1 @c This is part of the Emacs manual.
2 @c Copyright (C) 1985-1987, 1993-1995, 1997, 2000-2017 Free Software
4 @c See file emacs.texi for copying conditions.
6 @chapter Miscellaneous Commands
8 This chapter contains several brief topics that do not fit anywhere
9 else: reading Usenet news, host and network security,
10 viewing PDFs and other such documents, web
11 browsing, running shell commands and shell subprocesses, using a
12 single shared Emacs for utilities that expect to run an editor as a
13 subprocess, printing, sorting text, editing binary files, saving an
14 Emacs session for later resumption, recursive editing level, following
15 hyperlinks, and various diversions and amusements.
29 Gnus is an Emacs package primarily designed for reading and posting
30 Usenet news. It can also be used to read and respond to messages from
31 a number of other sources---email, remote directories, digests, and so
32 on. Here we introduce Gnus and describe several basic features.
34 For full details, see @ref{Top, Gnus,, gnus, The Gnus Manual}.
37 For full details on Gnus, type @kbd{C-h i} and then select the Gnus
42 * Buffers of Gnus:: The group, summary, and article buffers.
43 * Gnus Startup:: What you should know about starting Gnus.
44 * Gnus Group Buffer:: A short description of Gnus group commands.
45 * Gnus Summary Buffer:: A short description of Gnus summary commands.
49 @subsection Gnus Buffers
51 Gnus uses several buffers to display information and to receive
52 commands. The three most commonly-used Gnus buffers are the
53 @dfn{group buffer}, the @dfn{summary buffer} and the @dfn{article
56 The @dfn{group buffer} contains a list of article sources (e.g.,
57 newsgroups and email inboxes), which are collectively referred to as
58 @dfn{groups}. This is the first buffer Gnus displays when it starts
59 up. It normally displays only the groups to which you subscribe and
60 that contain unread articles. From this buffer, you can select a
63 The @dfn{summary buffer} lists the articles in a single group,
64 showing one article per line. By default, it displays each article's
65 author, subject, and line
70 number, but this is customizable; @xref{Summary Buffer Format,,, gnus,
73 The summary buffer is created when you select a group in the group
74 buffer, and is killed when you exit the group.
76 From the summary buffer, you can choose an article to view. The
77 article is displayed in the @dfn{article buffer}. In normal Gnus
78 usage, you view this buffer but do not select it---all useful Gnus
79 commands can be invoked from the summary buffer. But you can select
80 the article buffer, and execute Gnus commands from it, if you wish.
83 @subsection When Gnus Starts Up
86 @cindex @file{.newsrc} file
87 If your system has been set up for reading Usenet news, getting
88 started with Gnus is easy---just type @kbd{M-x gnus}.
90 On starting up, Gnus reads your @dfn{news initialization file}: a
91 file named @file{.newsrc} in your home directory which lists your
92 Usenet newsgroups and subscriptions (this file is not unique to Gnus;
93 it is used by many other newsreader programs). It then tries to
94 contact the system's default news server, which is typically specified
95 by the @env{NNTPSERVER} environment variable.
97 If your system does not have a default news server, or if you wish
98 to use Gnus for reading email, then before invoking @kbd{M-x gnus} you
99 need to tell Gnus where to get news and/or mail. To do this,
100 customize the variables @code{gnus-select-method} and/or
101 @code{gnus-secondary-select-methods}.
103 See the Gnus manual for details.
106 @xref{Finding the News,,, gnus, The Gnus Manual}.
109 Once Gnus has started up, it displays the group buffer. By default,
110 the group buffer shows only a small number of @dfn{subscribed groups}.
111 Groups with other statuses---@dfn{unsubscribed}, @dfn{killed}, or
112 @dfn{zombie}---are hidden. The first time you start Gnus, any group
113 to which you are not subscribed is made into a killed group; any group
114 that subsequently appears on the news server becomes a zombie group.
116 To proceed, you must select a group in the group buffer to open the
117 summary buffer for that group; then, select an article in the summary
118 buffer to view its article buffer in a separate window. The following
119 sections explain how to use the group and summary buffers to do this.
121 To quit Gnus, type @kbd{q} in the group buffer. This automatically
122 records your group statuses in the files @file{.newsrc} and
123 @file{.newsrc.eld}, so that they take effect in subsequent Gnus
126 @node Gnus Group Buffer
127 @subsection Using the Gnus Group Buffer
129 The following commands are available in the Gnus group buffer:
132 @kindex SPC @r{(Gnus Group mode)}
133 @findex gnus-group-read-group
135 Switch to the summary buffer for the group on the current line.
137 @kindex l @r{(Gnus Group mode)}
138 @kindex A s @r{(Gnus Group mode)}
139 @findex gnus-group-list-groups
142 In the group buffer, list only the groups to which you subscribe and
143 which contain unread articles (this is the default listing).
145 @kindex L @r{(Gnus Group mode)}
146 @kindex A u @r{(Gnus Group mode)}
147 @findex gnus-group-list-all-groups
150 List all subscribed and unsubscribed groups, but not killed or zombie
153 @kindex A k @r{(Gnus Group mode)}
154 @findex gnus-group-list-all-groups
158 @kindex A z @r{(Gnus Group mode)}
159 @findex gnus-group-list-all-groups
163 @kindex u @r{(Gnus Group mode)}
164 @findex gnus-group-unsubscribe-current-group
165 @cindex subscribe groups
166 @cindex unsubscribe groups
168 Toggle the subscription status of the group on the current line
169 (i.e., turn a subscribed group into an unsubscribed group, or vice
170 versa). Invoking this on a killed or zombie group turns it into an
173 @kindex C-k @r{(Gnus Group mode)}
174 @findex gnus-group-kill-group
176 Kill the group on the current line. Killed groups are not recorded in
177 the @file{.newsrc} file, and they are not shown in the @kbd{l} or
180 @kindex DEL @r{(Gnus Group mode)}
182 Move point to the previous group containing unread articles.
184 @kindex n @r{(Gnus Group mode)}
185 @findex gnus-group-next-unread-group
186 @findex gnus-summary-next-unread-article
188 Move point to the next unread group.
190 @kindex p @r{(Gnus Group mode)}
191 @findex gnus-group-prev-unread-group
192 @findex gnus-summary-prev-unread-article
194 Move point to the previous unread group.
196 @kindex q @r{(Gnus Group mode)}
197 @findex gnus-group-exit
199 Update your Gnus settings, and quit Gnus.
202 @node Gnus Summary Buffer
203 @subsection Using the Gnus Summary Buffer
205 The following commands are available in the Gnus summary buffer:
208 @kindex SPC @r{(Gnus Summary mode)}
209 @findex gnus-group-read-group
211 If there is no article selected, select the article on the current
212 line and display its article buffer. Otherwise, try scrolling the
213 selected article buffer in its window; on reaching the end of the
214 buffer, select the next unread article.
216 Thus, you can read through all articles by repeatedly typing
219 @kindex DEL @r{(Gnus Summary mode)}
220 @findex gnus-summary-prev-page
222 Scroll the text of the article backwards.
224 @kindex n @r{(Gnus Summary mode)}
225 @findex gnus-group-next-unread-group
226 @findex gnus-summary-next-unread-article
228 Select the next unread article.
230 @kindex p @r{(Gnus Summary mode)}
231 @findex gnus-group-prev-unread-group
232 @findex gnus-summary-prev-unread-article
234 Select the previous unread article.
236 @kindex s @r{(Gnus Summary mode)}
237 @findex gnus-summary-isearch-article
239 Do an incremental search on the selected article buffer, as if you
240 switched to the buffer and typed @kbd{C-s} (@pxref{Incremental
243 @kindex M-s @r{(Gnus Summary mode)}
244 @findex gnus-summary-search-article-forward
245 @item M-s @var{regexp} @key{RET}
246 Search forward for articles containing a match for @var{regexp}.
248 @kindex q @r{(Gnus Summary mode)}
250 Exit the summary buffer and return to the group buffer.
254 @section Host Security
257 Emacs runs inside an operating system such as GNU/Linux, and relies on
258 the operating system to check security constraints such as accesses to
259 files. The default settings for Emacs are designed for typical use;
260 they may require some tailoring in environments where security is more
261 of a concern, or less of a concern, than usual. For example,
262 file-local variables can be risky, and you can set the variable
263 @code{enable-local-variables} to @code{:safe} or (even more
264 conservatively) to @code{nil}; conversely, if your files can all be
265 trusted and the default checking for these variables is irritating,
266 you can set @code{enable-local-variables} to @code{:all}. @xref{Safe
269 @xref{Security Considerations,,, elisp, The Emacs Lisp Reference
270 Manual}, for more information about security considerations when using
271 Emacs as part of a larger application.
273 @node Network Security
274 @section Network Security
275 @cindex network security manager
282 Whenever Emacs establishes any network connection, it passes the
283 established connection to the @dfn{Network Security Manager}
284 (@acronym{NSM}). @acronym{NSM} is responsible for enforcing the
285 network security under your control.
287 @vindex network-security-level
288 The @code{network-security-level} variable determines the security
289 level that @acronym{NSM} enforces. If its value is @code{low}, no
290 security checks are performed.
292 If this variable is @code{medium} (which is the default), a number of
293 checks will be performed. If as result @acronym{NSM} determines that
294 the network connection might not be trustworthy, it will make you
295 aware of that, and will ask you what to do about the network
298 You can decide to register a permanent security exception for an
299 unverified connection, a temporary exception, or refuse the connection
302 Below is a list of the checks done on the @code{medium} level.
306 @item unable to verify a @acronym{TLS} certificate
307 If the connection is a @acronym{TLS}, @acronym{SSL} or
308 @acronym{STARTTLS} connection, @acronym{NSM} will check whether
309 the certificate used to establish the identity of the server we're
310 connecting to can be verified.
312 While an invalid certificate is often the cause for concern (there
313 could be a Man-in-the-Middle hijacking your network connection and
314 stealing your password), there may be valid reasons for going ahead
315 with the connection anyway. For instance, the server may be using a
316 self-signed certificate, or the certificate may have expired. It's up
317 to you to determine whether it's acceptable to continue with the
320 @item a self-signed certificate has changed
321 If you've previously accepted a self-signed certificate, but it has
322 now changed, that could mean that the server has just changed the
323 certificate, but it might also mean that the network connection has
326 @item previously encrypted connection now unencrypted
327 If the connection is unencrypted, but it was encrypted in previous
328 sessions, this might mean that there is a proxy between you and the
329 server that strips away @acronym{STARTTLS} announcements, leaving the
330 connection unencrypted. This is usually very suspicious.
332 @item talking to an unencrypted service when sending a password
333 When connecting to an @acronym{IMAP} or @acronym{POP3} server, these
334 should usually be encrypted, because it's common to send passwords
335 over these connections. Similarly, if you're sending email via
336 @acronym{SMTP} that requires a password, you usually want that
337 connection to be encrypted. If the connection isn't encrypted,
338 @acronym{NSM} will warn you.
342 If @code{network-security-level} is @code{high}, the following checks
343 will be made, in addition to the above:
346 @item a validated certificate changes the public key
347 Servers change their keys occasionally, and that is normally nothing
348 to be concerned about. However, if you are worried that your network
349 connections are being hijacked by agencies who have access to pliable
350 Certificate Authorities which issue new certificates for third-party
351 services, you may want to keep track of these changes.
353 @item Diffie-Hellman low prime bits
354 When doing the public key exchange, the number of prime bits
355 should be high to ensure that the channel can't be eavesdropped on by
356 third parties. If this number is too low, you will be warned.
358 @item @acronym{RC4} stream cipher
359 The @acronym{RC4} stream cipher is believed to be of low quality and
360 may allow eavesdropping by third parties.
362 @item @acronym{SSL1}, @acronym{SSL2} and @acronym{SSL3}
363 The protocols older than @acronym{TLS1.0} are believed to be
364 vulnerable to a variety of attacks, and you may want to avoid using
365 these if what you're doing requires higher security.
368 Finally, if @code{network-security-level} is @code{paranoid}, you will
369 also be notified the first time @acronym{NSM} sees any new
370 certificate. This will allow you to inspect all the certificates from
371 all the connections that Emacs makes.
373 The following additional variables can be used to control details of
374 @acronym{NSM} operation:
377 @item nsm-settings-file
378 @vindex nsm-settings-file
379 This is the file where @acronym{NSM} stores details about connections.
380 It defaults to @file{~/.emacs.d/network-security.data}.
382 @item nsm-save-host-names
383 @vindex nsm-save-host-names
384 By default, host names will not be saved for non-@code{STARTTLS}
385 connections. Instead a host/port hash is used to identify connections.
386 This means that one can't casually read the settings file to see what
387 servers the user has connected to. If this variable is @code{t},
388 @acronym{NSM} will also save host names in the nsm-settings-file.
393 @section Document Viewing
397 @cindex PostScript file
398 @cindex OpenDocument file
399 @cindex Microsoft Office file
401 @cindex mode, DocView
402 @cindex document viewer (DocView)
403 @findex doc-view-mode
405 DocView mode is a major mode for viewing DVI, PostScript (PS), PDF,
406 OpenDocument, and Microsoft Office documents. It provides features
407 such as slicing, zooming, and searching inside documents. It works by
408 converting the document to a set of images using the @command{gs}
409 (GhostScript) or @command{mudraw}/@command{pdfdraw} (MuPDF) commands
410 and other external tools @footnote{For PostScript files, GhostScript
411 is a hard requirement. For DVI files, @code{dvipdf} or @code{dvipdfm}
412 is needed. For OpenDocument and Microsoft Office documents, the
413 @code{unoconv} tool is needed.}, and displaying those images.
415 @findex doc-view-toggle-display
416 @findex doc-view-toggle-display
417 @cindex doc-view-minor-mode
418 When you visit a document file that can be displayed with DocView
419 mode, Emacs automatically uses DocView mode @footnote{The needed
420 external tools for the document type must be available, and Emacs must
421 be running in a graphical frame and have PNG image support. If any of
422 these requirements is not fulfilled, Emacs falls back to another major
423 mode.}. As an exception, when you visit a PostScript file, Emacs
424 switches to PS mode, a major mode for editing PostScript files as
425 text; however, it also enables DocView minor mode, so you can type
426 @kbd{C-c C-c} to view the document with DocView. In either DocView
427 mode or DocView minor mode, repeating @kbd{C-c C-c}
428 (@code{doc-view-toggle-display}) toggles between DocView and the
429 underlying file contents.
431 @findex doc-view-open-text
432 When you visit a file which would normally be handled by DocView
433 mode but some requirement is not met (e.g., you operate in a terminal
434 frame or emacs has no PNG support), you are queried if you want to
435 view the document's contents as plain text. If you confirm, the
436 buffer is put in text mode and DocView minor mode is activated. Thus,
437 by typing @kbd{C-c C-c} you switch to the fallback mode. With another
438 @kbd{C-c C-c} you return to DocView mode. The plain text contents can
439 also be displayed from within DocView mode by typing @kbd{C-c C-t}
440 (@code{doc-view-open-text}).
442 You can explicitly enable DocView mode with the command @code{M-x
443 doc-view-mode}. You can toggle DocView minor mode with @code{M-x
444 doc-view-minor-mode}.
446 When DocView mode starts, it displays a welcome screen and begins
447 formatting the file, page by page. It displays the first page once
448 that has been formatted.
450 To kill the DocView buffer, type @kbd{k}
451 (@code{doc-view-kill-proc-and-buffer}). To bury it, type @kbd{q}
452 (@code{quit-window}).
455 * Navigation: DocView Navigation. Navigating DocView buffers.
456 * Searching: DocView Searching. Searching inside documents.
457 * Slicing: DocView Slicing. Specifying which part of a page is displayed.
458 * Conversion: DocView Conversion. Influencing and triggering conversion.
461 @node DocView Navigation
462 @subsection DocView Navigation
464 In DocView mode, you can scroll the current page using the usual
465 Emacs movement keys: @kbd{C-p}, @kbd{C-n}, @kbd{C-b}, @kbd{C-f}, and
468 @vindex doc-view-continuous
469 By default, the line-motion keys @kbd{C-p} and @kbd{C-n} stop
470 scrolling at the beginning and end of the current page, respectively.
471 However, if you change the variable @code{doc-view-continuous} to a
472 non-@code{nil} value, then @kbd{C-p} displays the previous page if you
473 are already at the beginning of the current page, and @kbd{C-n}
474 displays the next page if you are at the end of the current page.
476 @findex doc-view-next-page
477 @findex doc-view-previous-page
478 @kindex n @r{(DocView mode)}
479 @kindex p @r{(DocView mode)}
480 @kindex C-x ] @r{(DocView mode)}
481 @kindex C-x [ @r{(DocView mode)}
482 You can also display the next page by typing @kbd{n}, @key{next} or
483 @kbd{C-x ]} (@code{doc-view-next-page}). To display the previous
484 page, type @kbd{p}, @key{prior} or @kbd{C-x [}
485 (@code{doc-view-previous-page}).
487 @findex doc-view-scroll-up-or-next-page
488 @findex doc-view-scroll-down-or-previous-page
489 @kindex SPC @r{(DocView mode)}
490 @kindex DEL @r{(DocView mode)}
491 @key{SPC} (@code{doc-view-scroll-up-or-next-page}) is a convenient
492 way to advance through the document. It scrolls within the current
493 page or advances to the next. @key{DEL} moves backwards in a similar
494 way (@code{doc-view-scroll-down-or-previous-page}).
496 @findex doc-view-first-page
497 @findex doc-view-last-page
498 @findex doc-view-goto-page
499 @kindex M-< @r{(DocView mode)}
500 @kindex M-> @r{(DocView mode)}
501 To go to the first page, type @kbd{M-<}
502 (@code{doc-view-first-page}); to go to the last one, type @kbd{M->}
503 (@code{doc-view-last-page}). To jump to a page by its number, type
504 @kbd{M-g M-g} or @kbd{M-g g} (@code{doc-view-goto-page}).
506 @findex doc-view-enlarge
507 @findex doc-view-shrink
508 @vindex doc-view-resolution
509 @kindex + @r{(DocView mode)}
510 @kindex - @r{(DocView mode)}
511 You can enlarge or shrink the document with @kbd{+}
512 (@code{doc-view-enlarge}) and @kbd{-} (@code{doc-view-shrink}). These
513 commands work by reconverting the document at the new size. To
514 specify the default size for DocView, customize the variable
515 @code{doc-view-resolution}.
517 @node DocView Searching
518 @subsection DocView Searching
520 In DocView mode, you can search the file's text for a regular
521 expression (@pxref{Regexps}). The interface for searching is inspired
522 by @code{isearch} (@pxref{Incremental Search}).
524 @findex doc-view-search
525 @findex doc-view-search-backward
526 @findex doc-view-show-tooltip
527 To begin a search, type @kbd{C-s} (@code{doc-view-search}) or
528 @kbd{C-r} (@code{doc-view-search-backward}). This reads a regular
529 expression using a minibuffer, then echoes the number of matches found
530 within the document. You can move forward and back among the matches
531 by typing @kbd{C-s} and @kbd{C-r}. DocView mode has no way to show
532 the match inside the page image; instead, it displays a tooltip (at
533 the mouse position) listing all matching lines in the current page.
534 To force display of this tooltip, type @kbd{C-t}
535 (@code{doc-view-show-tooltip}).
537 To start a new search, use the search command with a prefix
538 argument; i.e., @kbd{C-u C-s} for a forward search or @kbd{C-u C-r}
539 for a backward search.
541 @node DocView Slicing
542 @subsection DocView Slicing
544 Documents often have wide margins for printing. They are annoying
545 when reading the document on the screen, because they use up screen
546 space and can cause inconvenient scrolling.
548 @findex doc-view-set-slice
549 @findex doc-view-set-slice-using-mouse
550 With DocView you can hide these margins by selecting a @dfn{slice}
551 of pages to display. A slice is a rectangle within the page area;
552 once you specify a slice in DocView, it applies to whichever page you
555 To specify the slice numerically, type @kbd{s s}
556 (@code{doc-view-set-slice}); then enter the top left pixel position
557 and the slice's width and height.
558 @c ??? how does this work?
560 A more convenient graphical way to specify the slice is with @kbd{s
561 m} (@code{doc-view-set-slice-using-mouse}), where you use the mouse to
562 select the slice. Simply press and hold the left mouse button at the
563 upper-left corner of the region you want to have in the slice, then
564 move the mouse pointer to the lower-right corner and release the
567 The most convenient way is to set the optimal slice by using
568 BoundingBox information automatically determined from the document by
569 typing @kbd{s b} (@code{doc-view-set-slice-from-bounding-box}).
571 @findex doc-view-reset-slice
572 To cancel the selected slice, type @kbd{s r}
573 (@code{doc-view-reset-slice}). Then DocView shows the entire page
574 including its entire margins.
576 @node DocView Conversion
577 @subsection DocView Conversion
579 @vindex doc-view-cache-directory
580 @findex doc-view-clear-cache
581 For efficiency, DocView caches the images produced by @command{gs}.
582 The name of this directory is given by the variable
583 @code{doc-view-cache-directory}. You can clear the cache directory by
584 typing @code{M-x doc-view-clear-cache}.
586 @findex doc-view-kill-proc
587 @findex doc-view-kill-proc-and-buffer
588 To force reconversion of the currently viewed document, type @kbd{r}
589 or @kbd{g} (@code{revert-buffer}). To kill the converter process
590 associated with the current buffer, type @kbd{K}
591 (@code{doc-view-kill-proc}). The command @kbd{k}
592 (@code{doc-view-kill-proc-and-buffer}) kills the converter process and
596 @section Web Browsing with EWW
599 @findex eww-open-file
600 @dfn{EWW}, the Emacs Web Wowser, is a web browser package for Emacs.
601 It allows browsing URLs within an Emacs buffer. The command @kbd{M-x
602 eww} will open a URL or search the web. You can open a file
603 using the command @kbd{M-x eww-open-file}. You can use EWW as the
604 web browser for @code{browse-url}, @pxref{Browse-URL}. For full
605 details, @pxref{Top, EWW,, eww, The Emacs Web Wowser Manual}.
607 @node Embedded WebKit Widgets
608 @section Embedded WebKit Widgets
610 @cindex webkit widgets
611 @cindex embedded widgets
613 @findex xwidget-webkit-browse-url
614 @findex xwidget-webkit-mode
615 @cindex Xwidget-WebKit mode
616 If Emacs was compiled with the appropriate support packages, it is
617 able to show browser widgets in its buffers. The command @kbd{M-x
618 xwidget-webkit-browse-url} asks for a URL to display in the browser
619 widget. The URL normally defaults to the URL at or before point, but
620 if there is an active region (@pxref{Mark}), the default URL comes
621 from the region instead, after removing any whitespace from it. The
622 command then creates a new buffer with the embedded browser showing
623 the specified URL. The buffer is put in the Xwidget-WebKit mode
624 (similar to Image mode, @pxref{File Conveniences}), which provides
625 one-key commands for scrolling the widget, changing its size, and
626 reloading it. Type @w{@kbd{C-h b}} in that buffer to see the key
630 @section Running Shell Commands from Emacs
632 @cindex shell commands
634 Emacs has commands for passing single command lines to shell
635 subprocesses, and for running a shell interactively with input and
636 output to an Emacs buffer, and for running a shell in a terminal
640 @item M-! @var{cmd} @key{RET}
641 Run the shell command @var{cmd} and display the output
642 (@code{shell-command}).
643 @item M-| @var{cmd} @key{RET}
644 Run the shell command @var{cmd} with region contents as input;
645 optionally replace the region with the output
646 (@code{shell-command-on-region}).
647 @item M-& @var{cmd} @key{RET}
648 Run the shell command @var{cmd} asynchronously, and display the output
649 (@code{async-shell-command}).
651 Run a subshell with input and output through an Emacs buffer. You can
652 then give commands interactively.
654 Run a subshell with input and output through an Emacs buffer. You can
655 then give commands interactively. Full terminal emulation is
660 Whenever you specify a relative file name for an executable program
661 (either in the @var{cmd} argument to one of the above commands, or in
662 other contexts), Emacs searches for the program in the directories
663 specified by the variable @code{exec-path}. The value of this
664 variable must be a list of directories; the default value is
665 initialized from the environment variable @env{PATH} when Emacs is
666 started (@pxref{General Variables}).
668 @kbd{M-x eshell} invokes a shell implemented entirely in Emacs. It
669 is documented in its own manual.
671 @xref{Top,Eshell,Eshell, eshell, Eshell: The Emacs Shell}.
674 See the Eshell Info manual, which is distributed with Emacs.
678 * Single Shell:: How to run one shell command and return.
679 * Interactive Shell:: Permanent shell taking input via Emacs.
680 * Shell Mode:: Special Emacs commands used with permanent shell.
681 * Shell Prompts:: Two ways to recognize shell prompts.
682 * History: Shell History. Repeating previous commands in a shell buffer.
683 * Directory Tracking:: Keeping track when the subshell changes directory.
684 * Options: Shell Options. Options for customizing Shell mode.
685 * Terminal emulator:: An Emacs window as a terminal emulator.
686 * Term Mode:: Special Emacs commands used in Term mode.
687 * Remote Host:: Connecting to another computer.
688 * Serial Terminal:: Connecting to a serial port.
692 @subsection Single Shell Commands
695 @findex shell-command
696 @kbd{M-!} (@code{shell-command}) reads a line of text using the
697 minibuffer and executes it as a shell command, in a subshell made just
698 for that command. Standard input for the command comes from the null
699 device. If the shell command produces any output, the output appears
700 either in the echo area (if it is short), or in an Emacs buffer named
701 @file{*Shell Command Output*}, displayed in another window (if the
704 For instance, one way to decompress a file named @file{foo.gz} is to
705 type @kbd{M-! gunzip foo.gz @key{RET}}. That shell command normally
706 creates the file @file{foo} and produces no terminal output.
708 A numeric argument to @code{shell-command}, e.g., @kbd{M-1 M-!},
709 causes it to insert terminal output into the current buffer instead of
710 a separate buffer. It puts point before the output, and sets the mark
711 after the output. For instance, @kbd{M-1 M-! gunzip < foo.gz
712 @key{RET}} would insert the uncompressed form of the file
713 @file{foo.gz} into the current buffer.
715 Provided the specified shell command does not end with @samp{&}, it
716 runs @dfn{synchronously}, and you must wait for it to exit before
717 continuing to use Emacs. To stop waiting, type @kbd{C-g} to quit;
718 this sends a @code{SIGINT} signal to terminate the shell command (this
719 is the same signal that @kbd{C-c} normally generates in the shell).
720 Emacs then waits until the command actually terminates. If the shell
721 command doesn't stop (because it ignores the @code{SIGINT} signal),
722 type @kbd{C-g} again; this sends the command a @code{SIGKILL} signal,
723 which is impossible to ignore.
726 @findex async-shell-command
727 A shell command that ends in @samp{&} is executed
728 @dfn{asynchronously}, and you can continue to use Emacs as it runs.
729 You can also type @kbd{M-&} (@code{async-shell-command}) to execute a
730 shell command asynchronously; this is exactly like calling @kbd{M-!}
731 with a trailing @samp{&}, except that you do not need the @samp{&}.
732 The default output buffer for asynchronous shell commands is named
733 @samp{*Async Shell Command*}. Emacs inserts the output into this
734 buffer as it comes in, whether or not the buffer is visible in a
737 @vindex async-shell-command-buffer
738 If you want to run more than one asynchronous shell command at the
739 same time, they could end up competing for the output buffer. The
740 option @code{async-shell-command-buffer} specifies what to do about
741 this; e.g., whether to rename the pre-existing output buffer, or to
742 use a different buffer for the new command. Consult the variable's
743 documentation for more possibilities.
745 @vindex async-shell-command-display-buffer
746 If you want the output buffer for asynchronous shell commands to be
747 displayed only when the command generates output, set
748 @code{async-shell-command-display-buffer} to @code{nil}.
751 @findex shell-command-on-region
752 @kbd{M-|} (@code{shell-command-on-region}) is like @kbd{M-!}, but
753 passes the contents of the region as the standard input to the shell
754 command, instead of no input. With a numeric argument, it deletes the
755 old region and replaces it with the output from the shell command.
757 For example, you can use @kbd{M-|} with the @command{gpg} program to
758 see what keys are in the buffer. If the buffer contains a GnuPG key,
759 type @kbd{C-x h M-| gpg @key{RET}} to feed the entire buffer contents
760 to @command{gpg}. This will output the list of keys to the
761 @file{*Shell Command Output*} buffer.
763 @vindex shell-file-name
764 The above commands use the shell specified by the variable
765 @code{shell-file-name}. Its default value is determined by the
766 @env{SHELL} environment variable when Emacs is started. If the file
767 name is relative, Emacs searches the directories listed in
768 @code{exec-path} (@pxref{Shell}).
770 To specify a coding system for @kbd{M-!} or @kbd{M-|}, use the command
771 @kbd{C-x @key{RET} c} immediately beforehand. @xref{Communication Coding}.
773 @vindex shell-command-default-error-buffer
774 By default, error output is intermixed with the regular output in
775 the output buffer. But if you change the value of the variable
776 @code{shell-command-default-error-buffer} to a string, error output is
777 inserted into a buffer of that name.
779 @vindex shell-command-dont-erase-buffer
780 By default, the output buffer is erased between shell commands.
781 If you change the value of the variable
782 @code{shell-command-dont-erase-buffer} to a non-@code{nil} value,
783 the output buffer is not erased. This variable also controls where to
784 set the point in the output buffer after the command completes; see the
785 documentation of the variable for details.
787 @node Interactive Shell
788 @subsection Interactive Subshell
791 To run a subshell interactively, type @kbd{M-x shell}. This creates
792 (or reuses) a buffer named @file{*shell*}, and runs a shell subprocess
793 with input coming from and output going to that buffer. That is to
794 say, any terminal output from the subshell goes into the buffer,
795 advancing point, and any terminal input for the subshell comes from
796 text in the buffer. To give input to the subshell, go to the end of
797 the buffer and type the input, terminated by @key{RET}.
799 By default, when the subshell is invoked interactively, the
800 @file{*shell*} buffer is displayed in a new window. This behavior can
801 be customized via @code{display-buffer-alist} (@pxref{Window Choice}).
803 While the subshell is waiting or running a command, you can switch
804 windows or buffers and perform other editing in Emacs. Emacs inserts
805 the output from the subshell into the Shell buffer whenever it has
806 time to process it (e.g., while waiting for keyboard input).
808 @cindex @code{comint-highlight-input} face
809 @cindex @code{comint-highlight-prompt} face
810 In the Shell buffer, prompts are displayed with the face
811 @code{comint-highlight-prompt}, and submitted input lines are
812 displayed with the face @code{comint-highlight-input}. This makes it
813 easier to distinguish input lines from the shell output.
816 To make multiple subshells, invoke @kbd{M-x shell} with a prefix
817 argument (e.g., @kbd{C-u M-x shell}). Then the command will read a
818 buffer name, and create (or reuse) a subshell in that buffer. You can
819 also rename the @file{*shell*} buffer using @kbd{M-x rename-uniquely},
820 then create a new @file{*shell*} buffer using plain @kbd{M-x shell}.
821 Subshells in different buffers run independently and in parallel.
823 @vindex explicit-shell-file-name
824 @cindex environment variables for subshells
825 @cindex @env{ESHELL} environment variable
826 @cindex @env{SHELL} environment variable
827 To specify the shell file name used by @kbd{M-x shell}, customize
828 the variable @code{explicit-shell-file-name}. If this is @code{nil}
829 (the default), Emacs uses the environment variable @env{ESHELL} if it
830 exists. Otherwise, it usually uses the variable
831 @code{shell-file-name} (@pxref{Single Shell}); but if the default
832 directory is remote (@pxref{Remote Files}), it prompts you for the
835 Emacs sends the new shell the contents of the file
836 @file{~/.emacs_@var{shellname}} as input, if it exists, where
837 @var{shellname} is the name of the file that the shell was loaded
838 from. For example, if you use bash, the file sent to it is
839 @file{~/.emacs_bash}. If this file is not found, Emacs tries with
840 @file{~/.emacs.d/init_@var{shellname}.sh}.
842 To specify a coding system for the shell, you can use the command
843 @kbd{C-x @key{RET} c} immediately before @kbd{M-x shell}. You can
844 also change the coding system for a running subshell by typing
845 @kbd{C-x @key{RET} p} in the shell buffer. @xref{Communication
848 @cindex @env{INSIDE_EMACS} environment variable
849 Emacs sets the environment variable @env{INSIDE_EMACS} in the
850 subshell to @samp{@var{version},comint}, where @var{version} is the
851 Emacs version (e.g., @samp{24.1}). Programs can check this variable
852 to determine whether they are running inside an Emacs subshell.
855 @subsection Shell Mode
859 The major mode for Shell buffers is Shell mode. Many of its special
860 commands are bound to the @kbd{C-c} prefix, and resemble the usual
861 editing and job control characters present in ordinary shells, except
862 that you must type @kbd{C-c} first. Here is a list of Shell mode
867 @kindex RET @r{(Shell mode)}
868 @findex comint-send-input
869 Send the current line as input to the subshell
870 (@code{comint-send-input}). Any shell prompt at the beginning of the
871 line is omitted (@pxref{Shell Prompts}). If point is at the end of
872 buffer, this is like submitting the command line in an ordinary
873 interactive shell. However, you can also invoke @key{RET} elsewhere
874 in the shell buffer to submit the current line as input.
877 @kindex TAB @r{(Shell mode)}
878 @findex completion-at-point
879 @cindex shell completion
880 Complete the command name or file name before point in the shell
881 buffer (@code{completion-at-point}). This uses the usual Emacs
882 completion rules (@pxref{Completion}), with the completion
883 alternatives being file names, environment variable names, the shell
884 command history, and history references (@pxref{History References}).
885 For options controlling the completion, @pxref{Shell Options}.
888 @kindex M-? @r{(Shell mode)}
889 @findex comint-dynamic-list-filename@dots{}
890 Display temporarily a list of the possible completions of the file
891 name before point (@code{comint-dynamic-list-filename-completions}).
894 @kindex C-d @r{(Shell mode)}
895 @findex comint-delchar-or-maybe-eof
896 Either delete a character or send @acronym{EOF}
897 (@code{comint-delchar-or-maybe-eof}). Typed at the end of the shell
898 buffer, this sends @acronym{EOF} to the subshell. Typed at any other
899 position in the buffer, this deletes a character as usual.
902 @kindex C-c C-a @r{(Shell mode)}
903 @findex comint-bol-or-process-mark
904 Move to the beginning of the line, but after the prompt if any
905 (@code{comint-bol-or-process-mark}). If you repeat this command twice
906 in a row, the second time it moves back to the process mark, which is
907 the beginning of the input that you have not yet sent to the subshell.
908 (Normally that is the same place---the end of the prompt on this
909 line---but after @kbd{C-c @key{SPC}} the process mark may be in a
913 Accumulate multiple lines of input, then send them together. This
914 command inserts a newline before point, but does not send the preceding
915 text as input to the subshell---at least, not yet. Both lines, the one
916 before this newline and the one after, will be sent together (along with
917 the newline that separates them), when you type @key{RET}.
920 @kindex C-c C-u @r{(Shell mode)}
921 @findex comint-kill-input
922 Kill all text pending at end of buffer to be sent as input
923 (@code{comint-kill-input}). If point is not at end of buffer,
924 this only kills the part of this text that precedes point.
927 @kindex C-c C-w @r{(Shell mode)}
928 Kill a word before point (@code{backward-kill-word}).
931 @kindex C-c C-c @r{(Shell mode)}
932 @findex comint-interrupt-subjob
933 Interrupt the shell or its current subjob if any
934 (@code{comint-interrupt-subjob}). This command also kills
935 any shell input pending in the shell buffer and not yet sent.
938 @kindex C-c C-z @r{(Shell mode)}
939 @findex comint-stop-subjob
940 Stop the shell or its current subjob if any (@code{comint-stop-subjob}).
941 This command also kills any shell input pending in the shell buffer and
945 @findex comint-quit-subjob
946 @kindex C-c C-\ @r{(Shell mode)}
947 Send quit signal to the shell or its current subjob if any
948 (@code{comint-quit-subjob}). This command also kills any shell input
949 pending in the shell buffer and not yet sent.
952 @kindex C-c C-o @r{(Shell mode)}
953 @findex comint-delete-output
954 Delete the last batch of output from a shell command
955 (@code{comint-delete-output}). This is useful if a shell command spews
956 out lots of output that just gets in the way.
959 @kindex C-c C-s @r{(Shell mode)}
960 @findex comint-write-output
961 Write the last batch of output from a shell command to a file
962 (@code{comint-write-output}). With a prefix argument, the file is
963 appended to instead. Any prompt at the end of the output is not
968 @kindex C-c C-r @r{(Shell mode)}
969 @kindex C-M-l @r{(Shell mode)}
970 @findex comint-show-output
971 Scroll to display the beginning of the last batch of output at the top
972 of the window; also move the cursor there (@code{comint-show-output}).
975 @kindex C-c C-e @r{(Shell mode)}
976 @findex comint-show-maximum-output
977 Scroll to put the end of the buffer at the bottom of the window
978 (@code{comint-show-maximum-output}).
981 @kindex C-c C-f @r{(Shell mode)}
982 @findex shell-forward-command
983 @vindex shell-command-regexp
984 Move forward across one shell command, but not beyond the current line
985 (@code{shell-forward-command}). The variable @code{shell-command-regexp}
986 specifies how to recognize the end of a command.
989 @kindex C-c C-b @r{(Shell mode)}
990 @findex shell-backward-command
991 Move backward across one shell command, but not beyond the current line
992 (@code{shell-backward-command}).
995 Ask the shell for its working directory, and update the Shell buffer's
996 default directory. @xref{Directory Tracking}.
998 @item M-x send-invisible @key{RET} @var{text} @key{RET}
999 @findex send-invisible
1000 Send @var{text} as input to the shell, after reading it without
1001 echoing. This is useful when a shell command runs a program that asks
1004 Please note that Emacs will not echo passwords by default. If you
1005 really want them to be echoed, evaluate (@pxref{Lisp Eval}) the
1006 following Lisp expression:
1009 (remove-hook 'comint-output-filter-functions
1010 'comint-watch-for-password-prompt)
1013 @item M-x comint-continue-subjob
1014 @findex comint-continue-subjob
1015 Continue the shell process. This is useful if you accidentally suspend
1016 the shell process.@footnote{You should not suspend the shell process.
1017 Suspending a subjob of the shell is a completely different matter---that
1018 is normal practice, but you must use the shell to continue the subjob;
1019 this command won't do it.}
1021 @item M-x comint-strip-ctrl-m
1022 @findex comint-strip-ctrl-m
1023 Discard all control-M characters from the current group of shell output.
1024 The most convenient way to use this command is to make it run
1025 automatically when you get output from the subshell. To do that,
1026 evaluate this Lisp expression:
1029 (add-hook 'comint-output-filter-functions
1030 'comint-strip-ctrl-m)
1033 @item M-x comint-truncate-buffer
1034 @findex comint-truncate-buffer
1035 This command truncates the shell buffer to a certain maximum number of
1036 lines, specified by the variable @code{comint-buffer-maximum-size}.
1037 Here's how to do this automatically each time you get output from the
1041 (add-hook 'comint-output-filter-functions
1042 'comint-truncate-buffer)
1047 @cindex mode, Comint
1048 Shell mode is a derivative of Comint mode, a general-purpose mode for
1049 communicating with interactive subprocesses. Most of the features of
1050 Shell mode actually come from Comint mode, as you can see from the
1051 command names listed above. The special features of Shell mode include
1052 the directory tracking feature, and a few user commands.
1054 Other Emacs features that use variants of Comint mode include GUD
1055 (@pxref{Debuggers}) and @kbd{M-x run-lisp} (@pxref{External Lisp}).
1058 You can use @kbd{M-x comint-run} to execute any program of your choice
1059 in a subprocess using unmodified Comint mode---without the
1060 specializations of Shell mode.
1063 @subsection Shell Prompts
1065 @cindex prompt, shell
1066 A prompt is text output by a program to show that it is ready to
1067 accept new user input. Normally, Comint mode (and thus Shell mode)
1068 automatically figures out part of the buffer is a prompt, based on the
1069 output of the subprocess. (Specifically, it assumes that any received
1070 output line which doesn't end with a newline is a prompt.)
1072 Comint mode divides the buffer into two types of @dfn{fields}: input
1073 fields (where user input is typed) and output fields (everywhere
1074 else). Prompts are part of the output fields. Most Emacs motion
1075 commands do not cross field boundaries, unless they move over multiple
1076 lines. For instance, when point is in the input field on a shell
1077 command line, @kbd{C-a} puts point at the beginning of the input
1078 field, after the prompt. Internally, the fields are implemented using
1079 the @code{field} text property (@pxref{Text Properties,,, elisp, the
1080 Emacs Lisp Reference Manual}).
1082 @vindex comint-use-prompt-regexp
1083 @vindex shell-prompt-pattern
1084 If you change the variable @code{comint-use-prompt-regexp} to a
1085 non-@code{nil} value, then Comint mode recognize prompts using a
1086 regular expression (@pxref{Regexps}). In Shell mode, the regular
1087 expression is specified by the variable @code{shell-prompt-pattern}.
1088 The default value of @code{comint-use-prompt-regexp} is @code{nil},
1089 because this method for recognizing prompts is unreliable, but you may
1090 want to set it to a non-@code{nil} value in unusual circumstances. In
1091 that case, Emacs does not divide the Comint buffer into fields, so the
1092 general motion commands behave as they normally do in buffers without
1093 special text properties. However, you can use the paragraph motion
1094 commands to conveniently navigate the buffer (@pxref{Paragraphs}); in
1095 Shell mode, Emacs uses @code{shell-prompt-pattern} as paragraph
1099 @subsection Shell Command History
1101 Shell buffers support three ways of repeating earlier commands. You
1102 can use keys like those used for the minibuffer history; these work
1103 much as they do in the minibuffer, inserting text from prior commands
1104 while point remains always at the end of the buffer. You can move
1105 through the buffer to previous inputs in their original place, then
1106 resubmit them or copy them to the end. Or you can use a
1107 @samp{!}-style history reference.
1110 * Ring: Shell Ring. Fetching commands from the history list.
1111 * Copy: Shell History Copying. Moving to a command and then copying it.
1112 * History References:: Expanding @samp{!}-style history references.
1116 @subsubsection Shell History Ring
1119 @findex comint-previous-input
1120 @kindex M-p @r{(Shell mode)}
1123 Fetch the next earlier old shell command.
1125 @kindex M-n @r{(Shell mode)}
1126 @findex comint-next-input
1129 Fetch the next later old shell command.
1131 @kindex M-r @r{(Shell mode)}
1132 @findex comint-history-isearch-backward-regexp
1134 Begin an incremental regexp search of old shell commands.
1137 @kindex C-c C-x @r{(Shell mode)}
1138 @findex comint-get-next-from-history
1139 Fetch the next subsequent command from the history.
1142 @kindex C-c . @r{(Shell mode)}
1143 @findex comint-input-previous-argument
1144 Fetch one argument from an old shell command.
1147 @kindex C-c C-l @r{(Shell mode)}
1148 @findex comint-dynamic-list-input-ring
1149 Display the buffer's history of shell commands in another window
1150 (@code{comint-dynamic-list-input-ring}).
1153 Shell buffers provide a history of previously entered shell
1154 commands. To reuse shell commands from the history, use the editing
1155 commands @kbd{M-p}, @kbd{M-n}, @kbd{M-r} and @kbd{M-s}. These work
1156 just like the minibuffer history commands (@pxref{Minibuffer
1157 History}), except that they operate within the Shell buffer rather
1158 than the minibuffer.
1160 @kbd{M-p} fetches an earlier shell command to the end of the shell
1161 buffer. Successive use of @kbd{M-p} fetches successively earlier
1162 shell commands, each replacing any text that was already present as
1163 potential shell input. @kbd{M-n} does likewise except that it finds
1164 successively more recent shell commands from the buffer.
1165 @kbd{C-@key{UP}} works like @kbd{M-p}, and @kbd{C-@key{DOWN}} like
1168 The history search command @kbd{M-r} begins an incremental regular
1169 expression search of previous shell commands. After typing @kbd{M-r},
1170 start typing the desired string or regular expression; the last
1171 matching shell command will be displayed in the current line.
1172 Incremental search commands have their usual effects---for instance,
1173 @kbd{C-s} and @kbd{C-r} search forward and backward for the next match
1174 (@pxref{Incremental Search}). When you find the desired input, type
1175 @key{RET} to terminate the search. This puts the input in the command
1176 line. Any partial input you were composing before navigating the
1177 history list is restored when you go to the beginning or end of the
1180 Often it is useful to reexecute several successive shell commands that
1181 were previously executed in sequence. To do this, first find and
1182 reexecute the first command of the sequence. Then type @kbd{C-c C-x};
1183 that will fetch the following command---the one that follows the command
1184 you just repeated. Then type @key{RET} to reexecute this command. You
1185 can reexecute several successive commands by typing @kbd{C-c C-x
1186 @key{RET}} over and over.
1188 The command @kbd{C-c .}@: (@code{comint-input-previous-argument})
1189 copies an individual argument from a previous command, like
1190 @kbd{@key{ESC} .} in Bash. The simplest use copies the last argument from the
1191 previous shell command. With a prefix argument @var{n}, it copies the
1192 @var{n}th argument instead. Repeating @kbd{C-c .} copies from an
1193 earlier shell command instead, always using the same value of @var{n}
1194 (don't give a prefix argument when you repeat the @kbd{C-c .}
1197 These commands get the text of previous shell commands from a special
1198 history list, not from the shell buffer itself. Thus, editing the shell
1199 buffer, or even killing large parts of it, does not affect the history
1200 that these commands access.
1202 @vindex shell-input-ring-file-name
1203 Some shells store their command histories in files so that you can
1204 refer to commands from previous shell sessions. Emacs reads
1205 the command history file for your chosen shell, to initialize its own
1206 command history. The file name is @file{~/.bash_history} for bash,
1207 @file{~/.sh_history} for ksh, and @file{~/.history} for other shells.
1209 @node Shell History Copying
1210 @subsubsection Shell History Copying
1213 @kindex C-c C-p @r{(Shell mode)}
1214 @findex comint-previous-prompt
1216 Move point to the previous prompt (@code{comint-previous-prompt}).
1218 @kindex C-c C-n @r{(Shell mode)}
1219 @findex comint-next-prompt
1221 Move point to the following prompt (@code{comint-next-prompt}).
1223 @kindex C-c RET @r{(Shell mode)}
1224 @findex comint-copy-old-input
1226 Copy the input command at point, inserting the copy at the end of the
1227 buffer (@code{comint-copy-old-input}). This is useful if you move
1228 point back to a previous command. After you copy the command, you can
1229 submit the copy as input with @key{RET}. If you wish, you can edit
1230 the copy before resubmitting it. If you use this command on an output
1231 line, it copies that line to the end of the buffer.
1234 If @code{comint-use-prompt-regexp} is @code{nil} (the default), copy
1235 the old input command that you click on, inserting the copy at the end
1236 of the buffer (@code{comint-insert-input}). If
1237 @code{comint-use-prompt-regexp} is non-@code{nil}, or if the click is
1238 not over old input, just yank as usual.
1241 Moving to a previous input and then copying it with @kbd{C-c
1242 @key{RET}} or @kbd{mouse-2} produces the same results---the same
1243 buffer contents---that you would get by using @kbd{M-p} enough times
1244 to fetch that previous input from the history list. However, @kbd{C-c
1245 @key{RET}} copies the text from the buffer, which can be different
1246 from what is in the history list if you edit the input text in the
1247 buffer after it has been sent.
1249 @node History References
1250 @subsubsection Shell History References
1251 @cindex history reference
1253 Various shells including csh and bash support @dfn{history
1254 references} that begin with @samp{!} and @samp{^}. Shell mode
1255 recognizes these constructs, and can perform the history substitution
1258 If you insert a history reference and type @key{TAB}, this searches
1259 the input history for a matching command, performs substitution if
1260 necessary, and places the result in the buffer in place of the history
1261 reference. For example, you can fetch the most recent command
1262 beginning with @samp{mv} with @kbd{! m v @key{TAB}}. You can edit the
1263 command if you wish, and then resubmit the command to the shell by
1266 @vindex comint-input-autoexpand
1267 @findex comint-magic-space
1268 Shell mode can optionally expand history references in the buffer
1269 when you send them to the shell. To request this, set the variable
1270 @code{comint-input-autoexpand} to @code{input}. You can make
1271 @key{SPC} perform history expansion by binding @key{SPC} to the
1272 command @code{comint-magic-space}.
1274 Shell mode recognizes history references when they follow a prompt.
1275 @xref{Shell Prompts}, for how Shell mode recognizes prompts.
1277 @node Directory Tracking
1278 @subsection Directory Tracking
1279 @cindex directory tracking
1281 @vindex shell-pushd-regexp
1282 @vindex shell-popd-regexp
1283 @vindex shell-cd-regexp
1284 Shell mode keeps track of @samp{cd}, @samp{pushd} and @samp{popd}
1285 commands given to the subshell, in order to keep the Shell buffer's
1286 default directory (@pxref{File Names}) the same as the shell's working
1287 directory. It recognizes these commands by examining lines of input
1290 If you use aliases for these commands, you can tell Emacs to
1291 recognize them also, by setting the variables
1292 @code{shell-pushd-regexp}, @code{shell-popd-regexp}, and
1293 @code{shell-cd-regexp} to the appropriate regular expressions
1294 (@pxref{Regexps}). For example, if @code{shell-pushd-regexp} matches
1295 the beginning of a shell command line, that line is regarded as a
1296 @code{pushd} command. These commands are recognized only at the
1297 beginning of a shell command line.
1300 If Emacs gets confused about changes in the working directory of the
1301 subshell, type @kbd{M-x dirs}. This command asks the shell for its
1302 working directory and updates the default directory accordingly. It
1303 works for shells that support the most common command syntax, but may
1304 not work for unusual shells.
1306 @findex dirtrack-mode
1307 @cindex Dirtrack mode
1308 @cindex mode, Dirtrack
1309 @vindex dirtrack-list
1310 You can also use Dirtrack mode, a buffer-local minor mode that
1311 implements an alternative method of tracking the shell's working
1312 directory. To use this method, your shell prompt must contain the
1313 working directory at all times, and you must supply a regular
1314 expression for recognizing which part of the prompt contains the
1315 working directory; see the documentation of the variable
1316 @code{dirtrack-list} for details. To use Dirtrack mode, type @kbd{M-x
1317 dirtrack-mode} in the Shell buffer, or add @code{dirtrack-mode} to
1318 @code{shell-mode-hook} (@pxref{Hooks}).
1321 @subsection Shell Mode Options
1323 @vindex comint-scroll-to-bottom-on-input
1324 If the variable @code{comint-scroll-to-bottom-on-input} is
1325 non-@code{nil}, insertion and yank commands scroll the selected window
1326 to the bottom before inserting. The default is @code{nil}.
1328 @vindex comint-scroll-show-maximum-output
1329 If @code{comint-scroll-show-maximum-output} is non-@code{nil}, then
1330 arrival of output when point is at the end tries to scroll the last
1331 line of text to the bottom line of the window, showing as much useful
1332 text as possible. (This mimics the scrolling behavior of most
1333 terminals.) The default is @code{t}.
1335 @vindex comint-move-point-for-output
1336 By setting @code{comint-move-point-for-output}, you can opt for
1337 having point jump to the end of the buffer whenever output arrives---no
1338 matter where in the buffer point was before. If the value is
1339 @code{this}, point jumps in the selected window. If the value is
1340 @code{all}, point jumps in each window that shows the Comint buffer. If
1341 the value is @code{other}, point jumps in all nonselected windows that
1342 show the current buffer. The default value is @code{nil}, which means
1343 point does not jump to the end.
1345 @vindex comint-prompt-read-only
1346 If you set @code{comint-prompt-read-only}, the prompts in the Comint
1347 buffer are read-only.
1349 @vindex comint-input-ignoredups
1350 The variable @code{comint-input-ignoredups} controls whether successive
1351 identical inputs are stored in the input history. A non-@code{nil}
1352 value means to omit an input that is the same as the previous input.
1353 The default is @code{nil}, which means to store each input even if it is
1354 equal to the previous input.
1356 @vindex comint-completion-addsuffix
1357 @vindex comint-completion-recexact
1358 @vindex comint-completion-autolist
1359 Three variables customize file name completion. The variable
1360 @code{comint-completion-addsuffix} controls whether completion inserts a
1361 space or a slash to indicate a fully completed file or directory name
1362 (non-@code{nil} means do insert a space or slash).
1363 @code{comint-completion-recexact}, if non-@code{nil}, directs @key{TAB}
1364 to choose the shortest possible completion if the usual Emacs completion
1365 algorithm cannot add even a single character.
1366 @code{comint-completion-autolist}, if non-@code{nil}, says to list all
1367 the possible completions whenever completion is not exact.
1369 @vindex shell-completion-execonly
1370 Command completion normally considers only executable files.
1371 If you set @code{shell-completion-execonly} to @code{nil},
1372 it considers nonexecutable files as well.
1374 @vindex shell-completion-fignore
1375 @vindex comint-completion-fignore
1376 The variable @code{shell-completion-fignore} specifies a list of file
1377 name extensions to ignore in Shell mode completion. The default
1378 setting is @code{nil}, but some users prefer @code{("~" "#" "%")} to
1379 ignore file names ending in @samp{~}, @samp{#} or @samp{%}. Other
1380 related Comint modes use the variable @code{comint-completion-fignore}
1383 @findex shell-dynamic-complete-command
1384 Some implementation details of the shell command completion may also be found
1385 in the lisp documentation of the @code{shell-dynamic-complete-command}
1388 @findex shell-pushd-tohome
1389 @findex shell-pushd-dextract
1390 @findex shell-pushd-dunique
1391 You can configure the behavior of @samp{pushd}. Variables control
1392 whether @samp{pushd} behaves like @samp{cd} if no argument is given
1393 (@code{shell-pushd-tohome}), pop rather than rotate with a numeric
1394 argument (@code{shell-pushd-dextract}), and only add directories to the
1395 directory stack if they are not already on it
1396 (@code{shell-pushd-dunique}). The values you choose should match the
1397 underlying shell, of course.
1399 @node Terminal emulator
1400 @subsection Emacs Terminal Emulator
1403 To run a subshell in a text terminal emulator, use @kbd{M-x term}.
1404 This creates (or reuses) a buffer named @file{*terminal*}, and runs a
1405 subshell with input coming from your keyboard, and output going to
1408 @cindex line mode @r{(terminal emulator)}
1409 @cindex char mode @r{(terminal emulator)}
1410 The terminal emulator uses Term mode, which has two input modes. In
1411 @dfn{line mode}, Term basically acts like Shell mode (@pxref{Shell
1412 Mode}). In @dfn{char mode}, each character is sent directly to the
1413 subshell, as terminal input; the sole exception is the terminal escape
1414 character, which by default is @kbd{C-c} (@pxref{Term Mode}). Any
1415 echoing of your input is the responsibility of the subshell; any
1416 terminal output from the subshell goes into the buffer, advancing
1419 Some programs (such as Emacs itself) need to control the appearance
1420 of the terminal screen in detail. They do this by emitting special
1421 control codes. Term mode recognizes and handles ANSI-standard
1422 VT100-style escape sequences, which are accepted by most modern
1423 terminals, including @command{xterm}. (Hence, you can actually run
1424 Emacs inside an Emacs Term window.)
1426 The @code{term} face specifies the default appearance of text
1427 in the terminal emulator (the default is the same appearance as the
1428 @code{default} face). When terminal control codes are used to change
1429 the appearance of text, these are represented in the terminal emulator
1430 by the faces @code{term-color-black}, @code{term-color-red},
1431 @code{term-color-green}, @code{term-color-yellow}
1432 @code{term-color-blue}, @code{term-color-magenta},
1433 @code{term-color-cyan}, @code{term-color-white},
1434 @code{term-color-underline}, and @code{term-color-bold}.
1437 You can also Term mode to communicate with a device connected to a
1438 serial port. @xref{Serial Terminal}.
1440 The file name used to load the subshell is determined the same way
1441 as for Shell mode. To make multiple terminal emulators, rename the
1442 buffer @file{*terminal*} to something different using @kbd{M-x
1443 rename-uniquely}, just as with Shell mode.
1445 Unlike Shell mode, Term mode does not track the current directory by
1446 examining your input. But some shells can tell Term what the current
1447 directory is. This is done automatically by @code{bash} version 1.15
1454 @subsection Term Mode
1458 The terminal emulator uses Term mode, which has two input modes. In
1459 line mode, Term basically acts like Shell mode (@pxref{Shell Mode}).
1460 In char mode, each character is sent directly to the subshell, except
1461 for the Term escape character, normally @kbd{C-c}.
1463 To switch between line and char mode, use these commands:
1466 @kindex C-c C-j @r{(Term mode)}
1467 @findex term-line-mode
1469 Switch to line mode (@code{term-line-mode}). Do nothing if already in
1472 @kindex C-c C-k @r{(Term mode)}
1473 @findex term-char-mode
1475 Switch to char mode (@code{term-char-mode}). Do nothing if already in
1479 The following commands are only available in char mode:
1483 Send a literal @kbd{C-c} to the sub-shell.
1485 @item C-c @var{char}
1486 This is equivalent to @kbd{C-x @var{char}} in normal Emacs. For
1487 example, @kbd{C-c o} invokes the global binding of @kbd{C-x o}, which
1488 is normally @samp{other-window}.
1491 @cindex paging in Term mode
1492 Term mode has a page-at-a-time feature. When enabled, it makes
1493 output pause at the end of each screenful:
1496 @kindex C-c C-q @r{(Term mode)}
1497 @findex term-pager-toggle
1499 Toggle the page-at-a-time feature. This command works in both line
1500 and char modes. When the feature is enabled, the mode-line displays
1501 the word @samp{page}, and each time Term receives more than a
1502 screenful of output, it pauses and displays @samp{**MORE**} in the
1503 mode-line. Type @key{SPC} to display the next screenful of output, or
1504 @kbd{?} to see your other options. The interface is similar to the
1505 @code{more} program.
1509 @subsection Remote Host Shell
1511 @cindex connecting to remote host
1515 You can login to a remote computer, using whatever commands you
1516 would from a regular terminal (e.g., using the @code{telnet} or
1517 @code{rlogin} commands), from a Term window.
1519 A program that asks you for a password will normally suppress
1520 echoing of the password, so the password will not show up in the
1521 buffer. This will happen just as if you were using a real terminal,
1522 if the buffer is in char mode. If it is in line mode, the password is
1523 temporarily visible, but will be erased when you hit return. (This
1524 happens automatically; there is no special password processing.)
1526 When you log in to a different machine, you need to specify the type
1527 of terminal you're using, by setting the @env{TERM} environment
1528 variable in the environment for the remote login command. (If you use
1529 bash, you do that by writing the variable assignment before the remote
1530 login command, without a separating comma.) Terminal types
1531 @samp{ansi} or @samp{vt100} will work on most systems.
1533 @node Serial Terminal
1534 @subsection Serial Terminal
1535 @cindex terminal, serial
1538 If you have a device connected to a serial port of your computer,
1539 you can communicate with it by typing @kbd{M-x serial-term}. This
1540 command asks for a serial port name and speed, and switches to a new
1541 Term mode buffer. Emacs communicates with the serial device through
1542 this buffer just like it does with a terminal in ordinary Term mode.
1544 The speed of the serial port is measured in bits per second. The
1545 most common speed is 9600 bits per second. You can change the speed
1546 interactively by clicking on the mode line.
1548 A serial port can be configured even more by clicking on @samp{8N1} in
1549 the mode line. By default, a serial port is configured as @samp{8N1},
1550 which means that each byte consists of 8 data bits, No parity check
1553 If the speed or the configuration is wrong, you cannot communicate
1554 with your device and will probably only see garbage output in the
1558 @section Using Emacs as a Server
1560 @cindex Emacs as a server
1561 @cindex server, using Emacs as
1562 @cindex @env{EDITOR} environment variable
1564 Various programs can invoke your choice of editor to edit a
1565 particular piece of text. For instance, version control programs
1566 invoke an editor to enter version control logs (@pxref{Version
1567 Control}), and the Unix @command{mail} utility invokes an editor to
1568 enter a message to send. By convention, your choice of editor is
1569 specified by the environment variable @env{EDITOR}. If you set
1570 @env{EDITOR} to @samp{emacs}, Emacs would be invoked, but in an
1571 inconvenient way---by starting a new Emacs process. This is
1572 inconvenient because the new Emacs process doesn't share buffers, a
1573 command history, or other kinds of information with any existing Emacs
1576 You can solve this problem by setting up Emacs as an @dfn{edit
1577 server}, so that it ``listens'' for external edit requests and acts
1578 accordingly. There are various ways to start an Emacs server:
1581 @findex server-start
1583 Run the command @code{server-start} in an existing Emacs process:
1584 either type @kbd{M-x server-start}, or put the expression
1585 @code{(server-start)} in your init file (@pxref{Init File}). The
1586 existing Emacs process is the server; when you exit Emacs, the server
1587 dies with the Emacs process.
1589 @cindex daemon, Emacs
1591 Run Emacs as a @dfn{daemon}, using one of the @samp{--daemon} command-line
1592 options. @xref{Initial Options}. When Emacs is started this way, it
1593 calls @code{server-start} after initialization and does not open an
1594 initial frame. It then waits for edit requests from clients.
1596 @cindex systemd unit file
1598 If your operating system uses @command{systemd} to manage startup,
1599 you can automatically start Emacs in daemon mode when you login
1600 using the supplied @dfn{systemd unit file}. To activate this:
1602 systemctl --user enable emacs
1604 (If your Emacs was installed into a non-standard location, you may
1605 need to copy the @file{emacs.service} file to a standard directory
1606 such as @file{~/.config/systemd/user/}.)
1608 @cindex socket activation, systemd, Emacs
1610 An external process can invoke the Emacs server when a connection
1611 event occurs upon a specified socket and pass the socket to the new
1612 Emacs server process. An instance of this is the socket functionality
1613 of @command{systemd}: the @command{systemd} service creates a socket and
1614 listens for connections on it; when @command{emacsclient} connects to
1615 it for the first time, @command{systemd} can launch the Emacs server
1616 and hand over the socket to it for servicing @command{emacsclient}
1617 connections. A setup to use this functionality could be:
1619 @file{~/.config/systemd/user/emacs.socket}:
1622 ListenStream=/path/to/.emacs.socket
1625 WantedBy=sockets.target
1628 (The @file{emacs.service} file described above must also be installed.)
1630 The @code{ListenStream} path will be the path that Emacs listens for
1631 connections from @command{emacsclient}; this is a file of your choice.
1634 @cindex @env{TEXEDIT} environment variable
1635 Once an Emacs server is started, you can use a shell
1636 command called @command{emacsclient} to connect to the Emacs process
1637 and tell it to visit a file. You can then set the @env{EDITOR}
1638 environment variable to @samp{emacsclient}, so that external programs
1639 will use the existing Emacs process for editing.@footnote{Some
1640 programs use a different environment variable; for example, to make
1641 @TeX{} use @samp{emacsclient}, set the @env{TEXEDIT} environment
1642 variable to @samp{emacsclient +%d %s}.}
1645 You can run multiple Emacs servers on the same machine by giving
1646 each one a unique @dfn{server name}, using the variable
1647 @code{server-name}. For example, @kbd{M-x set-variable @key{RET}
1648 server-name @key{RET} "foo" @key{RET}} sets the server name to
1649 @samp{foo}. The @code{emacsclient} program can specify a server by
1650 name, using the @samp{-s} option (@pxref{emacsclient Options}).
1652 If you want to run multiple Emacs daemons (@pxref{Initial Options}),
1653 you can give each daemon its own server name like this:
1656 emacs --eval "(setq server-name \"foo\")" --daemon
1659 @findex server-eval-at
1660 If you have defined a server by a unique server name, it is possible
1661 to connect to the server from another Emacs instance and evaluate Lisp
1662 expressions on the server, using the @code{server-eval-at} function.
1663 For instance, @code{(server-eval-at "foo" '(+ 1 2))} evaluates the
1664 expression @code{(+ 1 2)} on the @samp{foo} server, and returns
1665 @code{3}. (If there is no server with that name, an error is
1666 signaled.) Currently, this feature is mainly useful for developers.
1669 * TCP Emacs server:: Listening to a TCP socket.
1670 * Invoking emacsclient:: Connecting to the Emacs server.
1671 * emacsclient Options:: Emacs client startup options.
1674 @node TCP Emacs server
1675 @subsection TCP Emacs server
1676 @cindex TCP Emacs server
1678 @vindex server-use-tcp
1679 An Emacs server usually listens to connections on a local Unix
1680 domain socket. Some operating systems, such as MS-Windows, do not
1681 support local sockets; in that case, the server uses TCP sockets
1682 instead. In some cases it is useful to have the server listen on a
1683 TCP socket even if local sockets are supported, e.g., if you need to
1684 contact the Emacs server from a remote machine. You can set
1685 @code{server-use-tcp} to non-@code{nil} to have Emacs listen on a TCP
1686 socket instead of a local socket. This is the default if your OS does
1687 not support local sockets.
1691 If the Emacs server is set to use TCP, it will by default listen to
1692 a random port on the localhost interface. This can be changed to
1693 another interface and/or a fixed port using the variables
1694 @code{server-host} and @code{server-port}.
1696 @vindex server-auth-key
1697 A TCP socket is not subject to file system permissions. To retain
1698 some control over which users can talk to an Emacs server over TCP
1699 sockets, the @command{emacsclient} program must send an authorization
1700 key to the server. This key is normally randomly generated by the
1701 Emacs server. This is the recommended mode of operation.
1703 @findex server-generate-key
1704 If needed, you can set the authorization key to a static value by
1705 setting the @code{server-auth-key} variable. The key must consist of
1706 64 ASCII printable characters except for space (this means characters
1707 from @samp{!} to @samp{~}, or from decimal code 33 to 126). You can
1708 use @kbd{M-x server-generate-key} to get a random key.
1710 @vindex server-auth-dir
1712 When you start a TCP Emacs server, Emacs creates a @dfn{server file}
1713 containing the TCP information to be used by @command{emacsclient} to
1714 connect to the server. The variable @code{server-auth-dir} specifies
1715 the directory containing the server file; by default, this is
1716 @file{~/.emacs.d/server/}. In the absence of a local socket with file
1717 permissions, the permissions of this directory determine which users
1718 can have their @command{emacsclient} processes talk to the Emacs
1721 @vindex EMACS_SERVER_FILE@r{, environment variable}
1722 To tell @command{emacsclient} to connect to the server over TCP with
1723 a specific server file, use the @samp{-f} or @samp{--server-file}
1724 option, or set the @env{EMACS_SERVER_FILE} environment variable
1725 (@pxref{emacsclient Options}). If @code{server-auth-dir} is set to a
1726 non-standard value, @command{emacsclient} needs an absolute file name
1727 to the server file, as the default @code{server-auth-dir} is
1728 hard-coded in @command{emacsclient} to be used as the directory for
1729 resolving relative filenames.
1731 @node Invoking emacsclient
1732 @subsection Invoking @code{emacsclient}
1733 @cindex @code{emacsclient} invocation
1735 The simplest way to use the @command{emacsclient} program is to run
1736 the shell command @samp{emacsclient @var{file}}, where @var{file} is a
1737 file name. This connects to an Emacs server, and tells that Emacs
1738 process to visit @var{file} in one of its existing frames---either a
1739 graphical frame, or one in a text terminal (@pxref{Frames}). You
1740 can then select that frame to begin editing.
1742 If there is no Emacs server, the @command{emacsclient} program halts
1743 with an error message. If the Emacs process has no existing
1744 frame---which can happen if it was started as a daemon (@pxref{Emacs
1745 Server})---then Emacs opens a frame on the terminal in which you
1746 called @command{emacsclient}.
1748 You can also force @command{emacsclient} to open a new frame on a
1749 graphical display, or on a text terminal, using the @samp{-c} and
1750 @samp{-t} options. @xref{emacsclient Options}.
1752 If you are running on a single text terminal, you can switch between
1753 @command{emacsclient}'s shell and the Emacs server using one of two
1754 methods: (i) run the Emacs server and @command{emacsclient} on
1755 different virtual terminals, and switch to the Emacs server's virtual
1756 terminal after calling @command{emacsclient}; or (ii) call
1757 @command{emacsclient} from within the Emacs server itself, using Shell
1758 mode (@pxref{Interactive Shell}) or Term mode (@pxref{Term Mode});
1759 @code{emacsclient} blocks only the subshell under Emacs, and you can
1760 still use Emacs to edit the file.
1764 When you finish editing @var{file} in the Emacs server, type
1765 @kbd{C-x #} (@code{server-edit}) in its buffer. This saves the file
1766 and sends a message back to the @command{emacsclient} program, telling
1767 it to exit. Programs that use @env{EDITOR} usually wait for the
1768 editor---in this case @command{emacsclient}---to exit before doing
1771 You can also call @command{emacsclient} with multiple file name
1772 arguments: @samp{emacsclient @var{file1} @var{file2} ...} tells the
1773 Emacs server to visit @var{file1}, @var{file2}, and so forth. Emacs
1774 selects the buffer visiting @var{file1}, and buries the other buffers
1775 at the bottom of the buffer list (@pxref{Buffers}). The
1776 @command{emacsclient} program exits once all the specified files are
1777 finished (i.e., once you have typed @kbd{C-x #} in each server
1780 @vindex server-kill-new-buffers
1781 @vindex server-temp-file-regexp
1782 Finishing with a server buffer also kills the buffer, unless it
1783 already existed in the Emacs session before the server was asked to
1784 create it. However, if you set @code{server-kill-new-buffers} to
1785 @code{nil}, then a different criterion is used: finishing with a
1786 server buffer kills it if the file name matches the regular expression
1787 @code{server-temp-file-regexp}. This is set up to distinguish certain
1790 Each @kbd{C-x #} checks for other pending external requests to edit
1791 various files, and selects the next such file. You can switch to a
1792 server buffer manually if you wish; you don't have to arrive at it
1793 with @kbd{C-x #}. But @kbd{C-x #} is the way to tell
1794 @command{emacsclient} that you are finished.
1796 @vindex server-window
1797 If you set the value of the variable @code{server-window} to a
1798 window or a frame, @kbd{C-x #} always displays the next server buffer
1799 in that window or in that frame.
1801 @node emacsclient Options
1802 @subsection @code{emacsclient} Options
1803 @cindex @code{emacsclient} options
1805 You can pass some optional arguments to the @command{emacsclient}
1809 emacsclient -c +12 @var{file1} +4:3 @var{file2}
1813 The @samp{+@var{line}} or @samp{+@var{line}:@var{column}} arguments
1814 specify line numbers, or line and column numbers, for the next file
1815 argument. These behave like the command line arguments for Emacs
1816 itself. @xref{Action Arguments}.
1818 The other optional arguments recognized by @command{emacsclient} are
1822 @item -a @var{command}
1823 @itemx --alternate-editor=@var{command}
1824 Specify a shell command to run if @code{emacsclient} fails to contact Emacs.
1825 This is useful when running @code{emacsclient} in a script.
1826 The command may include arguments, which may be quoted "like this".
1827 Currently, escaping of quotes is not supported.
1829 As a special exception, if @var{command} is the empty string, then
1830 @code{emacsclient} starts Emacs in daemon mode (as @command{emacs
1831 --daemon}) and then tries connecting again.
1833 @cindex @env{ALTERNATE_EDITOR} environment variable
1834 The environment variable @env{ALTERNATE_EDITOR} has the same effect as
1835 the @samp{-a} option. If both are present, the latter takes
1838 @cindex client frame
1840 @itemx --create-frame
1841 Create a new graphical @dfn{client frame}, instead of using an
1842 existing Emacs frame. See below for the special behavior of @kbd{C-x
1843 C-c} in a client frame. If Emacs cannot create a new graphical frame
1844 (e.g., if it cannot connect to the X server), it tries to create a
1845 text terminal client frame, as though you had supplied the @samp{-t}
1848 On MS-Windows, a single Emacs session cannot display frames on both
1849 graphical and text terminals, nor on multiple text terminals. Thus,
1850 if the Emacs server is running on a text terminal, the @samp{-c}
1851 option, like the @samp{-t} option, creates a new frame in the server's
1852 current text terminal. @xref{Windows Startup}.
1854 If you omit a filename argument while supplying the @samp{-c} option,
1855 the new frame displays the @file{*scratch*} buffer by default. You
1856 can customize this behavior with the variable @code{initial-buffer-choice}
1857 (@pxref{Entering Emacs}).
1859 @item -F @var{alist}
1860 @itemx --frame-parameters=@var{alist}
1861 Set the parameters for a newly-created graphical frame
1862 (@pxref{Frame Parameters}).
1864 @item -d @var{display}
1865 @itemx --display=@var{display}
1866 Tell Emacs to open the given files on the X display @var{display}
1867 (assuming there is more than one X display available).
1871 Tell Emacs to evaluate some Emacs Lisp code, instead of visiting some
1872 files. When this option is given, the arguments to
1873 @command{emacsclient} are interpreted as a list of expressions to
1874 evaluate, @emph{not} as a list of files to visit.
1876 @item -f @var{server-file}
1877 @itemx --server-file=@var{server-file}
1878 Specify a server file (@pxref{TCP Emacs server}) for connecting to an
1879 Emacs server via TCP. Alternatively, you can set the
1880 @env{EMACS_SERVER_FILE} environment variable to point to the server
1883 An Emacs server usually uses a local socket to listen for connections,
1884 but also supports connections over TCP. To connect to a TCP Emacs
1885 server, @command{emacsclient} needs to read a @dfn{server file}
1886 containing the connection details of the Emacs server. The name of
1887 this file is specified with this option, either as a file name
1888 relative to @file{~/.emacs.d/server} or as an absolute file name.
1889 @xref{TCP Emacs server}.
1893 Let @command{emacsclient} exit immediately, instead of waiting until
1894 all server buffers are finished. You can take as long as you like to
1895 edit the server buffers within Emacs, and they are @emph{not} killed
1896 when you type @kbd{C-x #} in them.
1898 @item --parent-id @var{id}
1899 Open an @command{emacsclient} frame as a client frame in the parent X
1900 window with id @var{id}, via the XEmbed protocol. Currently, this
1901 option is mainly useful for developers.
1905 Do not let @command{emacsclient} display messages about waiting for
1906 Emacs or connecting to remote server sockets.
1909 @itemx --suppress-output
1910 Do not let @command{emacsclient} display results returned from the
1911 server. Mostly useful in combination with @samp{-e} when the
1912 evaluation performed is for side-effect rather than result.
1914 @item -s @var{server-name}
1915 @itemx --socket-name=@var{server-name}
1916 Connect to the Emacs server named @var{server-name}. The server name
1917 is given by the variable @code{server-name} on the Emacs server. If
1918 this option is omitted, @command{emacsclient} connects to the first
1919 server it finds. (This option is not supported on MS-Windows.)
1924 Create a new client frame on the current text terminal, instead of
1925 using an existing Emacs frame. This behaves just like the @samp{-c}
1926 option, described above, except that it creates a text terminal frame
1927 (@pxref{Non-Window Terminals}).
1929 On MS-Windows, @samp{-t} behaves just like @samp{-c} if the Emacs
1930 server is using the graphical display, but if the Emacs server is
1931 running on a text terminal, it creates a new frame in the current text
1934 @item -T @var{tramp-prefix}
1935 @itemx --tramp-prefix=@var{tramp-prefix}
1936 Set the prefix to add to filenames for Emacs to locate files on remote
1937 machines using TRAMP (@pxref{Top, The Tramp Manual,, tramp, The Tramp
1938 Manual}). This is mostly useful in combination with using the Emacs
1939 server over TCP (@pxref{TCP Emacs server}). By ssh-forwarding the
1940 listening port and making the @var{server-file} available on a remote
1941 machine, programs on the remote machine can use @command{emacsclient}
1942 as the value for the @env{EDITOR} and similar environment variables,
1943 but instead of talking to an Emacs server on the remote machine, the
1944 files will be visited in the local Emacs session using TRAMP.
1946 @vindex EMACSCLIENT_TRAMP@r{, environment variable}
1947 Setting the environment variable @env{EMACSCLIENT_TRAMP} has the same
1948 effect as using the @samp{-T} option. If both are specified, the
1949 command-line option takes precedence.
1951 For example, assume two hosts, @samp{local} and @samp{remote}, and
1952 that the local Emacs listens on tcp port 12345. Assume further that
1953 @file{/home} is on a shared file system, so that the server file
1954 @file{~/.emacs.d/server/server} is readable on both hosts.
1957 local$ ssh -R12345:localhost:12345 remote
1958 remote$ export EDITOR="emacsclient \
1959 --server-file=server \
1960 --tramp=/ssh:remote:"
1961 remote$ $EDITOR /tmp/foo.txt #Should open in local emacs.
1966 The new graphical or text terminal frames created by the @samp{-c}
1967 or @samp{-t} options are considered @dfn{client frames}. Any new
1968 frame that you create from a client frame is also considered a client
1969 frame. If you type @kbd{C-x C-c} (@code{save-buffers-kill-terminal})
1970 in a client frame, that command does not kill the Emacs session as it
1971 normally does (@pxref{Exiting}). Instead, Emacs deletes the client
1972 frame; furthermore, if the client frame has an @command{emacsclient}
1973 waiting to regain control (i.e., if you did not supply the @samp{-n}
1974 option), Emacs deletes all other frames of the same client, and marks
1975 the client's server buffers as finished, as though you had typed
1976 @kbd{C-x #} in all of them. If it so happens that there are no
1977 remaining frames after the client frame(s) are deleted, the Emacs
1980 As an exception, when Emacs is started as a daemon, all frames are
1981 considered client frames, and @kbd{C-x C-c} never kills Emacs. To
1982 kill a daemon session, type @kbd{M-x kill-emacs}.
1984 Note that the @samp{-t} and @samp{-n} options are contradictory:
1985 @samp{-t} says to take control of the current text terminal to create
1986 a new client frame, while @samp{-n} says not to take control of the
1987 text terminal. If you supply both options, Emacs visits the specified
1988 files(s) in an existing frame rather than a new client frame, negating
1989 the effect of @samp{-t}.
1992 @section Printing Hard Copies
1996 Emacs provides commands for printing hardcopies of either an entire
1997 buffer or part of one. You can invoke the printing commands directly,
1998 as detailed below, or using the @samp{File} menu on the menu bar.
2000 @findex htmlfontify-buffer
2001 Aside from the commands described in this section, you can also
2002 print hardcopies from Dired (@pxref{Operating on Files}) and the diary
2003 (@pxref{Displaying the Diary}). You can also ``print'' an Emacs
2004 buffer to HTML with the command @kbd{M-x htmlfontify-buffer}, which
2005 converts the current buffer to a HTML file, replacing Emacs faces with
2006 CSS-based markup. Furthermore, Org mode allows you to print Org
2007 files to a variety of formats, such as PDF (@pxref{Org Mode}).
2010 @item M-x print-buffer
2011 Print hardcopy of current buffer with page headings containing the
2012 file name and page number.
2013 @item M-x lpr-buffer
2014 Print hardcopy of current buffer without page headings.
2015 @item M-x print-region
2016 Like @code{print-buffer} but print only the current region.
2017 @item M-x lpr-region
2018 Like @code{lpr-buffer} but print only the current region.
2021 @findex print-buffer
2022 @findex print-region
2025 @vindex lpr-switches
2026 @vindex lpr-commands
2027 On most operating systems, the above hardcopy commands submit files
2028 for printing by calling the @command{lpr} program. To change the
2029 printer program, customize the variable @code{lpr-command}. To
2030 specify extra switches to give the printer program, customize the list
2031 variable @code{lpr-switches}. Its value should be a list of option
2032 strings, each of which should start with @samp{-} (e.g., the option
2033 string @code{"-w80"} specifies a line width of 80 columns). The
2034 default is the empty list, @code{nil}.
2036 @vindex printer-name
2037 @vindex lpr-printer-switch
2038 To specify the printer to use, set the variable @code{printer-name}.
2039 The default, @code{nil}, specifies the default printer. If you set it
2040 to a printer name (a string), that name is passed to @command{lpr}
2041 with the @samp{-P} switch; if you are not using @command{lpr}, you
2042 should specify the switch with @code{lpr-printer-switch}.
2044 @vindex lpr-headers-switches
2045 @vindex lpr-add-switches
2046 The variable @code{lpr-headers-switches} similarly specifies the
2047 extra switches to use to make page headers. The variable
2048 @code{lpr-add-switches} controls whether to supply @samp{-T} and
2049 @samp{-J} options (suitable for @command{lpr}) to the printer program:
2050 @code{nil} means don't add them (this should be the value if your
2051 printer program is not compatible with @command{lpr}).
2054 * PostScript:: Printing buffers or regions as PostScript.
2055 * PostScript Variables:: Customizing the PostScript printing commands.
2056 * Printing Package:: An optional advanced printing interface.
2060 @subsection PostScript Hardcopy
2062 These commands convert buffer contents to PostScript,
2063 either printing it or leaving it in another Emacs buffer.
2066 @item M-x ps-print-buffer
2067 Print hardcopy of the current buffer in PostScript form.
2068 @item M-x ps-print-region
2069 Print hardcopy of the current region in PostScript form.
2070 @item M-x ps-print-buffer-with-faces
2071 Print hardcopy of the current buffer in PostScript form, showing the
2072 faces used in the text by means of PostScript features.
2073 @item M-x ps-print-region-with-faces
2074 Print hardcopy of the current region in PostScript form, showing the
2075 faces used in the text.
2076 @item M-x ps-spool-buffer
2077 Generate and spool a PostScript image for the current buffer text.
2078 @item M-x ps-spool-region
2079 Generate and spool a PostScript image for the current region.
2080 @item M-x ps-spool-buffer-with-faces
2081 Generate and spool a PostScript image for the current buffer, showing the faces used.
2082 @item M-x ps-spool-region-with-faces
2083 Generate and spool a PostScript image for the current region, showing the faces used.
2084 @item M-x ps-despool
2085 Send the spooled PostScript to the printer.
2087 Generate/print PostScript for the current buffer as if handwritten.
2090 @findex ps-print-region
2091 @findex ps-print-buffer
2092 @findex ps-print-region-with-faces
2093 @findex ps-print-buffer-with-faces
2094 The @code{ps-print-buffer} and @code{ps-print-region} commands print
2095 buffer contents in PostScript form. One command prints the entire
2096 buffer; the other, just the region. The commands
2097 @code{ps-print-buffer-with-faces} and
2098 @code{ps-print-region-with-faces} behave similarly, but use PostScript
2099 features to show the faces (fonts and colors) of the buffer text.
2101 Interactively, when you use a prefix argument (@kbd{C-u}), the command
2102 prompts the user for a file name, and saves the PostScript image in that file
2103 instead of sending it to the printer.
2105 @findex ps-spool-region
2106 @findex ps-spool-buffer
2107 @findex ps-spool-region-with-faces
2108 @findex ps-spool-buffer-with-faces
2109 The commands whose names have @samp{spool} instead of @samp{print},
2110 generate the PostScript output in an Emacs buffer instead of sending
2114 Use the command @code{ps-despool} to send the spooled images to the
2115 printer. This command sends the PostScript generated by
2116 @samp{-spool-} commands (see commands above) to the printer. With a
2117 prefix argument (@kbd{C-u}), it prompts for a file name, and saves the
2118 spooled PostScript image in that file instead of sending it to the
2123 @kbd{M-x handwrite} is more frivolous. It generates a PostScript
2124 rendition of the current buffer as a cursive handwritten document. It
2125 can be customized in group @code{handwrite}. This function only
2126 supports ISO 8859-1 characters.
2128 @node PostScript Variables
2129 @subsection Variables for PostScript Hardcopy
2131 @vindex ps-lpr-command
2132 @vindex ps-lpr-switches
2133 @vindex ps-printer-name
2134 All the PostScript hardcopy commands use the variables
2135 @code{ps-lpr-command} and @code{ps-lpr-switches} to specify how to print
2136 the output. @code{ps-lpr-command} specifies the command name to run,
2137 @code{ps-lpr-switches} specifies command line options to use, and
2138 @code{ps-printer-name} specifies the printer. If you don't set the
2139 first two variables yourself, they take their initial values from
2140 @code{lpr-command} and @code{lpr-switches}. If @code{ps-printer-name}
2141 is @code{nil}, @code{printer-name} is used.
2143 @vindex ps-print-header
2144 The variable @code{ps-print-header} controls whether these commands
2145 add header lines to each page---set it to @code{nil} to turn headers
2148 @cindex color emulation on black-and-white printers
2149 @vindex ps-print-color-p
2150 If your printer doesn't support colors, you should turn off color
2151 processing by setting @code{ps-print-color-p} to @code{nil}. By
2152 default, if the display supports colors, Emacs produces hardcopy output
2153 with color information; on black-and-white printers, colors are emulated
2154 with shades of gray. This might produce illegible output, even if your
2155 screen colors only use shades of gray.
2157 Alternatively, you can set @code{ps-print-color-p} to @code{black-white} to
2158 print colors on black/white printers.
2160 @vindex ps-use-face-background
2161 By default, PostScript printing ignores the background colors of the
2162 faces, unless the variable @code{ps-use-face-background} is
2163 non-@code{nil}. This is to avoid unwanted interference with the zebra
2164 stripes and background image/text.
2166 @vindex ps-paper-type
2167 @vindex ps-page-dimensions-database
2168 The variable @code{ps-paper-type} specifies which size of paper to
2169 format for; legitimate values include @code{a4}, @code{a3},
2170 @code{a4small}, @code{b4}, @code{b5}, @code{executive}, @code{ledger},
2171 @code{legal}, @code{letter}, @code{letter-small}, @code{statement},
2172 @code{tabloid}. The default is @code{letter}. You can define
2173 additional paper sizes by changing the variable
2174 @code{ps-page-dimensions-database}.
2176 @vindex ps-landscape-mode
2177 The variable @code{ps-landscape-mode} specifies the orientation of
2178 printing on the page. The default is @code{nil}, which stands for
2179 portrait mode. Any non-@code{nil} value specifies landscape
2182 @vindex ps-number-of-columns
2183 The variable @code{ps-number-of-columns} specifies the number of
2184 columns; it takes effect in both landscape and portrait mode. The
2187 @vindex ps-font-family
2188 @vindex ps-font-size
2189 @vindex ps-font-info-database
2190 The variable @code{ps-font-family} specifies which font family to use
2191 for printing ordinary text. Legitimate values include @code{Courier},
2192 @code{Helvetica}, @code{NewCenturySchlbk}, @code{Palatino} and
2193 @code{Times}. The variable @code{ps-font-size} specifies the size of
2194 the font for ordinary text. It defaults to 8.5 points.
2196 @vindex ps-multibyte-buffer
2197 @cindex Intlfonts for PostScript printing
2198 @cindex fonts for PostScript printing
2199 Emacs supports more scripts and characters than a typical PostScript
2200 printer. Thus, some of the characters in your buffer might not be
2201 printable using the fonts built into your printer. You can augment
2202 the fonts supplied with the printer with those from the GNU Intlfonts
2203 package, or you can instruct Emacs to use Intlfonts exclusively. The
2204 variable @code{ps-multibyte-buffer} controls this: the default value,
2205 @code{nil}, is appropriate for printing @acronym{ASCII} and Latin-1
2206 characters; a value of @code{non-latin-printer} is for printers which
2207 have the fonts for @acronym{ASCII}, Latin-1, Japanese, and Korean
2208 characters built into them. A value of @code{bdf-font} arranges for
2209 the BDF fonts from the Intlfonts package to be used for @emph{all}
2210 characters. Finally, a value of @code{bdf-font-except-latin}
2211 instructs the printer to use built-in fonts for @acronym{ASCII} and Latin-1
2212 characters, and Intlfonts BDF fonts for the rest.
2214 @vindex bdf-directory-list
2215 To be able to use the BDF fonts, Emacs needs to know where to find
2216 them. The variable @code{bdf-directory-list} holds the list of
2217 directories where Emacs should look for the fonts; the default value
2218 includes a single directory @file{/usr/local/share/emacs/fonts/bdf}.
2220 Many other customization variables for these commands are defined and
2221 described in the Lisp files @file{ps-print.el} and @file{ps-mule.el}.
2223 @node Printing Package
2224 @subsection Printing Package
2225 @cindex Printing package
2227 The basic Emacs facilities for printing hardcopy can be extended
2228 using the Printing package. This provides an easy-to-use interface
2229 for choosing what to print, previewing PostScript files before
2230 printing, and setting various printing options such as print headers,
2231 landscape or portrait modes, duplex modes, and so forth. On GNU/Linux
2232 or Unix systems, the Printing package relies on the @file{gs} and
2233 @file{gv} utilities, which are distributed as part of the GhostScript
2234 program. On MS-Windows, the @file{gstools} port of Ghostscript can be
2237 @findex pr-interface
2238 To use the Printing package, add @code{(require 'printing)} to your
2239 init file (@pxref{Init File}), followed by @code{(pr-update-menus)}.
2240 This function replaces the usual printing commands in the menu bar
2241 with a @samp{Printing} submenu that contains various printing options.
2242 You can also type @kbd{M-x pr-interface @key{RET}}; this creates a
2243 @file{*Printing Interface*} buffer, similar to a customization buffer,
2244 where you can set the printing options. After selecting what and how
2245 to print, you start the print job using the @samp{Print} button (click
2246 @kbd{mouse-2} on it, or move point over it and type @key{RET}). For
2247 further information on the various options, use the @samp{Interface
2251 @section Sorting Text
2254 Emacs provides several commands for sorting text in the buffer. All
2255 operate on the contents of the region.
2256 They divide the text of the region into many @dfn{sort records},
2257 identify a @dfn{sort key} for each record, and then reorder the records
2258 into the order determined by the sort keys. The records are ordered so
2259 that their keys are in alphabetical order, or, for numeric sorting, in
2260 numeric order. In alphabetic sorting, all upper-case letters @samp{A}
2261 through @samp{Z} come before lower-case @samp{a}, in accordance with the
2262 @acronym{ASCII} character sequence.
2264 The various sort commands differ in how they divide the text into sort
2265 records and in which part of each record is used as the sort key. Most of
2266 the commands make each line a separate sort record, but some commands use
2267 paragraphs or pages as sort records. Most of the sort commands use each
2268 entire sort record as its own sort key, but some use only a portion of the
2269 record as the sort key.
2272 @findex sort-paragraphs
2275 @findex sort-numeric-fields
2276 @vindex sort-numeric-base
2278 @item M-x sort-lines
2279 Divide the region into lines, and sort by comparing the entire
2280 text of a line. A numeric argument means sort into descending order.
2282 @item M-x sort-paragraphs
2283 Divide the region into paragraphs, and sort by comparing the entire
2284 text of a paragraph (except for leading blank lines). A numeric
2285 argument means sort into descending order.
2287 @item M-x sort-pages
2288 Divide the region into pages, and sort by comparing the entire
2289 text of a page (except for leading blank lines). A numeric
2290 argument means sort into descending order.
2292 @item M-x sort-fields
2293 Divide the region into lines, and sort by comparing the contents of
2294 one field in each line. Fields are defined as separated by
2295 whitespace, so the first run of consecutive non-whitespace characters
2296 in a line constitutes field 1, the second such run constitutes field
2299 Specify which field to sort by with a numeric argument: 1 to sort by
2300 field 1, etc. A negative argument means count fields from the right
2301 instead of from the left; thus, minus 1 means sort by the last field.
2302 If several lines have identical contents in the field being sorted, they
2303 keep the same relative order that they had in the original buffer.
2305 @item M-x sort-numeric-fields
2306 Like @kbd{M-x sort-fields} except the specified field is converted
2307 to an integer for each line, and the numbers are compared. @samp{10}
2308 comes before @samp{2} when considered as text, but after it when
2309 considered as a number. By default, numbers are interpreted according
2310 to @code{sort-numeric-base}, but numbers beginning with @samp{0x} or
2311 @samp{0} are interpreted as hexadecimal and octal, respectively.
2313 @item M-x sort-columns
2314 Like @kbd{M-x sort-fields} except that the text within each line
2315 used for comparison comes from a fixed range of columns. With a
2316 prefix argument, sort in reverse order. See below for more details
2319 @findex reverse-region
2320 @item M-x reverse-region
2321 Reverse the order of the lines in the region. This is useful for
2322 sorting into descending order by fields, since those sort
2323 commands do not have a feature for doing that.
2326 For example, if the buffer contains this:
2329 On systems where clash detection (locking of files being edited) is
2330 implemented, Emacs also checks the first time you modify a buffer
2331 whether the file has changed on disk since it was last visited or
2332 saved. If it has, you are asked to confirm that you want to change
2337 applying @kbd{M-x sort-lines} to the entire buffer produces this:
2340 On systems where clash detection (locking of files being edited) is
2341 implemented, Emacs also checks the first time you modify a buffer
2342 saved. If it has, you are asked to confirm that you want to change
2344 whether the file has changed on disk since it was last visited or
2348 where the upper-case @samp{O} sorts before all lower-case letters. If
2349 you use @kbd{C-u 2 M-x sort-fields} instead, you get this:
2352 implemented, Emacs also checks the first time you modify a buffer
2353 saved. If it has, you are asked to confirm that you want to change
2355 On systems where clash detection (locking of files being edited) is
2356 whether the file has changed on disk since it was last visited or
2360 where the sort keys were @samp{Emacs}, @samp{If}, @samp{buffer},
2361 @samp{systems} and @samp{the}.
2363 @findex sort-columns
2364 @kbd{M-x sort-columns} requires more explanation. You specify the
2365 columns by putting point at one of the columns and the mark at the other
2366 column. Because this means you cannot put point or the mark at the
2367 beginning of the first line of the text you want to sort, this command
2368 uses an unusual definition of ``region'': all of the line point is in is
2369 considered part of the region, and so is all of the line the mark is in,
2370 as well as all the lines in between.
2372 For example, to sort a table by information found in columns 10 to 15,
2373 you could put the mark on column 10 in the first line of the table, and
2374 point on column 15 in the last line of the table, and then run
2375 @code{sort-columns}. Equivalently, you could run it with the mark on
2376 column 15 in the first line and point on column 10 in the last line.
2378 This can be thought of as sorting the rectangle specified by point and
2379 the mark, except that the text on each line to the left or right of the
2380 rectangle moves along with the text inside the rectangle.
2383 @vindex sort-fold-case
2384 Many of the sort commands ignore case differences when comparing, if
2385 @code{sort-fold-case} is non-@code{nil}.
2387 @c Picture Mode documentation
2389 @include picture-xtra.texi
2393 @node Editing Binary Files
2394 @section Editing Binary Files
2398 @cindex editing binary files
2400 There is a special major mode for editing binary files: Hexl mode. To
2401 use it, use @kbd{M-x hexl-find-file} instead of @kbd{C-x C-f} to visit
2402 the file. This command converts the file's contents to hexadecimal and
2403 lets you edit the translation. When you save the file, it is converted
2404 automatically back to binary.
2406 You can also use @kbd{M-x hexl-mode} to translate an existing buffer
2407 into hex. This is useful if you visit a file normally and then discover
2408 it is a binary file.
2410 Ordinary text characters overwrite in Hexl mode. This is to reduce
2411 the risk of accidentally spoiling the alignment of data in the file.
2412 There are special commands for insertion. Here is a list of the
2413 commands of Hexl mode:
2415 @c I don't think individual index entries for these commands are useful--RMS.
2418 Insert a byte with a code typed in decimal.
2421 Insert a byte with a code typed in octal.
2424 Insert a byte with a code typed in hex.
2427 Move to the beginning of a 1k-byte page.
2430 Move to the end of a 1k-byte page.
2433 Move to an address specified in hex.
2436 Move to an address specified in decimal.
2439 Leave Hexl mode, going back to the major mode this buffer had before you
2440 invoked @code{hexl-mode}.
2444 Other Hexl commands let you insert strings (sequences) of binary
2445 bytes, move by @code{short}s or @code{int}s, etc.; type @kbd{C-h a
2446 hexl-@key{RET}} for details.
2449 @node Saving Emacs Sessions
2450 @section Saving Emacs Sessions
2451 @cindex saving sessions
2452 @cindex restore session
2453 @cindex remember editing session
2454 @cindex reload files
2457 @vindex desktop-restore-frames
2458 Use the desktop library to save the state of Emacs from one session
2459 to another. Once you save the Emacs @dfn{desktop}---the buffers,
2460 their file names, major modes, buffer positions, and so on---then
2461 subsequent Emacs sessions reload the saved desktop. By default,
2462 the desktop also tries to save the frame and window configuration.
2463 To disable this, set @code{desktop-restore-frames} to @code{nil}.
2464 (See that variable's documentation for some related options
2465 that you can customize to fine-tune this behavior.)
2467 @vindex frameset-filter-alist
2468 When the desktop restores the frame and window configuration, it
2469 uses the recorded values of frame parameters, disregarding any
2470 settings for those parameters you have in your init file (@pxref{Init
2471 File}). This means that frame parameters such as fonts and faces for
2472 the restored frames will come from the desktop file, where they were
2473 saved when you exited your previous Emacs session; any settings for
2474 those parameters in your init file will be ignored. To disable this,
2475 customize the value of @code{frameset-filter-alist} to filter out the
2476 frame parameters you don't want to be restored.
2478 @findex desktop-save
2479 @vindex desktop-save-mode
2480 You can save the desktop manually with the command @kbd{M-x
2481 desktop-save}. You can also enable automatic saving of the desktop
2482 when you exit Emacs, and automatic restoration of the last saved
2483 desktop when Emacs starts: use the Customization buffer (@pxref{Easy
2484 Customization}) to set @code{desktop-save-mode} to @code{t} for future
2485 sessions, or add this line in your init file (@pxref{Init File}):
2488 (desktop-save-mode 1)
2491 @vindex desktop-auto-save-timeout
2493 When @code{desktop-save-mode} is active and the desktop file exists,
2494 Emacs auto-saves it every @code{desktop-auto-save-timeout}
2495 seconds, if that is non-@code{nil} and non-zero.
2497 @findex desktop-change-dir
2498 @findex desktop-revert
2499 @vindex desktop-path
2500 If you turn on @code{desktop-save-mode} in your init file, then when
2501 Emacs starts, it looks for a saved desktop in the current directory.
2502 (More precisely, it looks in the directories specified by
2503 @var{desktop-path}, and uses the first desktop it finds.)
2504 Thus, you can have separate saved desktops in different directories,
2505 and the starting directory determines which one Emacs reloads. You
2506 can save the current desktop and reload one saved in another directory
2507 by typing @kbd{M-x desktop-change-dir}. Typing @kbd{M-x
2508 desktop-revert} reverts to the desktop previously reloaded.
2510 Specify the option @samp{--no-desktop} on the command line when you
2511 don't want it to reload any saved desktop. This turns off
2512 @code{desktop-save-mode} for the current session. Starting Emacs with
2513 the @samp{--no-init-file} option also disables desktop reloading,
2514 since it bypasses the init file, where @code{desktop-save-mode} is
2517 @vindex desktop-restore-eager
2518 By default, all the buffers in the desktop are restored at one go.
2519 However, this may be slow if there are a lot of buffers in the
2520 desktop. You can specify the maximum number of buffers to restore
2521 immediately with the variable @code{desktop-restore-eager}; the
2522 remaining buffers are restored lazily, when Emacs is idle.
2524 @findex desktop-clear
2525 @vindex desktop-globals-to-clear
2526 @vindex desktop-clear-preserve-buffers-regexp
2527 Type @kbd{M-x desktop-clear} to empty the Emacs desktop. This kills
2528 all buffers except for internal ones, and clears the global variables
2529 listed in @code{desktop-globals-to-clear}. If you want this to
2530 preserve certain buffers, customize the variable
2531 @code{desktop-clear-preserve-buffers-regexp}, whose value is a regular
2532 expression matching the names of buffers not to kill.
2534 If you want to save minibuffer history from one session to
2535 another, use the @code{savehist} library.
2537 @node Recursive Edit
2538 @section Recursive Editing Levels
2539 @cindex recursive editing level
2540 @cindex editing level, recursive
2542 A @dfn{recursive edit} is a situation in which you are using Emacs
2543 commands to perform arbitrary editing while in the middle of another
2544 Emacs command. For example, when you type @kbd{C-r} inside of a
2545 @code{query-replace}, you enter a recursive edit in which you can change
2546 the current buffer. On exiting from the recursive edit, you go back to
2547 the @code{query-replace}. @xref{Query Replace}.
2550 @findex exit-recursive-edit
2551 @cindex exiting recursive edit
2552 @dfn{Exiting} the recursive edit means returning to the unfinished
2553 command, which continues execution. The command to exit is @kbd{C-M-c}
2554 (@code{exit-recursive-edit}).
2556 You can also @dfn{abort} the recursive edit. This is like exiting,
2557 but also quits the unfinished command immediately. Use the command
2558 @kbd{C-]} (@code{abort-recursive-edit}) to do this. @xref{Quitting}.
2560 The mode line shows you when you are in a recursive edit by displaying
2561 square brackets around the parentheses that always surround the major and
2562 minor mode names. Every window's mode line shows this in the same way,
2563 since being in a recursive edit is true of Emacs as a whole rather than
2564 any particular window or buffer.
2566 It is possible to be in recursive edits within recursive edits. For
2567 example, after typing @kbd{C-r} in a @code{query-replace}, you may type a
2568 command that enters the debugger. This begins a recursive editing level
2569 for the debugger, within the recursive editing level for @kbd{C-r}.
2570 Mode lines display a pair of square brackets for each recursive editing
2571 level currently in progress.
2573 Exiting the inner recursive edit (such as with the debugger @kbd{c}
2574 command) resumes the command running in the next level up. When that
2575 command finishes, you can then use @kbd{C-M-c} to exit another recursive
2576 editing level, and so on. Exiting applies to the innermost level only.
2577 Aborting also gets out of only one level of recursive edit; it returns
2578 immediately to the command level of the previous recursive edit. If you
2579 wish, you can then abort the next recursive editing level.
2581 Alternatively, the command @kbd{M-x top-level} aborts all levels of
2582 recursive edits, returning immediately to the top-level command
2583 reader. It also exits the minibuffer, if it is active.
2585 The text being edited inside the recursive edit need not be the same text
2586 that you were editing at top level. It depends on what the recursive edit
2587 is for. If the command that invokes the recursive edit selects a different
2588 buffer first, that is the buffer you will edit recursively. In any case,
2589 you can switch buffers within the recursive edit in the normal manner (as
2590 long as the buffer-switching keys have not been rebound). You could
2591 probably do all the rest of your editing inside the recursive edit,
2592 visiting files and all. But this could have surprising effects (such as
2593 stack overflow) from time to time. So remember to exit or abort the
2594 recursive edit when you no longer need it.
2596 In general, we try to minimize the use of recursive editing levels in
2597 GNU Emacs. This is because they constrain you to go back in a
2598 particular order---from the innermost level toward the top level. When
2599 possible, we present different activities in separate buffers so that
2600 you can switch between them as you please. Some commands switch to a
2601 new major mode which provides a command to switch back. These
2602 approaches give you more flexibility to go back to unfinished tasks in
2603 the order you choose.
2606 @c Apart from edt and viper, this is all obsolete.
2607 @c (Can't believe we were saying "most other editors" into 2014!)
2608 @c There seems no point having a node just for those, which both have
2609 @c their own manuals.
2612 @cindex emulating other editors
2613 @cindex other editors
2618 GNU Emacs can be programmed to emulate (more or less) most other
2619 editors. Standard facilities can emulate these:
2622 @item CRiSP/Brief (PC editor)
2624 @vindex crisp-override-meta-x
2625 @findex scroll-all-mode
2627 @cindex Brief emulation
2628 @cindex emulation of Brief
2630 @kbd{M-x crisp-mode} enables key bindings to emulate the CRiSP/Brief
2631 editor. Note that this rebinds @kbd{M-x} to exit Emacs unless you set
2632 the variable @code{crisp-override-meta-x}. You can also use the
2633 command @kbd{M-x scroll-all-mode} or set the variable
2634 @code{crisp-load-scroll-all} to emulate CRiSP's scroll-all feature
2635 (scrolling all windows together).
2637 @item EDT (DEC VMS editor)
2638 @findex edt-emulation-on
2639 @findex edt-emulation-off
2640 Turn on EDT emulation with @kbd{M-x edt-emulation-on}; restore normal
2641 command bindings with @kbd{M-x edt-emulation-off}.
2643 Most of the EDT emulation commands are keypad keys, and most standard
2644 Emacs key bindings are still available. The EDT emulation rebindings
2645 are done in the global keymap, so there is no problem switching
2646 buffers or major modes while in EDT emulation.
2648 @item TPU (DEC VMS editor)
2651 @kbd{M-x tpu-edt-on} turns on emulation of the TPU editor emulating EDT.
2653 @item vi (Berkeley editor)
2655 Viper is an emulator for vi. It implements several levels of
2656 emulation; level 1 is closest to vi itself, while level 5 departs
2657 somewhat from strict emulation to take advantage of the capabilities of
2658 Emacs. To invoke Viper, type @kbd{M-x viper-mode}; it will guide you
2659 the rest of the way and ask for the emulation level. @inforef{Top,
2662 @item vi (another emulator)
2664 @kbd{M-x vi-mode} enters a major mode that replaces the previously
2665 established major mode. All of the vi commands that, in real vi, enter
2666 input mode are programmed instead to return to the previous major
2667 mode. Thus, ordinary Emacs serves as vi's input mode.
2669 Because vi emulation works through major modes, it does not work
2670 to switch buffers during emulation. Return to normal Emacs first.
2672 If you plan to use vi emulation much, you probably want to bind a key
2673 to the @code{vi-mode} command.
2675 @item vi (alternate emulator)
2677 @kbd{M-x vip-mode} invokes another vi emulator, said to resemble real vi
2678 more thoroughly than @kbd{M-x vi-mode}. Input mode in this emulator
2679 is changed from ordinary Emacs so you can use @key{ESC} to go back to
2680 emulated vi command mode. To get from emulated vi command mode back to
2681 ordinary Emacs, type @kbd{C-z}.
2683 This emulation does not work through major modes, and it is possible
2684 to switch buffers in various ways within the emulator. It is not
2685 so necessary to assign a key to the command @code{vip-mode} as
2686 it is with @code{vi-mode} because terminating insert mode does
2689 @inforef{Top, VIP, vip}, for full information.
2691 @item WordStar (old wordprocessor)
2692 @findex wordstar-mode
2693 @kbd{M-x wordstar-mode} provides a major mode with WordStar-like
2700 @section Hyperlinking and Navigation Features
2702 The following subsections describe convenience features for handling
2703 URLs and other types of links occurring in Emacs buffer text.
2706 * Browse-URL:: Following URLs.
2707 * Goto Address mode:: Activating URLs.
2708 * FFAP:: Finding files etc. at point.
2712 @subsection Following URLs
2713 @cindex World Wide Web
2716 @findex browse-url-at-point
2717 @findex browse-url-at-mouse
2722 @item M-x browse-url @key{RET} @var{url} @key{RET}
2723 Load a URL into a Web browser.
2726 The Browse-URL package allows you to easily follow URLs from within
2727 Emacs. Most URLs are followed by invoking a web browser;
2728 @samp{mailto:} URLs are followed by invoking the @code{compose-mail}
2729 Emacs command to send mail to the specified address (@pxref{Sending
2732 The command @kbd{M-x browse-url} prompts for a URL, and follows it.
2733 If point is located near a plausible URL, that URL is offered as the
2734 default. The Browse-URL package also provides other commands which
2735 you might like to bind to keys, such as @code{browse-url-at-point} and
2736 @code{browse-url-at-mouse}.
2738 @vindex browse-url-mailto-function
2739 @vindex browse-url-browser-function
2740 You can customize Browse-URL's behavior via various options in the
2741 @code{browse-url} Customize group. In particular, the option
2742 @code{browse-url-mailto-function} lets you define how to follow
2743 @samp{mailto:} URLs, while @code{browse-url-browser-function} lets you
2744 define how to follow other types of URLs. For more information, view
2745 the package commentary by typing @kbd{C-h P browse-url @key{RET}}.
2747 @node Goto Address mode
2748 @subsection Activating URLs
2749 @findex goto-address-mode
2750 @cindex mode, Goto Address
2751 @cindex Goto Address mode
2752 @cindex URLs, activating
2755 @item M-x goto-address-mode
2756 Activate URLs and e-mail addresses in the current buffer.
2759 @kindex C-c RET @r{(Goto Address mode)}
2760 @findex goto-address-at-point
2761 You can make Emacs mark out URLs specially in the current buffer, by
2762 typing @kbd{M-x goto-address-mode}. When this buffer-local minor mode
2763 is enabled, it finds all the URLs in the buffer, highlights them, and
2764 turns them into clickable buttons. You can follow the URL by typing
2765 @kbd{C-c @key{RET}} (@code{goto-address-at-point}) while point is on
2766 its text; or by clicking with @kbd{mouse-2}, or by clicking
2767 @kbd{mouse-1} quickly (@pxref{Mouse References}). Following a URL is
2768 done by calling @code{browse-url} as a subroutine
2769 (@pxref{Browse-URL}).
2771 It can be useful to add @code{goto-address-mode} to mode hooks and
2772 hooks for displaying an incoming message
2773 (e.g., @code{rmail-show-message-hook} for Rmail, and
2774 @code{mh-show-mode-hook} for MH-E). This is not needed for Gnus,
2775 which has a similar feature of its own.
2778 @subsection Finding Files and URLs at Point
2779 @findex find-file-at-point
2781 @findex dired-at-point
2784 @cindex finding file at point
2786 The FFAP package replaces certain key bindings for finding files,
2787 such as @kbd{C-x C-f}, with commands that provide more sensitive
2788 defaults. These commands behave like the ordinary ones when given a
2789 prefix argument. Otherwise, they get the default file name or URL
2790 from the text around point. If what is found in the buffer has the
2791 form of a URL rather than a file name, the commands use
2792 @code{browse-url} to view it (@pxref{Browse-URL}).
2794 This feature is useful for following references in mail or news
2795 buffers, @file{README} files, @file{MANIFEST} files, and so on. For
2796 more information, view the package commentary by typing @kbd{C-h P
2799 @cindex FFAP minor mode
2801 To enable FFAP, type @kbd{M-x ffap-bindings}. This makes the
2802 following key bindings, and also installs hooks for additional FFAP
2803 functionality in Rmail, Gnus and VM article buffers.
2806 @item C-x C-f @var{filename} @key{RET}
2807 @kindex C-x C-f @r{(FFAP)}
2808 Find @var{filename}, guessing a default from text around point
2809 (@code{find-file-at-point}).
2811 @kindex C-x C-r @r{(FFAP)}
2812 @code{ffap-read-only}, analogous to @code{find-file-read-only}.
2814 @kindex C-x C-v @r{(FFAP)}
2815 @code{ffap-alternate-file}, analogous to @code{find-alternate-file}.
2816 @item C-x d @var{directory} @key{RET}
2817 @kindex C-x d @r{(FFAP)}
2818 Start Dired on @var{directory}, defaulting to the directory at
2819 point (@code{dired-at-point}).
2821 @code{ffap-list-directory}, analogous to @code{list-directory}.
2823 @kindex C-x 4 f @r{(FFAP)}
2824 @code{ffap-other-window}, analogous to @code{find-file-other-window}.
2826 @code{ffap-read-only-other-window}, analogous to
2827 @code{find-file-read-only-other-window}.
2829 @code{ffap-dired-other-window}, like @code{dired-other-window}.
2831 @kindex C-x 5 f @r{(FFAP)}
2832 @code{ffap-other-frame}, analogous to @code{find-file-other-frame}.
2834 @code{ffap-read-only-other-frame}, analogous to
2835 @code{find-file-read-only-other-frame}.
2837 @code{ffap-dired-other-frame}, analogous to @code{dired-other-frame}.
2839 Search buffer for next file name or URL, then find that file or URL.
2841 @kindex S-mouse-3 @r{(FFAP)}
2842 @code{ffap-at-mouse} finds the file guessed from text around the position
2845 @kindex C-S-mouse-3 @r{(FFAP)}
2846 Display a menu of files and URLs mentioned in current buffer, then
2847 find the one you select (@code{ffap-menu}).
2851 @section Other Amusements
2854 @findex animate-birthday-present
2856 The @code{animate} package makes text dance (e.g., @kbd{M-x
2857 animate-birthday-present}).
2863 @kbd{M-x blackbox}, @kbd{M-x mpuz} and @kbd{M-x 5x5} are puzzles.
2864 @code{blackbox} challenges you to determine the location of objects
2865 inside a box by tomography. @code{mpuz} displays a multiplication
2866 puzzle with letters standing for digits in a code that you must
2867 guess---to guess a value, type a letter and then the digit you think it
2868 stands for. The aim of @code{5x5} is to fill in all the squares.
2871 @kbd{M-x bubbles} is a game in which the object is to remove as many
2872 bubbles as you can in the smallest number of moves.
2876 @cindex cryptanalysis
2877 @kbd{M-x decipher} helps you to cryptanalyze a buffer which is
2878 encrypted in a simple monoalphabetic substitution cipher.
2880 @findex dissociated-press
2881 @kbd{M-x dissociated-press} scrambles the text in the current Emacs
2882 buffer, word by word or character by character, writing its output to
2883 a buffer named @file{*Dissociation*}. A positive argument tells it to
2884 operate character by character, and specifies the number of overlap
2885 characters. A negative argument tells it to operate word by word, and
2886 specifies the number of overlap words. Dissociated Press produces
2887 results fairly like those of a Markov chain, but is however, an
2888 independent, ignoriginal invention; it techniquitously copies several
2889 consecutive characters from the sample text between random jumps,
2890 unlike a Markov chain which would jump randomly after each word or
2891 character. Keep dissociwords out of your documentation, if you want
2892 it to be well userenced and properbose.
2895 @kbd{M-x dunnet} runs a text-based adventure game.
2899 If you want a little more personal involvement, try @kbd{M-x gomoku},
2900 which plays the game Go Moku with you.
2902 @cindex tower of Hanoi
2904 If you are a little bit bored, you can try @kbd{M-x hanoi}. If you are
2905 considerably bored, give it a numeric argument. If you are very, very
2906 bored, try an argument of 9. Sit back and watch.
2910 @kbd{M-x life} runs Conway's Game of Life cellular automaton.
2913 @cindex landmark game
2914 @kbd{M-x landmark} runs a relatively non-participatory game in which
2915 a robot attempts to maneuver towards a tree at the center of the
2916 window based on unique olfactory cues from each of the four
2919 @findex morse-region
2920 @findex unmorse-region
2923 @cindex --/---/.-./.../.
2924 @kbd{M-x morse-region} converts the text in the region to Morse
2925 code; @kbd{M-x unmorse-region} converts it back. @kbd{M-x
2926 nato-region} converts the text in the region to NATO phonetic
2927 alphabet; @kbd{M-x denato-region} converts it back.
2935 @kbd{M-x pong}, @kbd{M-x snake} and @kbd{M-x tetris} are
2936 implementations of the well-known Pong, Snake and Tetris games.
2940 @kbd{M-x solitaire} plays a game of solitaire in which you jump pegs
2944 The command @kbd{M-x zone} plays games with the display when Emacs
2949 Finally, if you find yourself frustrated, try describing your
2950 problems to the famous psychotherapist Eliza. Just do @kbd{M-x
2951 doctor}. End each input by typing @key{RET} twice.