
344 installed Emacs files. But they are quite likely to happen with Lisp |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
345 files that you are changing. The easiest way to prevent these problems |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
346 is to reload the new compiled file immediately after each recompilation. |
12067 | 347 |
348 The byte compiler uses the dynamic function loading feature if the | |
349 variable @code{byte-compile-dynamic} is non-@code{nil} at compilation | |
350 time. Do not set this variable globally, since dynamic loading is | |
351 desirable only for certain files. Instead, enable the feature for | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
352 specific source files with file-local variable bindings. For example, |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
353 you could do it by writing this text in the source file's first line: |
12067 | 354 |
355 @example | |
356 -*-byte-compile-dynamic: t;-*- | |
357 @end example | |
358 | |
359 @defvar byte-compile-dynamic | |
360 If this is non-@code{nil}, the byte compiler generates compiled files | |
361 that are set up for dynamic function loading. | |
362 @end defvar | |
363 | |
364 @defun fetch-bytecode function | |
365 This immediately finishes loading the definition of @var{function} from | |
366 its byte-compiled file, if it is not fully loaded already. The argument | |
367 @var{function} may be a byte-code function object or a function name. | |
368 @end defun | |
369 | |
5945 | 370 @node Eval During Compile |
371 @section Evaluation During Compilation | |
372 | |
12067 | 373 These features permit you to write code to be evaluated during |
5945 | 374 compilation of a program. |
375 | |
376 @defspec eval-and-compile body | |
377 This form marks @var{body} to be evaluated both when you compile the | |
378 containing code and when you run it (whether compiled or not). | |
379 | |
380 You can get a similar result by putting @var{body} in a separate file | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
381 and referring to that file with @code{require}. That method is |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
382 preferable when @var{body} is large. |
5945 | 383 @end defspec |
384 | |
385 @defspec eval-when-compile body | |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16736
diff
changeset
|
386 This form marks @var{body} to be evaluated at compile time but not when |
7212 | 387 the compiled program is loaded. The result of evaluation by the |
21007
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16736
diff
changeset
|
388 compiler becomes a constant which appears in the compiled program. If |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16736
diff
changeset
|
389 you load the source file, rather than compiling it, @var{body} is |
66d807bdc5b4
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
16736
diff
changeset
|
390 evaluated normally. |
5945 | 391 |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
392 @strong{Common Lisp Note:} At top level, this is analogous to the Common |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
393 Lisp idiom @code{(eval-when (compile eval) @dots{})}. Elsewhere, the |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
394 Common Lisp @samp{#.} reader macro (but not when interpreting) is closer |
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
395 to what @code{eval-when-compile} does. |
5945 | 396 @end defspec |
397 | |
398 @node Byte-Code Objects | |
12098 | 399 @section Byte-Code Function Objects |
5945 | 400 @cindex compiled function |
401 @cindex byte-code function | |
402 | |
403 Byte-compiled functions have a special data type: they are | |
404 @dfn{byte-code function objects}. | |
405 | |
406 Internally, a byte-code function object is much like a vector; | |
407 however, the evaluator handles this data type specially when it appears | |
408 as a function to be called. The printed representation for a byte-code | |
409 function object is like that for a vector, with an additional @samp{#} | |
410 before the opening @samp{[}. | |
411 | |
412 A byte-code function object must have at least four elements; there is | |
21682
90da2489c498
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
21007
diff
changeset
|
413 no maximum number, but only the first six elements have any normal use. |
5945 | 414 They are: |
415 | |
416 @table @var | |
417 @item arglist | |
418 The list of argument symbols. | |
419 | |
420 @item byte-code | |
421 The string containing the byte-code instructions. | |
422 | |
423 @item constants | |
7212 | 424 The vector of Lisp objects referenced by the byte code. These include |
425 symbols used as function names and variable names. | |
5945 | 426 |
427 @item stacksize | |
428 The maximum stack size this function needs. | |
429 | |
430 @item docstring | |
12098 | 431 The documentation string (if any); otherwise, @code{nil}. The value may |
432 be a number or a list, in case the documentation string is stored in a | |
433 file. Use the function @code{documentation} to get the real | |
434 documentation string (@pxref{Accessing Documentation}). | |
5945 | 435 |
436 @item interactive | |
437 The interactive spec (if any). This can be a string or a Lisp | |
438 expression. It is @code{nil} for a function that isn't interactive. | |
439 @end table | |
440 | |
441 Here's an example of a byte-code function object, in printed | |
442 representation. It is the definition of the command | |
443 @code{backward-sexp}. | |
444 | |
445 @example | |
446 #[(&optional arg) | |
447 "^H\204^F^@@\301^P\302^H[!\207" | |
448 [arg 1 forward-sexp] | |
449 2 | |
450 254435 | |
451 "p"] | |
452 @end example | |
453 | |
454 The primitive way to create a byte-code object is with | |
455 @code{make-byte-code}: | |
456 | |
457 @defun make-byte-code &rest elements | |
458 This function constructs and returns a byte-code function object | |
459 with @var{elements} as its elements. | |
460 @end defun | |
461 | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
462 You should not try to come up with the elements for a byte-code |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
463 function yourself, because if they are inconsistent, Emacs may crash |
7212 | 464 when you call the function. Always leave it to the byte compiler to |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
465 create these objects; it makes the elements consistent (we hope). |
5945 | 466 |
467 You can access the elements of a byte-code object using @code{aref}; | |
468 you can also use @code{vconcat} to create a vector with the same | |
469 elements. | |
470 | |
471 @node Disassembly | |
472 @section Disassembled Byte-Code | |
473 @cindex disassembled byte-code | |
474 | |
475 People do not write byte-code; that job is left to the byte compiler. | |
476 But we provide a disassembler to satisfy a cat-like curiosity. The | |
477 disassembler converts the byte-compiled code into humanly readable | |
478 form. | |
479 | |
480 The byte-code interpreter is implemented as a simple stack machine. | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
481 It pushes values onto a stack of its own, then pops them off to use them |
7212 | 482 in calculations whose results are themselves pushed back on the stack. |
483 When a byte-code function returns, it pops a value off the stack and | |
484 returns it as the value of the function. | |
5945 | 485 |
7212 | 486 In addition to the stack, byte-code functions can use, bind, and set |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
487 ordinary Lisp variables, by transferring values between variables and |
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
488 the stack. |
5945 | 489 |
490 @deffn Command disassemble object &optional stream | |
491 This function prints the disassembled code for @var{object}. If | |
492 @var{stream} is supplied, then output goes there. Otherwise, the | |
493 disassembled code is printed to the stream @code{standard-output}. The | |
494 argument @var{object} can be a function name or a lambda expression. | |
495 | |
496 As a special exception, if this function is used interactively, | |
497 it outputs to a buffer named @samp{*Disassemble*}. | |
498 @end deffn | |
499 | |
500 Here are two examples of using the @code{disassemble} function. We | |
501 have added explanatory comments to help you relate the byte-code to the | |
502 Lisp source; these do not appear in the output of @code{disassemble}. | |
503 These examples show unoptimized byte-code. Nowadays byte-code is | |
504 usually optimized, but we did not want to rewrite these examples, since | |
505 they still serve their purpose. | |
506 | |
507 @example | |
508 @group | |
509 (defun factorial (integer) | |
510 "Compute factorial of an integer." | |
511 (if (= 1 integer) 1 | |
512 (* integer (factorial (1- integer))))) | |
513 @result{} factorial | |
514 @end group | |
515 | |
516 @group | |
517 (factorial 4) | |
518 @result{} 24 | |
519 @end group | |
520 | |
521 @group | |
522 (disassemble 'factorial) | |
523 @print{} byte-code for factorial: | |
524 doc: Compute factorial of an integer. | |
525 args: (integer) | |
526 @end group | |
527 | |
528 @group | |
529 0 constant 1 ; @r{Push 1 onto stack.} | |
530 | |
531 1 varref integer ; @r{Get value of @code{integer}} | |
532 ; @r{from the environment} | |
533 ; @r{and push the value} | |
534 ; @r{onto the stack.} | |
535 @end group | |
536 | |
537 @group | |
538 2 eqlsign ; @r{Pop top two values off stack,} | |
539 ; @r{compare them,} | |
540 ; @r{and push result onto stack.} | |
541 @end group | |
542 | |
543 @group | |
544 3 goto-if-nil 10 ; @r{Pop and test top of stack;} | |
545 ; @r{if @code{nil}, go to 10,} | |
546 ; @r{else continue.} | |
547 @end group | |
548 | |
549 @group | |
550 6 constant 1 ; @r{Push 1 onto top of stack.} | |
551 | |
552 7 goto 17 ; @r{Go to 17 (in this case, 1 will be} | |
553 ; @r{returned by the function).} | |
554 @end group | |
555 | |
556 @group | |
557 10 constant * ; @r{Push symbol @code{*} onto stack.} | |
558 | |
559 11 varref integer ; @r{Push value of @code{integer} onto stack.} | |
560 @end group | |
561 | |
562 @group | |
563 12 constant factorial ; @r{Push @code{factorial} onto stack.} | |
564 | |
565 13 varref integer ; @r{Push value of @code{integer} onto stack.} | |
566 | |
567 14 sub1 ; @r{Pop @code{integer}, decrement value,} | |
568 ; @r{push new value onto stack.} | |
569 @end group | |
570 | |
571 @group | |
572 ; @r{Stack now contains:} | |
573 ; @minus{} @r{decremented value of @code{integer}} | |
574 ; @minus{} @r{@code{factorial}} | |
575 ; @minus{} @r{value of @code{integer}} | |
576 ; @minus{} @r{@code{*}} | |
577 @end group | |
578 | |
579 @group | |
580 15 call 1 ; @r{Call function @code{factorial} using} | |
581 ; @r{the first (i.e., the top) element} | |
582 ; @r{of the stack as the argument;} | |
583 ; @r{push returned value onto stack.} | |
584 @end group | |
585 | |
586 @group | |
587 ; @r{Stack now contains:} | |
7212 | 588 ; @minus{} @r{result of recursive} |
5945 | 589 ; @r{call to @code{factorial}} |
590 ; @minus{} @r{value of @code{integer}} | |
591 ; @minus{} @r{@code{*}} | |
592 @end group | |
593 | |
594 @group | |
595 16 call 2 ; @r{Using the first two} | |
596 ; @r{(i.e., the top two)} | |
597 ; @r{elements of the stack} | |
598 ; @r{as arguments,} | |
599 ; @r{call the function @code{*},} | |
600 ; @r{pushing the result onto the stack.} | |
601 @end group | |
602 | |
603 @group | |
604 17 return ; @r{Return the top element} | |
605 ; @r{of the stack.} | |
606 @result{} nil | |
607 @end group | |
608 @end example | |
609 | |
610 The @code{silly-loop} function is somewhat more complex: | |
611 | |
612 @example | |
613 @group | |
614 (defun silly-loop (n) | |
615 "Return time before and after N iterations of a loop." | |
616 (let ((t1 (current-time-string))) | |
617 (while (> (setq n (1- n)) | |
618 0)) | |
619 (list t1 (current-time-string)))) | |
620 @result{} silly-loop | |
621 @end group | |
622 | |
623 @group | |
624 (disassemble 'silly-loop) | |
625 @print{} byte-code for silly-loop: | |
626 doc: Return time before and after N iterations of a loop. | |
627 args: (n) | |
628 | |
629 0 constant current-time-string ; @r{Push} | |
630 ; @r{@code{current-time-string}} | |
631 ; @r{onto top of stack.} | |
632 @end group | |
633 | |
634 @group | |
635 1 call 0 ; @r{Call @code{current-time-string}} | |
636 ; @r{ with no argument,} | |
637 ; @r{ pushing result onto stack.} | |
638 @end group | |
639 | |
640 @group | |
641 2 varbind t1 ; @r{Pop stack and bind @code{t1}} | |
642 ; @r{to popped value.} | |
643 @end group | |
644 | |
645 @group | |
646 3 varref n ; @r{Get value of @code{n} from} | |
647 ; @r{the environment and push} | |
648 ; @r{the value onto the stack.} | |
649 @end group | |
650 | |
651 @group | |
652 4 sub1 ; @r{Subtract 1 from top of stack.} | |
653 @end group | |
654 | |
655 @group | |
656 5 dup ; @r{Duplicate the top of the stack;} | |
6452
8c7032348e93
*** empty log message ***
Richard M. Stallman <rms@gnu.org>
parents:
5945
diff
changeset
|
657 ; @r{i.e., copy the top of} |
5945 | 658 ; @r{the stack and push the} |
659 ; @r{copy onto the stack.} | |
660 @end group | |
661 | |
662 @group | |
663 6 varset n ; @r{Pop the top of the stack,} | |
664 ; @r{and bind @code{n} to the value.} | |
665 | |
666 ; @r{In effect, the sequence @code{dup varset}} | |
667 ; @r{copies the top of the stack} | |
668 ; @r{into the value of @code{n}} | |
669 ; @r{without popping it.} | |
670 @end group | |
671 | |
672 @group | |
673 7 constant 0 ; @r{Push 0 onto stack.} | |
674 @end group | |
675 | |
676 @group | |
677 8 gtr ; @r{Pop top two values off stack,} | |
678 ; @r{test if @var{n} is greater than 0} | |
679 ; @r{and push result onto stack.} | |
680 @end group | |
681 | |
682 @group | |
7212 | 683 9 goto-if-nil-else-pop 17 ; @r{Goto 17 if @code{n} <= 0} |
684 ; @r{(this exits the while loop).} | |
5945 | 685 ; @r{else pop top of stack} |
686 ; @r{and continue} | |
687 @end group | |
688 | |
689 @group | |
690 12 constant nil ; @r{Push @code{nil} onto stack} | |
691 ; @r{(this is the body of the loop).} | |
692 @end group | |
693 | |
694 @group | |
695 13 discard ; @r{Discard result of the body} | |
696 ; @r{of the loop (a while loop} | |
697 ; @r{is always evaluated for} | |
698 ; @r{its side effects).} | |
699 @end group | |
700 | |
701 @group | |
702 14 goto 3 ; @r{Jump back to beginning} | |
703 ; @r{of while loop.} | |
704 @end group | |
705 | |
706 @group | |
707 17 discard ; @r{Discard result of while loop} | |
708 ; @r{by popping top of stack.} | |
7212 | 709 ; @r{This result is the value @code{nil} that} |
710 ; @r{was not popped by the goto at 9.} | |
5945 | 711 @end group |
712 | |
713 @group | |
714 18 varref t1 ; @r{Push value of @code{t1} onto stack.} | |
715 @end group | |
716 | |
717 @group | |
718 19 constant current-time-string ; @r{Push} | |
719 ; @r{@code{current-time-string}} | |
720 ; @r{onto top of stack.} | |
721 @end group | |
722 | |
723 @group | |
724 20 call 0 ; @r{Call @code{current-time-string} again.} | |
725 @end group | |
726 | |
727 @group | |
728 21 list2 ; @r{Pop top two elements off stack,} | |
729 ; @r{create a list of them,} | |
730 ; @r{and push list onto stack.} | |
731 @end group | |
732 | |
733 @group | |
734 22 unbind 1 ; @r{Unbind @code{t1} in local environment.} | |
735 | |
736 23 return ; @r{Return value of the top of stack.} | |
737 | |
738 @result{} nil | |
739 @end group | |
740 @end example | |
741 | |
742 |