Commit | Line | Data |
---|---|---|
b8d4c8d0 GM |
1 | @c -*-texinfo-*- |
2 | @c This is part of the GNU Emacs Lisp Reference Manual. | |
29157371 | 3 | @c Copyright (C) 1990-1993, 1998-1999, 2001-2012 Free Software Foundation, Inc. |
b8d4c8d0 | 4 | @c See the file elisp.texi for copying conditions. |
ecc6530d | 5 | @node GNU Emacs Internals |
b8d4c8d0 GM |
6 | @appendix GNU Emacs Internals |
7 | ||
8 | This chapter describes how the runnable Emacs executable is dumped with | |
9 | the preloaded Lisp libraries in it, how storage is allocated, and some | |
10 | internal aspects of GNU Emacs that may be of interest to C programmers. | |
11 | ||
12 | @menu | |
13 | * Building Emacs:: How the dumped Emacs is made. | |
333f9019 | 14 | * Pure Storage:: Kludge to make preloaded Lisp functions shareable. |
b8d4c8d0 GM |
15 | * Garbage Collection:: Reclaiming space for Lisp objects no longer used. |
16 | * Memory Usage:: Info about total size of Lisp objects made so far. | |
17 | * Writing Emacs Primitives:: Writing C code for Emacs. | |
18 | * Object Internals:: Data formats of buffers, windows, processes. | |
19 | @end menu | |
20 | ||
21 | @node Building Emacs | |
520f2425 | 22 | @section Building Emacs |
b8d4c8d0 GM |
23 | @cindex building Emacs |
24 | @pindex temacs | |
25 | ||
26 | This section explains the steps involved in building the Emacs | |
27 | executable. You don't have to know this material to build and install | |
28 | Emacs, since the makefiles do all these things automatically. This | |
d15aac68 | 29 | information is pertinent to Emacs developers. |
b8d4c8d0 GM |
30 | |
31 | Compilation of the C source files in the @file{src} directory | |
32 | produces an executable file called @file{temacs}, also called a | |
d15aac68 CY |
33 | @dfn{bare impure Emacs}. It contains the Emacs Lisp interpreter and |
34 | I/O routines, but not the editing commands. | |
b8d4c8d0 GM |
35 | |
36 | @cindex @file{loadup.el} | |
d15aac68 CY |
37 | The command @w{@command{temacs -l loadup}} would run @file{temacs} |
38 | and direct it to load @file{loadup.el}. The @code{loadup} library | |
39 | loads additional Lisp libraries, which set up the normal Emacs editing | |
40 | environment. After this step, the Emacs executable is no longer | |
41 | @dfn{bare}. | |
b8d4c8d0 GM |
42 | |
43 | @cindex dumping Emacs | |
d15aac68 CY |
44 | Because it takes some time to load the standard Lisp files, the |
45 | @file{temacs} executable usually isn't run directly by users. | |
46 | Instead, as one of the last steps of building Emacs, the command | |
47 | @samp{temacs -batch -l loadup dump} is run. The special @samp{dump} | |
48 | argument causes @command{temacs} to dump out an executable program, | |
49 | called @file{emacs}, which has all the standard Lisp files preloaded. | |
50 | (The @samp{-batch} argument prevents @file{temacs} from trying to | |
51 | initialize any of its data on the terminal, so that the tables of | |
52 | terminal information are empty in the dumped Emacs.) | |
53 | ||
54 | @cindex preloaded Lisp files | |
b578c9cc | 55 | @vindex preloaded-file-list |
d15aac68 CY |
56 | The dumped @file{emacs} executable (also called a @dfn{pure} Emacs) |
57 | is the one which is installed. The variable | |
58 | @code{preloaded-file-list} stores a list of the Lisp files preloaded | |
59 | into the dumped Emacs. If you port Emacs to a new operating system, | |
60 | and are not able to implement dumping, then Emacs must load | |
61 | @file{loadup.el} each time it starts. | |
b8d4c8d0 GM |
62 | |
63 | @cindex @file{site-load.el} | |
b8d4c8d0 | 64 | You can specify additional files to preload by writing a library named |
520f2425 GM |
65 | @file{site-load.el} that loads them. You may need to rebuild Emacs |
66 | with an added definition | |
b8d4c8d0 GM |
67 | |
68 | @example | |
69 | #define SITELOAD_PURESIZE_EXTRA @var{n} | |
70 | @end example | |
71 | ||
72 | @noindent | |
520f2425 GM |
73 | to make @var{n} added bytes of pure space to hold the additional files; |
74 | see @file{src/puresize.h}. | |
b8d4c8d0 GM |
75 | (Try adding increments of 20000 until it is big enough.) However, the |
76 | advantage of preloading additional files decreases as machines get | |
77 | faster. On modern machines, it is usually not advisable. | |
78 | ||
79 | After @file{loadup.el} reads @file{site-load.el}, it finds the | |
80 | documentation strings for primitive and preloaded functions (and | |
81 | variables) in the file @file{etc/DOC} where they are stored, by | |
82 | calling @code{Snarf-documentation} (@pxref{Definition of | |
83 | Snarf-documentation,, Accessing Documentation}). | |
84 | ||
85 | @cindex @file{site-init.el} | |
86 | @cindex preloading additional functions and variables | |
87 | You can specify other Lisp expressions to execute just before dumping | |
88 | by putting them in a library named @file{site-init.el}. This file is | |
89 | executed after the documentation strings are found. | |
90 | ||
91 | If you want to preload function or variable definitions, there are | |
92 | three ways you can do this and make their documentation strings | |
93 | accessible when you subsequently run Emacs: | |
94 | ||
95 | @itemize @bullet | |
96 | @item | |
97 | Arrange to scan these files when producing the @file{etc/DOC} file, | |
98 | and load them with @file{site-load.el}. | |
99 | ||
100 | @item | |
101 | Load the files with @file{site-init.el}, then copy the files into the | |
102 | installation directory for Lisp files when you install Emacs. | |
103 | ||
104 | @item | |
5906db45 GM |
105 | Specify a @code{nil} value for @code{byte-compile-dynamic-docstrings} |
106 | as a local variable in each of these files, and load them with either | |
107 | @file{site-load.el} or @file{site-init.el}. (This method has the | |
108 | drawback that the documentation strings take up space in Emacs all the | |
109 | time.) | |
b8d4c8d0 GM |
110 | @end itemize |
111 | ||
112 | It is not advisable to put anything in @file{site-load.el} or | |
113 | @file{site-init.el} that would alter any of the features that users | |
114 | expect in an ordinary unmodified Emacs. If you feel you must override | |
115 | normal features for your site, do it with @file{default.el}, so that | |
116 | users can override your changes if they wish. @xref{Startup Summary}. | |
117 | ||
520f2425 GM |
118 | In a package that can be preloaded, it is sometimes necessary (or |
119 | useful) to delay certain evaluations until Emacs subsequently starts | |
120 | up. The vast majority of such cases relate to the values of | |
121 | customizable variables. For example, @code{tutorial-directory} is a | |
122 | variable defined in @file{startup.el}, which is preloaded. The default | |
123 | value is set based on @code{data-directory}. The variable needs to | |
124 | access the value of @code{data-directory} when Emacs starts, not when | |
125 | it is dumped, because the Emacs executable has probably been installed | |
126 | in a different location since it was dumped. | |
127 | ||
128 | @defun custom-initialize-delay symbol value | |
129 | This function delays the initialization of @var{symbol} to the next | |
130 | Emacs start. You normally use this function by specifying it as the | |
131 | @code{:initialize} property of a customizable variable. (The argument | |
f003f294 | 132 | @var{value} is unused, and is provided only for compatibility with the |
520f2425 GM |
133 | form Custom expects.) |
134 | @end defun | |
b8d4c8d0 | 135 | |
520f2425 GM |
136 | In the unlikely event that you need a more general functionality than |
137 | @code{custom-initialize-delay} provides, you can use | |
138 | @code{before-init-hook} (@pxref{Startup Summary}). | |
b8d4c8d0 GM |
139 | |
140 | @defun dump-emacs to-file from-file | |
141 | @cindex unexec | |
142 | This function dumps the current state of Emacs into an executable file | |
143 | @var{to-file}. It takes symbols from @var{from-file} (this is normally | |
144 | the executable file @file{temacs}). | |
145 | ||
146 | If you want to use this function in an Emacs that was already dumped, | |
147 | you must run Emacs with @samp{-batch}. | |
148 | @end defun | |
149 | ||
150 | @node Pure Storage | |
520f2425 | 151 | @section Pure Storage |
b8d4c8d0 GM |
152 | @cindex pure storage |
153 | ||
154 | Emacs Lisp uses two kinds of storage for user-created Lisp objects: | |
155 | @dfn{normal storage} and @dfn{pure storage}. Normal storage is where | |
999dd333 GM |
156 | all the new data created during an Emacs session are kept |
157 | (@pxref{Garbage Collection}). Pure storage is used for certain data | |
158 | in the preloaded standard Lisp files---data that should never change | |
159 | during actual use of Emacs. | |
b8d4c8d0 | 160 | |
d15aac68 | 161 | Pure storage is allocated only while @command{temacs} is loading the |
b8d4c8d0 GM |
162 | standard preloaded Lisp libraries. In the file @file{emacs}, it is |
163 | marked as read-only (on operating systems that permit this), so that | |
164 | the memory space can be shared by all the Emacs jobs running on the | |
165 | machine at once. Pure storage is not expandable; a fixed amount is | |
166 | allocated when Emacs is compiled, and if that is not sufficient for | |
167 | the preloaded libraries, @file{temacs} allocates dynamic memory for | |
999dd333 GM |
168 | the part that didn't fit. The resulting image will work, but garbage |
169 | collection (@pxref{Garbage Collection}) is disabled in this situation, | |
170 | causing a memory leak. Such an overflow normally won't happen unless | |
171 | you try to preload additional libraries or add features to the | |
172 | standard ones. Emacs will display a warning about the overflow when | |
173 | it starts. If this happens, you should increase the compilation | |
174 | parameter @code{SYSTEM_PURESIZE_EXTRA} in the file | |
175 | @file{src/puresize.h} and rebuild Emacs. | |
b8d4c8d0 GM |
176 | |
177 | @defun purecopy object | |
178 | This function makes a copy in pure storage of @var{object}, and returns | |
179 | it. It copies a string by simply making a new string with the same | |
180 | characters, but without text properties, in pure storage. It | |
181 | recursively copies the contents of vectors and cons cells. It does | |
182 | not make copies of other objects such as symbols, but just returns | |
183 | them unchanged. It signals an error if asked to copy markers. | |
184 | ||
185 | This function is a no-op except while Emacs is being built and dumped; | |
999dd333 | 186 | it is usually called only in preloaded Lisp files. |
b8d4c8d0 GM |
187 | @end defun |
188 | ||
189 | @defvar pure-bytes-used | |
190 | The value of this variable is the number of bytes of pure storage | |
191 | allocated so far. Typically, in a dumped Emacs, this number is very | |
192 | close to the total amount of pure storage available---if it were not, | |
193 | we would preallocate less. | |
194 | @end defvar | |
195 | ||
196 | @defvar purify-flag | |
197 | This variable determines whether @code{defun} should make a copy of the | |
198 | function definition in pure storage. If it is non-@code{nil}, then the | |
199 | function definition is copied into pure storage. | |
200 | ||
201 | This flag is @code{t} while loading all of the basic functions for | |
333f9019 | 202 | building Emacs initially (allowing those functions to be shareable and |
b8d4c8d0 GM |
203 | non-collectible). Dumping Emacs as an executable always writes |
204 | @code{nil} in this variable, regardless of the value it actually has | |
205 | before and after dumping. | |
206 | ||
207 | You should not change this flag in a running Emacs. | |
208 | @end defvar | |
209 | ||
210 | @node Garbage Collection | |
520f2425 | 211 | @section Garbage Collection |
b8d4c8d0 GM |
212 | |
213 | @cindex memory allocation | |
d15aac68 CY |
214 | When a program creates a list or the user defines a new function |
215 | (such as by loading a library), that data is placed in normal storage. | |
216 | If normal storage runs low, then Emacs asks the operating system to | |
217 | allocate more memory. Different types of Lisp objects, such as | |
f3372c87 DA |
218 | symbols, cons cells, small vectors, markers, etc., are segregated in |
219 | distinct blocks in memory. (Large vectors, long strings, buffers and | |
220 | certain other editing types, which are fairly large, are allocated in | |
221 | individual blocks, one per object; small strings are packed into blocks | |
222 | of 8k bytes, and small vectors are packed into blocks of 4k bytes). | |
223 | ||
224 | @cindex vector-like objects, storage | |
225 | @cindex storage of vector-like Lisp objects | |
226 | Beyond the basic vector, a lot of objects like window, buffer, and | |
227 | frame are managed as if they were vectors. The corresponding C data | |
228 | structures include the @code{struct vectorlike_header} field whose | |
229 | @code{next} field points to the next object in the chain: | |
230 | @code{header.next.buffer} points to the next buffer (which could be | |
231 | a killed buffer), and @code{header.next.vector} points to the next | |
232 | vector in a free list. If a vector is small (smaller than or equal to | |
dbab98c7 | 233 | @code{VBLOCK_BYTES_MAX} bytes, see @file{alloc.c}), then |
f3372c87 | 234 | @code{header.next.nbytes} contains the vector size in bytes. |
b8d4c8d0 | 235 | |
d15aac68 CY |
236 | @cindex garbage collection |
237 | It is quite common to use some storage for a while, then release it | |
238 | by (for example) killing a buffer or deleting the last pointer to an | |
239 | object. Emacs provides a @dfn{garbage collector} to reclaim this | |
240 | abandoned storage. The garbage collector operates by finding and | |
241 | marking all Lisp objects that are still accessible to Lisp programs. | |
242 | To begin with, it assumes all the symbols, their values and associated | |
243 | function definitions, and any data presently on the stack, are | |
244 | accessible. Any objects that can be reached indirectly through other | |
245 | accessible objects are also accessible. | |
b8d4c8d0 GM |
246 | |
247 | When marking is finished, all objects still unmarked are garbage. No | |
248 | matter what the Lisp program or the user does, it is impossible to refer | |
249 | to them, since there is no longer a way to reach them. Their space | |
250 | might as well be reused, since no one will miss them. The second | |
251 | (``sweep'') phase of the garbage collector arranges to reuse them. | |
252 | ||
253 | @c ??? Maybe add something describing weak hash tables here? | |
254 | ||
255 | @cindex free list | |
256 | The sweep phase puts unused cons cells onto a @dfn{free list} | |
257 | for future allocation; likewise for symbols and markers. It compacts | |
258 | the accessible strings so they occupy fewer 8k blocks; then it frees the | |
f3372c87 DA |
259 | other 8k blocks. Unreachable vectors from vector blocks are coalesced |
260 | to create largest possible free areas; if a free area spans a complete | |
261 | 4k block, that block is freed. Otherwise, the free area is recorded | |
262 | in a free list array, where each entry corresponds to a free list | |
263 | of areas of the same size. Large vectors, buffers, and other large | |
264 | objects are allocated and freed individually. | |
b8d4c8d0 GM |
265 | |
266 | @cindex CL note---allocate more storage | |
267 | @quotation | |
268 | @b{Common Lisp note:} Unlike other Lisps, GNU Emacs Lisp does not | |
269 | call the garbage collector when the free list is empty. Instead, it | |
270 | simply requests the operating system to allocate more storage, and | |
271 | processing continues until @code{gc-cons-threshold} bytes have been | |
272 | used. | |
273 | ||
274 | This means that you can make sure that the garbage collector will not | |
275 | run during a certain portion of a Lisp program by calling the garbage | |
276 | collector explicitly just before it (provided that portion of the | |
277 | program does not use so much space as to force a second garbage | |
278 | collection). | |
279 | @end quotation | |
280 | ||
281 | @deffn Command garbage-collect | |
282 | This command runs a garbage collection, and returns information on | |
283 | the amount of space in use. (Garbage collection can also occur | |
284 | spontaneously if you use more than @code{gc-cons-threshold} bytes of | |
285 | Lisp data since the previous garbage collection.) | |
286 | ||
287 | @code{garbage-collect} returns a list containing the following | |
288 | information: | |
289 | ||
290 | @example | |
291 | @group | |
292 | ((@var{used-conses} . @var{free-conses}) | |
293 | (@var{used-syms} . @var{free-syms}) | |
294 | @end group | |
295 | (@var{used-miscs} . @var{free-miscs}) | |
296 | @var{used-string-chars} | |
297 | @var{used-vector-slots} | |
298 | (@var{used-floats} . @var{free-floats}) | |
299 | (@var{used-intervals} . @var{free-intervals}) | |
300 | (@var{used-strings} . @var{free-strings})) | |
301 | @end example | |
302 | ||
303 | Here is an example: | |
304 | ||
305 | @example | |
306 | @group | |
307 | (garbage-collect) | |
308 | @result{} ((106886 . 13184) (9769 . 0) | |
309 | (7731 . 4651) 347543 121628 | |
310 | (31 . 94) (1273 . 168) | |
311 | (25474 . 3569)) | |
312 | @end group | |
313 | @end example | |
314 | ||
315 | Here is a table explaining each element: | |
316 | ||
317 | @table @var | |
318 | @item used-conses | |
319 | The number of cons cells in use. | |
320 | ||
321 | @item free-conses | |
322 | The number of cons cells for which space has been obtained from the | |
323 | operating system, but that are not currently being used. | |
324 | ||
325 | @item used-syms | |
326 | The number of symbols in use. | |
327 | ||
328 | @item free-syms | |
329 | The number of symbols for which space has been obtained from the | |
330 | operating system, but that are not currently being used. | |
331 | ||
332 | @item used-miscs | |
333 | The number of miscellaneous objects in use. These include markers and | |
334 | overlays, plus certain objects not visible to users. | |
335 | ||
336 | @item free-miscs | |
337 | The number of miscellaneous objects for which space has been obtained | |
338 | from the operating system, but that are not currently being used. | |
339 | ||
340 | @item used-string-chars | |
341 | The total size of all strings, in characters. | |
342 | ||
343 | @item used-vector-slots | |
344 | The total number of elements of existing vectors. | |
345 | ||
346 | @item used-floats | |
b8d4c8d0 GM |
347 | The number of floats in use. |
348 | ||
349 | @item free-floats | |
b8d4c8d0 GM |
350 | The number of floats for which space has been obtained from the |
351 | operating system, but that are not currently being used. | |
352 | ||
353 | @item used-intervals | |
354 | The number of intervals in use. Intervals are an internal | |
355 | data structure used for representing text properties. | |
356 | ||
357 | @item free-intervals | |
358 | The number of intervals for which space has been obtained | |
359 | from the operating system, but that are not currently being used. | |
360 | ||
361 | @item used-strings | |
362 | The number of strings in use. | |
363 | ||
364 | @item free-strings | |
365 | The number of string headers for which the space was obtained from the | |
366 | operating system, but which are currently not in use. (A string | |
367 | object consists of a header and the storage for the string text | |
368 | itself; the latter is only allocated when the string is created.) | |
369 | @end table | |
370 | ||
999dd333 | 371 | If there was overflow in pure space (@pxref{Pure Storage}), |
b8d4c8d0 | 372 | @code{garbage-collect} returns @code{nil}, because a real garbage |
9658dc2a | 373 | collection cannot be done. |
b8d4c8d0 GM |
374 | @end deffn |
375 | ||
376 | @defopt garbage-collection-messages | |
377 | If this variable is non-@code{nil}, Emacs displays a message at the | |
378 | beginning and end of garbage collection. The default value is | |
999dd333 | 379 | @code{nil}. |
b8d4c8d0 GM |
380 | @end defopt |
381 | ||
382 | @defvar post-gc-hook | |
383 | This is a normal hook that is run at the end of garbage collection. | |
384 | Garbage collection is inhibited while the hook functions run, so be | |
385 | careful writing them. | |
386 | @end defvar | |
387 | ||
388 | @defopt gc-cons-threshold | |
389 | The value of this variable is the number of bytes of storage that must | |
390 | be allocated for Lisp objects after one garbage collection in order to | |
391 | trigger another garbage collection. A cons cell counts as eight bytes, | |
392 | a string as one byte per character plus a few bytes of overhead, and so | |
393 | on; space allocated to the contents of buffers does not count. Note | |
394 | that the subsequent garbage collection does not happen immediately when | |
395 | the threshold is exhausted, but only the next time the Lisp evaluator is | |
396 | called. | |
397 | ||
999dd333 | 398 | The initial threshold value is 800,000. If you specify a larger |
b8d4c8d0 GM |
399 | value, garbage collection will happen less often. This reduces the |
400 | amount of time spent garbage collecting, but increases total memory use. | |
401 | You may want to do this when running a program that creates lots of | |
402 | Lisp data. | |
403 | ||
404 | You can make collections more frequent by specifying a smaller value, | |
405 | down to 10,000. A value less than 10,000 will remain in effect only | |
406 | until the subsequent garbage collection, at which time | |
407 | @code{garbage-collect} will set the threshold back to 10,000. | |
408 | @end defopt | |
409 | ||
410 | @defopt gc-cons-percentage | |
411 | The value of this variable specifies the amount of consing before a | |
412 | garbage collection occurs, as a fraction of the current heap size. | |
413 | This criterion and @code{gc-cons-threshold} apply in parallel, and | |
414 | garbage collection occurs only when both criteria are satisfied. | |
415 | ||
416 | As the heap size increases, the time to perform a garbage collection | |
417 | increases. Thus, it can be desirable to do them less frequently in | |
418 | proportion. | |
419 | @end defopt | |
420 | ||
421 | The value returned by @code{garbage-collect} describes the amount of | |
422 | memory used by Lisp data, broken down by data type. By contrast, the | |
423 | function @code{memory-limit} provides information on the total amount of | |
424 | memory Emacs is currently using. | |
425 | ||
b8d4c8d0 GM |
426 | @defun memory-limit |
427 | This function returns the address of the last byte Emacs has allocated, | |
428 | divided by 1024. We divide the value by 1024 to make sure it fits in a | |
429 | Lisp integer. | |
430 | ||
431 | You can use this to get a general idea of how your actions affect the | |
432 | memory usage. | |
433 | @end defun | |
434 | ||
435 | @defvar memory-full | |
d15aac68 | 436 | This variable is @code{t} if Emacs is nearly out of memory for Lisp |
b8d4c8d0 GM |
437 | objects, and @code{nil} otherwise. |
438 | @end defvar | |
439 | ||
440 | @defun memory-use-counts | |
441 | This returns a list of numbers that count the number of objects | |
442 | created in this Emacs session. Each of these counters increments for | |
443 | a certain kind of object. See the documentation string for details. | |
444 | @end defun | |
445 | ||
446 | @defvar gcs-done | |
447 | This variable contains the total number of garbage collections | |
448 | done so far in this Emacs session. | |
449 | @end defvar | |
450 | ||
451 | @defvar gc-elapsed | |
452 | This variable contains the total number of seconds of elapsed time | |
453 | during garbage collection so far in this Emacs session, as a floating | |
454 | point number. | |
455 | @end defvar | |
456 | ||
457 | @node Memory Usage | |
458 | @section Memory Usage | |
459 | @cindex memory usage | |
460 | ||
461 | These functions and variables give information about the total amount | |
462 | of memory allocation that Emacs has done, broken down by data type. | |
463 | Note the difference between these and the values returned by | |
01a6dcc8 | 464 | @code{garbage-collect}; those count objects that currently exist, but |
b8d4c8d0 GM |
465 | these count the number or size of all allocations, including those for |
466 | objects that have since been freed. | |
467 | ||
468 | @defvar cons-cells-consed | |
469 | The total number of cons cells that have been allocated so far | |
470 | in this Emacs session. | |
471 | @end defvar | |
472 | ||
473 | @defvar floats-consed | |
474 | The total number of floats that have been allocated so far | |
475 | in this Emacs session. | |
476 | @end defvar | |
477 | ||
478 | @defvar vector-cells-consed | |
479 | The total number of vector cells that have been allocated so far | |
480 | in this Emacs session. | |
481 | @end defvar | |
482 | ||
483 | @defvar symbols-consed | |
484 | The total number of symbols that have been allocated so far | |
485 | in this Emacs session. | |
486 | @end defvar | |
487 | ||
488 | @defvar string-chars-consed | |
489 | The total number of string characters that have been allocated so far | |
9658dc2a | 490 | in this session. |
b8d4c8d0 GM |
491 | @end defvar |
492 | ||
493 | @defvar misc-objects-consed | |
494 | The total number of miscellaneous objects that have been allocated so | |
9658dc2a | 495 | far in this session. These include markers and overlays, plus |
b8d4c8d0 GM |
496 | certain objects not visible to users. |
497 | @end defvar | |
498 | ||
499 | @defvar intervals-consed | |
500 | The total number of intervals that have been allocated so far | |
501 | in this Emacs session. | |
502 | @end defvar | |
503 | ||
504 | @defvar strings-consed | |
505 | The total number of strings that have been allocated so far in this | |
506 | Emacs session. | |
507 | @end defvar | |
508 | ||
509 | @node Writing Emacs Primitives | |
520f2425 | 510 | @section Writing Emacs Primitives |
b8d4c8d0 GM |
511 | @cindex primitive function internals |
512 | @cindex writing Emacs primitives | |
513 | ||
514 | Lisp primitives are Lisp functions implemented in C. The details of | |
515 | interfacing the C function so that Lisp can call it are handled by a few | |
516 | C macros. The only way to really understand how to write new C code is | |
517 | to read the source, but we can explain some things here. | |
518 | ||
519 | An example of a special form is the definition of @code{or}, from | |
520 | @file{eval.c}. (An ordinary function would have the same general | |
521 | appearance.) | |
522 | ||
523 | @cindex garbage collection protection | |
524 | @smallexample | |
525 | @group | |
526 | DEFUN ("or", For, Sor, 0, UNEVALLED, 0, | |
d15aac68 CY |
527 | doc: /* Eval args until one of them yields non-nil, then return |
528 | that value. | |
529 | The remaining args are not evalled at all. | |
b8d4c8d0 GM |
530 | If all args return nil, return nil. |
531 | @end group | |
532 | @group | |
533 | usage: (or CONDITIONS ...) */) | |
b88746ba | 534 | (Lisp_Object args) |
b8d4c8d0 GM |
535 | @{ |
536 | register Lisp_Object val = Qnil; | |
537 | struct gcpro gcpro1; | |
538 | @end group | |
539 | ||
540 | @group | |
541 | GCPRO1 (args); | |
542 | @end group | |
543 | ||
544 | @group | |
545 | while (CONSP (args)) | |
546 | @{ | |
7d2d7cc0 | 547 | val = eval_sub (XCAR (args)); |
b8d4c8d0 GM |
548 | if (!NILP (val)) |
549 | break; | |
550 | args = XCDR (args); | |
551 | @} | |
552 | @end group | |
553 | ||
554 | @group | |
555 | UNGCPRO; | |
556 | return val; | |
557 | @} | |
558 | @end group | |
559 | @end smallexample | |
560 | ||
561 | @cindex @code{DEFUN}, C macro to define Lisp primitives | |
562 | Let's start with a precise explanation of the arguments to the | |
563 | @code{DEFUN} macro. Here is a template for them: | |
564 | ||
565 | @example | |
566 | DEFUN (@var{lname}, @var{fname}, @var{sname}, @var{min}, @var{max}, @var{interactive}, @var{doc}) | |
567 | @end example | |
568 | ||
569 | @table @var | |
570 | @item lname | |
571 | This is the name of the Lisp symbol to define as the function name; in | |
572 | the example above, it is @code{or}. | |
573 | ||
574 | @item fname | |
d15aac68 CY |
575 | This is the C function name for this function. This is the name that |
576 | is used in C code for calling the function. The name is, by | |
577 | convention, @samp{F} prepended to the Lisp name, with all dashes | |
578 | (@samp{-}) in the Lisp name changed to underscores. Thus, to call | |
579 | this function from C code, call @code{For}. | |
b8d4c8d0 GM |
580 | |
581 | @item sname | |
582 | This is a C variable name to use for a structure that holds the data for | |
583 | the subr object that represents the function in Lisp. This structure | |
584 | conveys the Lisp symbol name to the initialization routine that will | |
585 | create the symbol and store the subr object as its definition. By | |
586 | convention, this name is always @var{fname} with @samp{F} replaced with | |
587 | @samp{S}. | |
588 | ||
589 | @item min | |
590 | This is the minimum number of arguments that the function requires. The | |
591 | function @code{or} allows a minimum of zero arguments. | |
592 | ||
593 | @item max | |
594 | This is the maximum number of arguments that the function accepts, if | |
595 | there is a fixed maximum. Alternatively, it can be @code{UNEVALLED}, | |
596 | indicating a special form that receives unevaluated arguments, or | |
597 | @code{MANY}, indicating an unlimited number of evaluated arguments (the | |
598 | equivalent of @code{&rest}). Both @code{UNEVALLED} and @code{MANY} are | |
9658dc2a GM |
599 | macros. If @var{max} is a number, it must be more than @var{min} but |
600 | less than 8. | |
b8d4c8d0 GM |
601 | |
602 | @item interactive | |
603 | This is an interactive specification, a string such as might be used as | |
604 | the argument of @code{interactive} in a Lisp function. In the case of | |
605 | @code{or}, it is 0 (a null pointer), indicating that @code{or} cannot be | |
606 | called interactively. A value of @code{""} indicates a function that | |
2c30e468 EZ |
607 | should receive no arguments when called interactively. If the value |
608 | begins with a @samp{(}, the string is evaluated as a Lisp form. | |
7d2d7cc0 GM |
609 | For examples of the last two forms, see @code{widen} and |
610 | @code{narrow-to-region} in @file{editfns.c}. | |
b8d4c8d0 GM |
611 | |
612 | @item doc | |
613 | This is the documentation string. It uses C comment syntax rather | |
614 | than C string syntax because comment syntax requires nothing special | |
615 | to include multiple lines. The @samp{doc:} identifies the comment | |
616 | that follows as the documentation string. The @samp{/*} and @samp{*/} | |
617 | delimiters that begin and end the comment are not part of the | |
618 | documentation string. | |
619 | ||
620 | If the last line of the documentation string begins with the keyword | |
621 | @samp{usage:}, the rest of the line is treated as the argument list | |
622 | for documentation purposes. This way, you can use different argument | |
623 | names in the documentation string from the ones used in the C code. | |
624 | @samp{usage:} is required if the function has an unlimited number of | |
625 | arguments. | |
626 | ||
627 | All the usual rules for documentation strings in Lisp code | |
628 | (@pxref{Documentation Tips}) apply to C code documentation strings | |
629 | too. | |
630 | @end table | |
631 | ||
b88746ba | 632 | After the call to the @code{DEFUN} macro, you must write the |
d15aac68 CY |
633 | argument list for the C function, including the types for the |
634 | arguments. If the primitive accepts a fixed maximum number of Lisp | |
635 | arguments, there must be one C argument for each Lisp argument, and | |
636 | each argument must be of type @code{Lisp_Object}. (Various macros and | |
637 | functions for creating values of type @code{Lisp_Object} are declared | |
638 | in the file @file{lisp.h}.) If the primitive has no upper limit on | |
639 | the number of Lisp arguments, it must have exactly two C arguments: | |
640 | the first is the number of Lisp arguments, and the second is the | |
641 | address of a block containing their values. These have types | |
642 | @code{int} and @w{@code{Lisp_Object *}} respectively. | |
b8d4c8d0 GM |
643 | |
644 | @cindex @code{GCPRO} and @code{UNGCPRO} | |
645 | @cindex protect C variables from garbage collection | |
646 | Within the function @code{For} itself, note the use of the macros | |
d15aac68 CY |
647 | @code{GCPRO1} and @code{UNGCPRO}. These macros are defined for the |
648 | sake of the few platforms which do not use Emacs' default | |
649 | stack-marking garbage collector. The @code{GCPRO1} macro ``protects'' | |
650 | a variable from garbage collection, explicitly informing the garbage | |
651 | collector that that variable and all its contents must be as | |
652 | accessible. GC protection is necessary in any function which can | |
653 | perform Lisp evaluation by calling @code{eval_sub} or @code{Feval} as | |
654 | a subroutine, either directly or indirectly. | |
b8d4c8d0 GM |
655 | |
656 | It suffices to ensure that at least one pointer to each object is | |
d15aac68 | 657 | GC-protected. Thus, a particular local variable can do without |
b8d4c8d0 | 658 | protection if it is certain that the object it points to will be |
7d2d7cc0 | 659 | preserved by some other pointer (such as another local variable that |
d15aac68 CY |
660 | has a @code{GCPRO}). Otherwise, the local variable needs a |
661 | @code{GCPRO}. | |
b8d4c8d0 GM |
662 | |
663 | The macro @code{GCPRO1} protects just one local variable. If you | |
664 | want to protect two variables, use @code{GCPRO2} instead; repeating | |
665 | @code{GCPRO1} will not work. Macros @code{GCPRO3}, @code{GCPRO4}, | |
666 | @code{GCPRO5}, and @code{GCPRO6} also exist. All these macros | |
667 | implicitly use local variables such as @code{gcpro1}; you must declare | |
668 | these explicitly, with type @code{struct gcpro}. Thus, if you use | |
669 | @code{GCPRO2}, you must declare @code{gcpro1} and @code{gcpro2}. | |
b8d4c8d0 GM |
670 | |
671 | @code{UNGCPRO} cancels the protection of the variables that are | |
672 | protected in the current function. It is necessary to do this | |
673 | explicitly. | |
674 | ||
b8d4c8d0 GM |
675 | You must not use C initializers for static or global variables unless |
676 | the variables are never written once Emacs is dumped. These variables | |
677 | with initializers are allocated in an area of memory that becomes | |
678 | read-only (on certain operating systems) as a result of dumping Emacs. | |
679 | @xref{Pure Storage}. | |
680 | ||
b8d4c8d0 GM |
681 | @cindex @code{defsubr}, Lisp symbol for a primitive |
682 | Defining the C function is not enough to make a Lisp primitive | |
683 | available; you must also create the Lisp symbol for the primitive and | |
684 | store a suitable subr object in its function cell. The code looks like | |
685 | this: | |
686 | ||
687 | @example | |
7d2d7cc0 | 688 | defsubr (&@var{sname}); |
b8d4c8d0 GM |
689 | @end example |
690 | ||
691 | @noindent | |
7d2d7cc0 | 692 | Here @var{sname} is the name you used as the third argument to @code{DEFUN}. |
b8d4c8d0 GM |
693 | |
694 | If you add a new primitive to a file that already has Lisp primitives | |
695 | defined in it, find the function (near the end of the file) named | |
696 | @code{syms_of_@var{something}}, and add the call to @code{defsubr} | |
697 | there. If the file doesn't have this function, or if you create a new | |
698 | file, add to it a @code{syms_of_@var{filename}} (e.g., | |
699 | @code{syms_of_myfile}). Then find the spot in @file{emacs.c} where all | |
700 | of these functions are called, and add a call to | |
701 | @code{syms_of_@var{filename}} there. | |
702 | ||
703 | @anchor{Defining Lisp variables in C} | |
704 | @vindex byte-boolean-vars | |
705 | @cindex defining Lisp variables in C | |
706 | @cindex @code{DEFVAR_INT}, @code{DEFVAR_LISP}, @code{DEFVAR_BOOL} | |
707 | The function @code{syms_of_@var{filename}} is also the place to define | |
708 | any C variables that are to be visible as Lisp variables. | |
709 | @code{DEFVAR_LISP} makes a C variable of type @code{Lisp_Object} visible | |
710 | in Lisp. @code{DEFVAR_INT} makes a C variable of type @code{int} | |
711 | visible in Lisp with a value that is always an integer. | |
712 | @code{DEFVAR_BOOL} makes a C variable of type @code{int} visible in Lisp | |
713 | with a value that is either @code{t} or @code{nil}. Note that variables | |
714 | defined with @code{DEFVAR_BOOL} are automatically added to the list | |
715 | @code{byte-boolean-vars} used by the byte compiler. | |
716 | ||
7d2d7cc0 GM |
717 | @cindex defining customization variables in C |
718 | If you want to make a Lisp variables that is defined in C behave | |
719 | like one declared with @code{defcustom}, add an appropriate entry to | |
720 | @file{cus-start.el}. | |
721 | ||
b8d4c8d0 GM |
722 | @cindex @code{staticpro}, protection from GC |
723 | If you define a file-scope C variable of type @code{Lisp_Object}, | |
724 | you must protect it from garbage-collection by calling @code{staticpro} | |
725 | in @code{syms_of_@var{filename}}, like this: | |
726 | ||
727 | @example | |
728 | staticpro (&@var{variable}); | |
729 | @end example | |
730 | ||
731 | Here is another example function, with more complicated arguments. | |
732 | This comes from the code in @file{window.c}, and it demonstrates the use | |
733 | of macros and functions to manipulate Lisp objects. | |
734 | ||
735 | @smallexample | |
736 | @group | |
737 | DEFUN ("coordinates-in-window-p", Fcoordinates_in_window_p, | |
7d2d7cc0 GM |
738 | Scoordinates_in_window_p, 2, 2, 0, |
739 | doc: /* Return non-nil if COORDINATES are in WINDOW. | |
740 | ... | |
b8d4c8d0 GM |
741 | @end group |
742 | @group | |
7d2d7cc0 GM |
743 | or `right-margin' is returned. */) |
744 | (register Lisp_Object coordinates, Lisp_Object window) | |
b8d4c8d0 | 745 | @{ |
7d2d7cc0 GM |
746 | struct window *w; |
747 | struct frame *f; | |
b8d4c8d0 | 748 | int x, y; |
7d2d7cc0 | 749 | Lisp_Object lx, ly; |
b8d4c8d0 GM |
750 | @end group |
751 | ||
752 | @group | |
7d2d7cc0 GM |
753 | CHECK_LIVE_WINDOW (window); |
754 | w = XWINDOW (window); | |
755 | f = XFRAME (w->frame); | |
756 | CHECK_CONS (coordinates); | |
757 | lx = Fcar (coordinates); | |
758 | ly = Fcdr (coordinates); | |
759 | CHECK_NUMBER_OR_FLOAT (lx); | |
760 | CHECK_NUMBER_OR_FLOAT (ly); | |
761 | x = FRAME_PIXEL_X_FROM_CANON_X (f, lx) + FRAME_INTERNAL_BORDER_WIDTH(f); | |
762 | y = FRAME_PIXEL_Y_FROM_CANON_Y (f, ly) + FRAME_INTERNAL_BORDER_WIDTH(f); | |
b8d4c8d0 GM |
763 | @end group |
764 | ||
765 | @group | |
7d2d7cc0 | 766 | switch (coordinates_in_window (w, x, y)) |
b8d4c8d0 | 767 | @{ |
7d2d7cc0 | 768 | case ON_NOTHING: /* NOT in window at all. */ |
b8d4c8d0 GM |
769 | return Qnil; |
770 | @end group | |
771 | ||
7d2d7cc0 | 772 | ... |
b8d4c8d0 GM |
773 | |
774 | @group | |
7d2d7cc0 | 775 | case ON_MODE_LINE: /* In mode line of window. */ |
b8d4c8d0 GM |
776 | return Qmode_line; |
777 | @end group | |
778 | ||
7d2d7cc0 GM |
779 | ... |
780 | ||
b8d4c8d0 | 781 | @group |
7d2d7cc0 GM |
782 | case ON_SCROLL_BAR: /* On scroll-bar of window. */ |
783 | /* Historically we are supposed to return nil in this case. */ | |
784 | return Qnil; | |
b8d4c8d0 GM |
785 | @end group |
786 | ||
787 | @group | |
788 | default: | |
789 | abort (); | |
790 | @} | |
791 | @} | |
792 | @end group | |
793 | @end smallexample | |
794 | ||
795 | Note that C code cannot call functions by name unless they are defined | |
796 | in C. The way to call a function written in Lisp is to use | |
797 | @code{Ffuncall}, which embodies the Lisp function @code{funcall}. Since | |
798 | the Lisp function @code{funcall} accepts an unlimited number of | |
799 | arguments, in C it takes two: the number of Lisp-level arguments, and a | |
800 | one-dimensional array containing their values. The first Lisp-level | |
801 | argument is the Lisp function to call, and the rest are the arguments to | |
802 | pass to it. Since @code{Ffuncall} can call the evaluator, you must | |
803 | protect pointers from garbage collection around the call to | |
804 | @code{Ffuncall}. | |
805 | ||
806 | The C functions @code{call0}, @code{call1}, @code{call2}, and so on, | |
807 | provide handy ways to call a Lisp function conveniently with a fixed | |
808 | number of arguments. They work by calling @code{Ffuncall}. | |
809 | ||
810 | @file{eval.c} is a very good file to look through for examples; | |
811 | @file{lisp.h} contains the definitions for some important macros and | |
812 | functions. | |
813 | ||
814 | If you define a function which is side-effect free, update the code | |
7d2d7cc0 | 815 | in @file{byte-opt.el} that binds @code{side-effect-free-fns} and |
b8d4c8d0 GM |
816 | @code{side-effect-and-error-free-fns} so that the compiler optimizer |
817 | knows about it. | |
818 | ||
819 | @node Object Internals | |
520f2425 | 820 | @section Object Internals |
b8d4c8d0 GM |
821 | @cindex object internals |
822 | ||
7d2d7cc0 | 823 | @c FIXME Is this still true? Does --with-wide-int affect anything? |
b8d4c8d0 GM |
824 | GNU Emacs Lisp manipulates many different types of data. The actual |
825 | data are stored in a heap and the only access that programs have to it | |
c773345a CY |
826 | is through pointers. Each pointer is 32 bits wide on 32-bit machines, |
827 | and 64 bits wide on 64-bit machines; three of these bits are used for | |
828 | the tag that identifies the object's type, and the remainder are used | |
829 | to address the object. | |
b8d4c8d0 GM |
830 | |
831 | Because Lisp objects are represented as tagged pointers, it is always | |
832 | possible to determine the Lisp data type of any object. The C data type | |
833 | @code{Lisp_Object} can hold any Lisp object of any data type. Ordinary | |
834 | variables have type @code{Lisp_Object}, which means they can hold any | |
835 | type of Lisp value; you can determine the actual data type only at run | |
836 | time. The same is true for function arguments; if you want a function | |
837 | to accept only a certain type of argument, you must check the type | |
838 | explicitly using a suitable predicate (@pxref{Type Predicates}). | |
839 | @cindex type checking internals | |
840 | ||
841 | @menu | |
842 | * Buffer Internals:: Components of a buffer structure. | |
843 | * Window Internals:: Components of a window structure. | |
844 | * Process Internals:: Components of a process structure. | |
845 | @end menu | |
846 | ||
847 | @node Buffer Internals | |
520f2425 | 848 | @subsection Buffer Internals |
b8d4c8d0 GM |
849 | @cindex internals, of buffer |
850 | @cindex buffer internals | |
851 | ||
7d2d7cc0 GM |
852 | Two structures (see @file{buffer.h}) are used to represent buffers |
853 | in C. The @code{buffer_text} structure contains fields describing the | |
854 | text of a buffer; the @code{buffer} structure holds other fields. In | |
855 | the case of indirect buffers, two or more @code{buffer} structures | |
856 | reference the same @code{buffer_text} structure. | |
b8d4c8d0 | 857 | |
c773345a | 858 | Here are some of the fields in @code{struct buffer_text}: |
b8d4c8d0 GM |
859 | |
860 | @table @code | |
861 | @item beg | |
c773345a | 862 | The address of the buffer contents. |
b8d4c8d0 GM |
863 | |
864 | @item gpt | |
c773345a CY |
865 | @itemx gpt_byte |
866 | The character and byte positions of the buffer gap. @xref{Buffer | |
867 | Gap}. | |
b8d4c8d0 GM |
868 | |
869 | @item z | |
c773345a CY |
870 | @itemx z_byte |
871 | The character and byte positions of the end of the buffer text. | |
b8d4c8d0 GM |
872 | |
873 | @item gap_size | |
c773345a | 874 | The size of buffer's gap. @xref{Buffer Gap}. |
b8d4c8d0 GM |
875 | |
876 | @item modiff | |
c773345a CY |
877 | @itemx save_modiff |
878 | @itemx chars_modiff | |
879 | @itemx overlay_modiff | |
880 | These fields count the number of buffer-modification events performed | |
881 | in this buffer. @code{modiff} is incremented after each | |
882 | buffer-modification event, and is never otherwise changed; | |
883 | @code{save_modiff} contains the value of @code{modiff} the last time | |
884 | the buffer was visited or saved; @code{chars_modiff} counts only | |
885 | modifications to the characters in the buffer, ignoring all other | |
886 | kinds of changes; and @code{overlay_modiff} counts only modifications | |
887 | to the overlays. | |
b8d4c8d0 GM |
888 | |
889 | @item beg_unchanged | |
c773345a CY |
890 | @itemx end_unchanged |
891 | The number of characters at the start and end of the text that are | |
892 | known to be unchanged since the last complete redisplay. | |
b8d4c8d0 GM |
893 | |
894 | @item unchanged_modified | |
c773345a CY |
895 | @itemx overlay_unchanged_modified |
896 | The values of @code{modiff} and @code{overlay_modiff}, respectively, | |
da6062e6 | 897 | after the last complete redisplay. If their current values match |
c773345a | 898 | @code{modiff} or @code{overlay_modiff}, that means |
b8d4c8d0 GM |
899 | @code{beg_unchanged} and @code{end_unchanged} contain no useful |
900 | information. | |
901 | ||
902 | @item markers | |
903 | The markers that refer to this buffer. This is actually a single | |
904 | marker, and successive elements in its marker @code{chain} are the other | |
905 | markers referring to this buffer text. | |
906 | ||
907 | @item intervals | |
c773345a | 908 | The interval tree which records the text properties of this buffer. |
b8d4c8d0 GM |
909 | @end table |
910 | ||
c773345a | 911 | Some of the fields of @code{struct buffer} are: |
b8d4c8d0 GM |
912 | |
913 | @table @code | |
7d2d7cc0 GM |
914 | @item header |
915 | A @code{struct vectorlike_header} structure where @code{header.next} | |
916 | points to the next buffer, in the chain of all buffers (including | |
c773345a CY |
917 | killed buffers). This chain is used only for garbage collection, in |
918 | order to collect killed buffers properly. Note that vectors, and most | |
919 | kinds of objects allocated as vectors, are all on one chain, but | |
920 | buffers are on a separate chain of their own. | |
b8d4c8d0 GM |
921 | |
922 | @item own_text | |
c773345a CY |
923 | A @code{struct buffer_text} structure that ordinarily holds the buffer |
924 | contents. In indirect buffers, this field is not used. | |
b8d4c8d0 GM |
925 | |
926 | @item text | |
c773345a CY |
927 | A pointer to the @code{buffer_text} structure for this buffer. In an |
928 | ordinary buffer, this is the @code{own_text} field above. In an | |
929 | indirect buffer, this is the @code{own_text} field of the base buffer. | |
b8d4c8d0 GM |
930 | |
931 | @item pt | |
c773345a CY |
932 | @itemx pt_byte |
933 | The character and byte positions of point in a buffer. | |
b8d4c8d0 GM |
934 | |
935 | @item begv | |
c773345a CY |
936 | @itemx begv_byte |
937 | The character and byte positions of the beginning of the accessible | |
938 | range of text in the buffer. | |
b8d4c8d0 GM |
939 | |
940 | @item zv | |
c773345a CY |
941 | @itemx zv_byte |
942 | The character and byte positions of the end of the accessible range of | |
943 | text in the buffer. | |
b8d4c8d0 GM |
944 | |
945 | @item base_buffer | |
946 | In an indirect buffer, this points to the base buffer. In an ordinary | |
947 | buffer, it is null. | |
948 | ||
c773345a CY |
949 | @item local_flags |
950 | This field contains flags indicating that certain variables are local | |
951 | in this buffer. Such variables are declared in the C code using | |
952 | @code{DEFVAR_PER_BUFFER}, and their buffer-local bindings are stored | |
953 | in fields in the buffer structure itself. (Some of these fields are | |
b8d4c8d0 GM |
954 | described in this table.) |
955 | ||
956 | @item modtime | |
c773345a CY |
957 | The modification time of the visited file. It is set when the file is |
958 | written or read. Before writing the buffer into a file, this field is | |
959 | compared to the modification time of the file to see if the file has | |
960 | changed on disk. @xref{Buffer Modification}. | |
b8d4c8d0 GM |
961 | |
962 | @item auto_save_modified | |
c773345a | 963 | The time when the buffer was last auto-saved. |
b8d4c8d0 GM |
964 | |
965 | @item last_window_start | |
c773345a CY |
966 | The @code{window-start} position in the buffer as of the last time the |
967 | buffer was displayed in a window. | |
b8d4c8d0 GM |
968 | |
969 | @item clip_changed | |
c773345a CY |
970 | This flag indicates that narrowing has changed in the buffer. |
971 | @xref{Narrowing}. | |
b8d4c8d0 GM |
972 | |
973 | @item prevent_redisplay_optimizations_p | |
c773345a CY |
974 | This flag indicates that redisplay optimizations should not be used to |
975 | display this buffer. | |
b8d4c8d0 | 976 | |
c773345a CY |
977 | @item overlay_center |
978 | This field holds the current overlay center position. @xref{Managing | |
979 | Overlays}. | |
980 | ||
981 | @item overlays_before | |
982 | @itemx overlays_after | |
983 | These fields hold, respectively, a list of overlays that end at or | |
984 | before the current overlay center, and a list of overlays that end | |
985 | after the current overlay center. @xref{Managing Overlays}. | |
986 | @code{overlays_before} is sorted in order of decreasing end position, | |
987 | and @code{overlays_after} is sorted in order of increasing beginning | |
988 | position. | |
b8d4c8d0 | 989 | |
7d2d7cc0 GM |
990 | @c FIXME? the following are now all Lisp_Object BUFFER_INTERNAL_FIELD (foo). |
991 | ||
b8d4c8d0 | 992 | @item name |
c773345a CY |
993 | A Lisp string that names the buffer. It is guaranteed to be unique. |
994 | @xref{Buffer Names}. | |
b8d4c8d0 | 995 | |
c773345a CY |
996 | @item save_length |
997 | The length of the file this buffer is visiting, when last read or | |
998 | saved. This and other fields concerned with saving are not kept in | |
999 | the @code{buffer_text} structure because indirect buffers are never | |
1000 | saved. | |
b8d4c8d0 GM |
1001 | |
1002 | @item directory | |
c773345a CY |
1003 | The directory for expanding relative file names. This is the value of |
1004 | the buffer-local variable @code{default-directory} (@pxref{File Name Expansion}). | |
b8d4c8d0 | 1005 | |
c773345a CY |
1006 | @item filename |
1007 | The name of the file visited in this buffer, or @code{nil}. This is | |
1008 | the value of the buffer-local variable @code{buffer-file-name} | |
1009 | (@pxref{Buffer File Name}). | |
b8d4c8d0 | 1010 | |
c773345a CY |
1011 | @item undo_list |
1012 | @itemx backed_up | |
1013 | @itemx auto_save_file_name | |
7d2d7cc0 | 1014 | @itemx auto_save_file_format |
c773345a CY |
1015 | @itemx read_only |
1016 | @itemx file_format | |
1017 | @itemx file_truename | |
1018 | @itemx invisibility_spec | |
1019 | @itemx display_count | |
1020 | @itemx display_time | |
1021 | These fields store the values of Lisp variables that are automatically | |
1022 | buffer-local (@pxref{Buffer-Local Variables}), whose corresponding | |
1023 | variable names have the additional prefix @code{buffer-} and have | |
1024 | underscores replaced with dashes. For instance, @code{undo_list} | |
29157371 | 1025 | stores the value of @code{buffer-undo-list}. |
b8d4c8d0 GM |
1026 | |
1027 | @item mark | |
c773345a CY |
1028 | The mark for the buffer. The mark is a marker, hence it is also |
1029 | included on the list @code{markers}. @xref{The Mark}. | |
b8d4c8d0 GM |
1030 | |
1031 | @item local_var_alist | |
c773345a CY |
1032 | The association list describing the buffer-local variable bindings of |
1033 | this buffer, not including the built-in buffer-local bindings that | |
1034 | have special slots in the buffer object. (Those slots are omitted | |
1035 | from this table.) @xref{Buffer-Local Variables}. | |
b8d4c8d0 GM |
1036 | |
1037 | @item major_mode | |
1038 | Symbol naming the major mode of this buffer, e.g., @code{lisp-mode}. | |
1039 | ||
1040 | @item mode_name | |
c773345a | 1041 | Pretty name of the major mode, e.g., @code{"Lisp"}. |
b8d4c8d0 GM |
1042 | |
1043 | @item keymap | |
c773345a CY |
1044 | @itemx abbrev_table |
1045 | @itemx syntax_table | |
1046 | @itemx category_table | |
1047 | @itemx display_table | |
1048 | These fields store the buffer's local keymap (@pxref{Keymaps}), abbrev | |
1049 | table (@pxref{Abbrev Tables}), syntax table (@pxref{Syntax Tables}), | |
1050 | category table (@pxref{Categories}), and display table (@pxref{Display | |
1051 | Tables}). | |
b8d4c8d0 GM |
1052 | |
1053 | @item downcase_table | |
c773345a CY |
1054 | @itemx upcase_table |
1055 | @itemx case_canon_table | |
1056 | These fields store the conversion tables for converting text to lower | |
1057 | case, upper case, and for canonicalizing text for case-fold search. | |
b8d4c8d0 GM |
1058 | @xref{Case Tables}. |
1059 | ||
b8d4c8d0 GM |
1060 | @item minor_modes |
1061 | An alist of the minor modes of this buffer. | |
1062 | ||
b8d4c8d0 | 1063 | @item pt_marker |
c773345a CY |
1064 | @itemx begv_marker |
1065 | @itemx zv_marker | |
1066 | These fields are only used in an indirect buffer, or in a buffer that | |
1067 | is the base of an indirect buffer. Each holds a marker that records | |
1068 | @code{pt}, @code{begv}, and @code{zv} respectively, for this buffer | |
b8d4c8d0 GM |
1069 | when the buffer is not current. |
1070 | ||
c773345a CY |
1071 | @item mode_line_format |
1072 | @itemx header_line_format | |
1073 | @itemx case_fold_search | |
1074 | @itemx tab_width | |
1075 | @itemx fill_column | |
1076 | @itemx left_margin | |
1077 | @itemx auto_fill_function | |
c773345a CY |
1078 | @itemx truncate_lines |
1079 | @itemx word_wrap | |
1080 | @itemx ctl_arrow | |
7d2d7cc0 GM |
1081 | @itemx bidi_display_reordering |
1082 | @itemx bidi_paragraph_direction | |
c773345a CY |
1083 | @itemx selective_display |
1084 | @itemx selective_display_ellipses | |
1085 | @itemx overwrite_mode | |
1086 | @itemx abbrev_mode | |
c773345a CY |
1087 | @itemx mark_active |
1088 | @itemx enable_multibyte_characters | |
1089 | @itemx buffer_file_coding_system | |
c773345a CY |
1090 | @itemx cache_long_line_scans |
1091 | @itemx point_before_scroll | |
1092 | @itemx left_fringe_width | |
1093 | @itemx right_fringe_width | |
1094 | @itemx fringes_outside_margins | |
1095 | @itemx scroll_bar_width | |
1096 | @itemx indicate_empty_lines | |
1097 | @itemx indicate_buffer_boundaries | |
1098 | @itemx fringe_indicator_alist | |
1099 | @itemx fringe_cursor_alist | |
1100 | @itemx scroll_up_aggressively | |
1101 | @itemx scroll_down_aggressively | |
1102 | @itemx cursor_type | |
1103 | @itemx cursor_in_non_selected_windows | |
1104 | These fields store the values of Lisp variables that are automatically | |
1105 | buffer-local (@pxref{Buffer-Local Variables}), whose corresponding | |
1106 | variable names have underscores replaced with dashes. For instance, | |
1107 | @code{mode_line_format} stores the value of @code{mode-line-format}. | |
b8d4c8d0 GM |
1108 | |
1109 | @item last_selected_window | |
1110 | This is the last window that was selected with this buffer in it, or @code{nil} | |
1111 | if that window no longer displays this buffer. | |
b8d4c8d0 GM |
1112 | @end table |
1113 | ||
1114 | @node Window Internals | |
520f2425 | 1115 | @subsection Window Internals |
b8d4c8d0 GM |
1116 | @cindex internals, of window |
1117 | @cindex window internals | |
1118 | ||
7d2d7cc0 GM |
1119 | The fields of a window (for a complete list, see the definition of |
1120 | @code{struct window} in @file{window.h}) include: | |
b8d4c8d0 GM |
1121 | |
1122 | @table @code | |
1123 | @item frame | |
1124 | The frame that this window is on. | |
1125 | ||
1126 | @item mini_p | |
1127 | Non-@code{nil} if this window is a minibuffer window. | |
1128 | ||
1129 | @item parent | |
1130 | Internally, Emacs arranges windows in a tree; each group of siblings has | |
1131 | a parent window whose area includes all the siblings. This field points | |
1132 | to a window's parent. | |
1133 | ||
1134 | Parent windows do not display buffers, and play little role in display | |
1135 | except to shape their child windows. Emacs Lisp programs usually have | |
1136 | no access to the parent windows; they operate on the windows at the | |
1137 | leaves of the tree, which actually display buffers. | |
1138 | ||
b8d4c8d0 | 1139 | @item hchild |
ee2d5b5e CY |
1140 | @itemx vchild |
1141 | These fields contain the window's leftmost child and its topmost child | |
1142 | respectively. @code{hchild} is used if the window is subdivided | |
1143 | horizontally by child windows, and @code{vchild} if it is subdivided | |
7d2d7cc0 GM |
1144 | vertically. In a live window, only one of @code{hchild}, @code{vchild}, |
1145 | and @code{buffer} (q.v.) is non-@code{nil}. | |
b8d4c8d0 GM |
1146 | |
1147 | @item next | |
ee2d5b5e CY |
1148 | @itemx prev |
1149 | The next sibling and previous sibling of this window. @code{next} is | |
520f2425 GM |
1150 | @code{nil} if the window is the right-most or bottom-most in its group; |
1151 | @code{prev} is @code{nil} if it is the left-most or top-most in its | |
ee2d5b5e CY |
1152 | group. |
1153 | ||
1154 | @item left_col | |
1155 | The left-hand edge of the window, measured in columns, relative to the | |
1156 | leftmost column in the frame (column 0). | |
1157 | ||
1158 | @item top_line | |
1159 | The top edge of the window, measured in lines, relative to the topmost | |
1160 | line in the frame (line 0). | |
1161 | ||
1162 | @item total_cols | |
1163 | @itemx total_lines | |
1164 | The width and height of the window, measured in columns and lines | |
1165 | respectively. The width includes the scroll bar and fringes, and/or | |
1166 | the separator line on the right of the window (if any). | |
b8d4c8d0 GM |
1167 | |
1168 | @item buffer | |
ee2d5b5e | 1169 | The buffer that the window is displaying. |
b8d4c8d0 GM |
1170 | |
1171 | @item start | |
ee2d5b5e CY |
1172 | A marker pointing to the position in the buffer that is the first |
1173 | character displayed in the window. | |
b8d4c8d0 GM |
1174 | |
1175 | @item pointm | |
1176 | @cindex window point internals | |
1177 | This is the value of point in the current buffer when this window is | |
1178 | selected; when it is not selected, it retains its previous value. | |
1179 | ||
1180 | @item force_start | |
1181 | If this flag is non-@code{nil}, it says that the window has been | |
1182 | scrolled explicitly by the Lisp program. This affects what the next | |
1183 | redisplay does if point is off the screen: instead of scrolling the | |
1184 | window to show the text around point, it moves point to a location that | |
1185 | is on the screen. | |
1186 | ||
1187 | @item frozen_window_start_p | |
1188 | This field is set temporarily to 1 to indicate to redisplay that | |
1189 | @code{start} of this window should not be changed, even if point | |
1190 | gets invisible. | |
1191 | ||
1192 | @item start_at_line_beg | |
1193 | Non-@code{nil} means current value of @code{start} was the beginning of a line | |
1194 | when it was chosen. | |
1195 | ||
b8d4c8d0 GM |
1196 | @item use_time |
1197 | This is the last time that the window was selected. The function | |
1198 | @code{get-lru-window} uses this field. | |
1199 | ||
1200 | @item sequence_number | |
1201 | A unique number assigned to this window when it was created. | |
1202 | ||
1203 | @item last_modified | |
1204 | The @code{modiff} field of the window's buffer, as of the last time | |
1205 | a redisplay completed in this window. | |
1206 | ||
1207 | @item last_overlay_modified | |
1208 | The @code{overlay_modiff} field of the window's buffer, as of the last | |
1209 | time a redisplay completed in this window. | |
1210 | ||
1211 | @item last_point | |
1212 | The buffer's value of point, as of the last time a redisplay completed | |
1213 | in this window. | |
1214 | ||
1215 | @item last_had_star | |
1216 | A non-@code{nil} value means the window's buffer was ``modified'' when the | |
1217 | window was last updated. | |
1218 | ||
1219 | @item vertical_scroll_bar | |
1220 | This window's vertical scroll bar. | |
1221 | ||
7d2d7cc0 GM |
1222 | @item left_margin_cols |
1223 | @itemx right_margin_cols | |
ee2d5b5e | 1224 | The widths of the left and right margins in this window. A value of |
7d2d7cc0 GM |
1225 | @code{nil} means no margin. |
1226 | ||
1227 | @item left_fringe_width | |
1228 | @itemx right_fringe_width | |
1229 | The widths of the left and right fringes in this window. A value of | |
1230 | @code{nil} or @code{t} means use the values of the frame. | |
1231 | ||
1232 | @item fringes_outside_margins | |
1233 | A non-@code{nil} value means the fringes outside the display margins; | |
1234 | othersize they are between the margin and the text. | |
b8d4c8d0 GM |
1235 | |
1236 | @item window_end_pos | |
1237 | This is computed as @code{z} minus the buffer position of the last glyph | |
1238 | in the current matrix of the window. The value is only valid if | |
1239 | @code{window_end_valid} is not @code{nil}. | |
1240 | ||
1241 | @item window_end_bytepos | |
1242 | The byte position corresponding to @code{window_end_pos}. | |
1243 | ||
1244 | @item window_end_vpos | |
1245 | The window-relative vertical position of the line containing | |
1246 | @code{window_end_pos}. | |
1247 | ||
1248 | @item window_end_valid | |
1249 | This field is set to a non-@code{nil} value if @code{window_end_pos} is truly | |
7d2d7cc0 | 1250 | valid. This is @code{nil} if nontrivial redisplay is pre-empted, since in that |
b8d4c8d0 GM |
1251 | case the display that @code{window_end_pos} was computed for did not get |
1252 | onto the screen. | |
1253 | ||
b8d4c8d0 GM |
1254 | @item cursor |
1255 | A structure describing where the cursor is in this window. | |
1256 | ||
1257 | @item last_cursor | |
1258 | The value of @code{cursor} as of the last redisplay that finished. | |
1259 | ||
1260 | @item phys_cursor | |
1261 | A structure describing where the cursor of this window physically is. | |
1262 | ||
1263 | @item phys_cursor_type | |
7d2d7cc0 GM |
1264 | @c FIXME What is this? |
1265 | @c itemx phys_cursor_ascent | |
1266 | @itemx phys_cursor_height | |
1267 | @itemx phys_cursor_width | |
1268 | The type, height, and width of the cursor that was last displayed on | |
1269 | this window. | |
b8d4c8d0 GM |
1270 | |
1271 | @item phys_cursor_on_p | |
1272 | This field is non-zero if the cursor is physically on. | |
1273 | ||
1274 | @item cursor_off_p | |
7d2d7cc0 GM |
1275 | Non-zero means the cursor in this window is logically off. This is |
1276 | used for blinking the cursor. | |
b8d4c8d0 GM |
1277 | |
1278 | @item last_cursor_off_p | |
1279 | This field contains the value of @code{cursor_off_p} as of the time of | |
1280 | the last redisplay. | |
1281 | ||
1282 | @item must_be_updated_p | |
1283 | This is set to 1 during redisplay when this window must be updated. | |
1284 | ||
1285 | @item hscroll | |
1286 | This is the number of columns that the display in the window is scrolled | |
1287 | horizontally to the left. Normally, this is 0. | |
1288 | ||
1289 | @item vscroll | |
1290 | Vertical scroll amount, in pixels. Normally, this is 0. | |
1291 | ||
1292 | @item dedicated | |
1293 | Non-@code{nil} if this window is dedicated to its buffer. | |
1294 | ||
1295 | @item display_table | |
1296 | The window's display table, or @code{nil} if none is specified for it. | |
1297 | ||
1298 | @item update_mode_line | |
1299 | Non-@code{nil} means this window's mode line needs to be updated. | |
1300 | ||
1301 | @item base_line_number | |
1302 | The line number of a certain position in the buffer, or @code{nil}. | |
1303 | This is used for displaying the line number of point in the mode line. | |
1304 | ||
1305 | @item base_line_pos | |
1306 | The position in the buffer for which the line number is known, or | |
7d2d7cc0 GM |
1307 | @code{nil} meaning none is known. If it is a buffer, don't display |
1308 | the line number as long as the window shows that buffer. | |
b8d4c8d0 GM |
1309 | |
1310 | @item region_showing | |
1311 | If the region (or part of it) is highlighted in this window, this field | |
1312 | holds the mark position that made one end of that region. Otherwise, | |
1313 | this field is @code{nil}. | |
1314 | ||
1315 | @item column_number_displayed | |
1316 | The column number currently displayed in this window's mode line, or @code{nil} | |
1317 | if column numbers are not being displayed. | |
1318 | ||
1319 | @item current_matrix | |
7d2d7cc0 GM |
1320 | @itemx desired_matrix |
1321 | Glyph matrices describing the current and desired display of this window. | |
b8d4c8d0 GM |
1322 | @end table |
1323 | ||
1324 | @node Process Internals | |
520f2425 | 1325 | @subsection Process Internals |
b8d4c8d0 GM |
1326 | @cindex internals, of process |
1327 | @cindex process internals | |
1328 | ||
7d2d7cc0 GM |
1329 | The fields of a process (for a complete list, see the definition of |
1330 | @code{struct Lisp_Process} in @file{process.h}) include: | |
b8d4c8d0 GM |
1331 | |
1332 | @table @code | |
1333 | @item name | |
1334 | A string, the name of the process. | |
1335 | ||
1336 | @item command | |
1337 | A list containing the command arguments that were used to start this | |
c73e02fa GM |
1338 | process. For a network or serial process, it is @code{nil} if the |
1339 | process is running or @code{t} if the process is stopped. | |
b8d4c8d0 GM |
1340 | |
1341 | @item filter | |
7d2d7cc0 GM |
1342 | If non-@code{nil}, a function used to accept output from the process |
1343 | instead of a buffer. | |
b8d4c8d0 GM |
1344 | |
1345 | @item sentinel | |
7d2d7cc0 GM |
1346 | If non-@code{nil}, a function called whenever the state of the process |
1347 | changes. | |
b8d4c8d0 GM |
1348 | |
1349 | @item buffer | |
1350 | The associated buffer of the process. | |
1351 | ||
1352 | @item pid | |
1353 | An integer, the operating system's process @acronym{ID}. | |
7d2d7cc0 | 1354 | Pseudo-processes such as network or serial connections use a value of 0. |
b8d4c8d0 GM |
1355 | |
1356 | @item childp | |
7d2d7cc0 GM |
1357 | A flag, @code{t} if this is really a child process. For a network or |
1358 | serial connection, it is a plist based on the arguments to | |
1359 | @code{make-network-process} or @code{make-serial-process}. | |
b8d4c8d0 GM |
1360 | |
1361 | @item mark | |
1362 | A marker indicating the position of the end of the last output from this | |
1363 | process inserted into the buffer. This is often but not always the end | |
1364 | of the buffer. | |
1365 | ||
1366 | @item kill_without_query | |
ee2d5b5e CY |
1367 | If this is non-zero, killing Emacs while this process is still running |
1368 | does not ask for confirmation about killing the process. | |
b8d4c8d0 | 1369 | |
7d2d7cc0 GM |
1370 | @item raw_status |
1371 | The raw process status, as returned by the @code{wait} system call. | |
b8d4c8d0 GM |
1372 | |
1373 | @item status | |
1374 | The process status, as @code{process-status} should return it. | |
1375 | ||
1376 | @item tick | |
1377 | @itemx update_tick | |
1378 | If these two fields are not equal, a change in the status of the process | |
1379 | needs to be reported, either by running the sentinel or by inserting a | |
1380 | message in the process buffer. | |
1381 | ||
1382 | @item pty_flag | |
1383 | Non-@code{nil} if communication with the subprocess uses a @acronym{PTY}; | |
1384 | @code{nil} if it uses a pipe. | |
1385 | ||
1386 | @item infd | |
1387 | The file descriptor for input from the process. | |
1388 | ||
1389 | @item outfd | |
1390 | The file descriptor for output to the process. | |
1391 | ||
b8d4c8d0 GM |
1392 | @item tty_name |
1393 | The name of the terminal that the subprocess is using, | |
1394 | or @code{nil} if it is using pipes. | |
1395 | ||
1396 | @item decode_coding_system | |
1397 | Coding-system for decoding the input from this process. | |
1398 | ||
1399 | @item decoding_buf | |
1400 | A working buffer for decoding. | |
1401 | ||
1402 | @item decoding_carryover | |
1403 | Size of carryover in decoding. | |
1404 | ||
1405 | @item encode_coding_system | |
1406 | Coding-system for encoding the output to this process. | |
1407 | ||
1408 | @item encoding_buf | |
1409 | A working buffer for encoding. | |
1410 | ||
b8d4c8d0 GM |
1411 | @item inherit_coding_system_flag |
1412 | Flag to set @code{coding-system} of the process buffer from the | |
1413 | coding system used to decode process output. | |
c73e02fa GM |
1414 | |
1415 | @item type | |
1416 | Symbol indicating the type of process: @code{real}, @code{network}, | |
7d2d7cc0 | 1417 | @code{serial}. |
c73e02fa | 1418 | |
b8d4c8d0 | 1419 | @end table |
7d2d7cc0 GM |
1420 | |
1421 | @c FIXME Mention src/globals.h somewhere in this file? |