[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3. Customization

Customization can be done in 2 ways.

3.1 Rudimentary Changes  Simple constant definitions.
3.2 Key Bindings  Enabling Emacs Keys, Rebinding keys, etc.
3.2.1 Packages that Change Keymaps  How to deal with such beasts.
3.3 Viper Specials  Special Viper commands.
3.4 Vi Macros  How to do Vi style macros.


[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.1 Rudimentary Changes

An easy way to customize Viper is to change the values of constants used in Viper. Here is the list of the constants used in Viper and their default values. The corresponding :se command is also indicated. (The symbols t and nil represent "true" and "false" in Lisp).

Viper supports both the abbreviated Vi variable names and their full names. Variable completion is done on full names only. TAB and SPC complete variable names. Typing `=' will complete the name and then will prompt for a value, if applicable. For instance, :se au SPC will complete the command to :set autoindent; :se ta SPC will complete the command and prompt further like this: :set tabstop = . However, typing :se ts SPC will produce a "No match" message because ts is an abbreviation for tabstop and Viper supports completion on full names only. However, you can still hit RET or =, which will complete the command like this: :set ts = and Viper will be waiting for you to type a value for the tabstop variable. To get the full list of Vi variables, type :se SPC TAB.

viper-auto-indent nil
:se ai (:se autoindent)
:se ai-g (:se autoindent-global)
If t, enable auto indentation. by RET, o or O command.

viper-auto-indent is a local variable. To change the value globally, use setq-default. It may be useful for certain major modes to have their own values of viper-auto-indent. This can be achieved by using setq to change the local value of this variable in the hooks to the appropriate major modes.

:se ai changes the value of viper-auto-indent in the current buffer only; :se ai-g does the same globally.

viper-electric-mode t
If not nil, auto-indentation becomes electric, which means that RET, O, and o indent cursor according to the current major mode. In the future, this variable may control additional electric features.

This is a local variable: setq changes the value of this variable in the current buffer only. Use setq-default to change the value in all buffers.

viper-case-fold-search nil
:se ic (:se ignorecase)
If not nil, search ignores cases. This can also be toggled by quickly hitting / twice.
viper-re-search nil
:se magic
If not nil, search will use regular expressions; if nil then use vanilla search. This behavior can also be toggled by quickly hitting / trice.
buffer-read-only
:se ro (:se readonly)
Set current buffer to read only. To change globally put (setq-default buffer-read-only t) in your `.emacs' file.
blink-matching-paren t
:se sm (:se showmatch)
Show matching parens by blinking cursor.
tab-width t (default setting via setq-default)
:se ts=value (:se tabstop=value)
:se ts-g=value (:se tabstop-global=value)
tab-width is a local variable that controls the width of the tab stops. To change the value globally, use setq-default; for local settings, use setq.

The command :se ts sets the tab width in the current buffer only; it has no effect on other buffers.

The command :se ts-g sets tab width globally, for all buffers where the tab is not yet set locally, including the new buffers.

Note that typing TAB normally doesn't insert the tab, since this key is usually bound to a text-formatting function, indent-for-tab-command (which facilitates programming and document writing). Instead, the tab is inserted via the command viper-insert-tab, which is bound to S-tab (shift + tab).

On some non-windowing terminals, Shift doesn't modify the TAB key, so S-tab behaves as if it were TAB. In such a case, you will have to bind viper-insert-tab to some other convenient key.

viper-shift-width 8
:se sw=value (:se shiftwidth=value)
The number of columns shifted by > and < commands.
viper-search-wrap-around t
:se ws (:se wrapscan)
If not nil, search wraps around the end/beginning of buffer.
viper-search-scroll-threshold 2
If search lands within this many lines of the window top or bottom, the window will be scrolled up or down by about 1/7-th of its size, to reveal the context. If the value is negative--don't scroll.
viper-tags-file-name "TAGS"
The name of the file used as the tag table.
viper-re-query-replace nil
If not nil, use reg-exp replace in query replace.
viper-want-ctl-h-help nil
If not nil, C-h is bound to help-command; otherwise, C-h is bound as usual in Vi.
viper-vi-style-in-minibuffer t
If not nil, Viper provides a high degree of compatibility with Vi insert mode when you type text in the minibuffer; if nil, typing in the minibuffer feels like plain Emacs.
viper-no-multiple-ESC t
If you set this to nil, you can use ESC as Meta in Vi state. Normally, this is not necessary, since graphical displays have separate Meta keys (usually on each side of the space bar). On a dumb terminal, Viper sets this variable to twice, which is almost like nil, except that double ESC beeps. This, too, lets ESC to be used as a Meta.
viper-fast-keyseq-timeout 200
Key sequences separated by this many milliseconds are treated as Vi-style keyboard macros. If the key sequence is defined as such a macro, it will be executed. Otherwise, it is processed as an ordinary sequence of typed keys.

Setting this variable too high may slow down your typing. Setting it too low may make it hard to type macros quickly enough.

viper-ex-style-motion t
Set this to nil, if you want l,h to cross lines, etc. See section 2.8 Movement and Markers, for more info.
viper-ex-style-editing t
Set this to nil, if you want C-h and DEL to not stop at the beginning of a line in Insert state, X and x to delete characters across lines in Vi command state, etc.
viper-ESC-moves-cursor-back t
It t, cursor moves back 1 character when switching from insert state to vi state. If nil, the cursor stays where it was before the switch.
viper-always t
t means: leave it to Viper to decide when a buffer must be brought up in Vi state, Insert state, or Emacs state. This heuristics works well in virtually all cases. nil means you either has to invoke viper-mode manually for each buffer (or you can add viper-mode to the appropriate major mode hooks using viper-load-hook).

This option must be set in the file `~/.viper'.

viper-custom-file-name "~/.viper"
File used for Viper-specific customization. Change this setting, if you want. Must be set in `.emacs' (not `.viper'!) before Viper is loaded. Note that you have to set it as a string inside double quotes.
viper-spell-function 'ispell-region
Function used by the command #c<move> to spell.
viper-glob-function
The value of this variable is the function symbol used to expand wildcard symbols. This is platform-dependent. The default tries to set this variable to work with most shells, MS Windows, OS/2, etc. However, if it doesn't work the way you expect, you should write your own. Use viper-glob-unix-files and viper-glob-mswindows-files in `viper-util.el' as examples.

This feature is used to expand wildcards in the Ex command :e. Note that Viper doesn't support wildcards in the :r and :w commands, because file completion is a better mechanism.

ex-cycle-other-window t
If not nil, :n and :b will cycle through files in another window, if one exists.
ex-cycle-through-non-files nil
:n does not normally cycle through buffers. Set this to get buffers also.
viper-want-emacs-keys-in-insert
This is set to nil for user levels 1 and 2 and to t for user levels 3 and 4. Users who specify level 5 are allowed to set this variable as they please (the default for this level is t). If set to nil, complete Vi compatibility is provided in Insert state. This is really not recommended, as this precludes you from using language-specific features provided by the major modes.
viper-want-emacs-keys-in-vi
This is set to nil for user level 1 and to t for user levels 2--4. At level 5, users are allowed to set this variable as they please (the default for this level is t). If set to nil, complete Vi compatibility is provided in Vi command state. Setting this to nil is really a bad idea, unless you are a novice, as this precludes the use of language-specific features provided by the major modes.
viper-keep-point-on-repeat t
If not nil, point is not moved when the user repeats the previous command by typing `.' This is very useful for doing repeated changes with the . key.
viper-repeat-from-history-key 'f12
Prefix key used to invoke the macros f12 1 and f12 2 that repeat the second-last and the third-last destructive command. Both these macros are bound (as Viper macros) to viper-repeat-from-history, which checks the second key by which it is invoked to see which of the previous commands to invoke. Viper binds f12 1 and f12 2 only, but the user can bind more in `~/.viper'. See section 3.4 Vi Macros, for how to do this.
viper-keep-point-on-undo nil
If not nil, Viper tries to not move point when undoing commands. Instead, it will briefly move the cursor to the place where change has taken place. However, if the undone piece of text is not seen in window, then point will be moved to the place where the change took place. Set it to t and see if you like it better.
viper-delete-backwards-in-replace nil
If not nil, DEL key will delete characters while moving the cursor backwards. If nil, the cursor will move backwards without deleting anything.
viper-replace-overlay-face 'viper-replace-overlay-face
On a graphical display, Viper highlights replacement regions instead of putting a `$' at the end. This variable controls the so called face used to highlight the region.

By default, viper-replace-overlay-face underlines the replacement on monochrome displays and also lays a stipple over them. On color displays, replacement regions are highlighted with color.

If you know something about Emacs faces and don't like how Viper highlights replacement regions, you can change viper-replace-overlay-face by specifying a new face. (Emacs faces are described in the Emacs Lisp reference.) On a color display, the following customization method is usually most effective:

 
(set-face-foreground viper-replace-overlay-face "DarkSlateBlue")
(set-face-background viper-replace-overlay-face "yellow")
For a complete list of colors available to you, evaluate the expression (x-defined-colors). (Type it in the buffer *scratch* and then hit the C-j key.

viper-replace-overlay-cursor-color "Red"
Cursor color when it is inside the replacement region. This has effect only on color displays and only when Emacs runs as an X application.
viper-insert-state-cursor-color nil
If set to a valid color, this will be the cursor color when Viper is in insert state.
viper-emacs-state-cursor-color nil
If set to a valid color, this will be the cursor color when Viper is in emacs state.
viper-replace-region-end-delimiter "$"
A string used to mark the end of replacement regions. It is used only on TTYs or if viper-use-replace-region-delimiters is non-nil.
viper-replace-region-start-delimiter ""
A string used to mark the beginning of replacement regions. It is used only on TTYs or if viper-use-replace-region-delimiters is non-nil.
viper-use-replace-region-delimiters
If non-nil, Viper will always use viper-replace-region-end-delimiter and viper-replace-region-start-delimiter to delimit replacement regions, even on color displays (where this is unnecessary). By default, this variable is non-nil only on TTYs or monochrome displays.
viper-allow-multiline-replace-regions t
If non-nil, multi-line text replacement regions, such as those produced by commands c55w, 3C, etc., will stay around until the user exits the replacement mode. In this variable is set to nil, Viper will emulate the standard Vi behavior, which supports only intra-line replacement regions (and multi-line replacement regions are deleted).
viper-toggle-key "\C-z"
Specifies the key used to switch from Emacs to Vi and back. Must be set in `.viper'. This variable can't be changed interactively after Viper is loaded.

In Insert state, this key acts as a temporary escape to Vi state, i.e., it will set Viper up so that the very next command will be executed as if it were typed in Vi state.

viper-buffer-search-char nil
Key used for buffer search. See section 3.3 Viper Specials, for details.
viper-surrounding-word-function 'viper-surrounding-word
The value of this variable is a function name that is used to determine what constitutes a word clicked upon by the mouse. This is used by mouse search and insert.
viper-search-face 'viper-search-face
Variable that controls how search patterns are highlighted when they are found.
viper-vi-state-hook nil
List of parameterless functions to be run just after entering the Vi command state.
viper-insert-state-hook nil
Same for Insert state. This hook is also run after entering Replace state.
viper-replace-state-hook nil
List of (parameterless) functions called just after entering Replace state (and after all viper-insert-state-hook).
viper-emacs-state-hook nil
List of (parameterless) functions called just after switching from Vi state to Emacs state.
viper-load-hook nil
List of (parameterless) functions called just after loading Viper. This is the last chance to do customization before Viper is up and running.
You can reset some of these constants in Viper with the Ex command :set (when so indicated in the table). Or you can include a line like this in your `.viper' file:
 
(setq viper-case-fold-search t)


[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.2 Key Bindings

Viper lets you define hot keys, i.e., you can associate keyboard keys such as F1, Help, PgDn, etc., with Emacs Lisp functions (that may already exist or that you will write). Each key has a "preferred form" in Emacs. For instance, the Up key's preferred form is [up], the Help key's preferred form is [help], and the Undo key has the preferred form [f14]. You can find out the preferred form of a key by typing M-x describe-key-briefly and then typing the key you want to know about.

Under the X Window System, every keyboard key emits its preferred form, so you can just type

 
(global-set-key [f11] 'calendar)                        ; L1, Stop
(global-set-key [f14] 'undo)                            ; L4, Undo

to bind L1 (a key that exists on some SUN workstations) so it will invoke the Emacs Calendar and to bind L4 so it will undo changes. However, on a dumb terminal or in an Xterm window, even the standard arrow keys may not emit the right signals for Emacs to understand. To let Emacs know about those keys, you will have to find out which key sequences they emit by typing C-q and then the key (you should switch to Emacs state first). Then you can bind those sequences to their preferred forms using input-decode-map as follows:

 
(cond ((string= (getenv "TERM") "xterm")
(define-key input-decode-map "\e[192z" [f11])    ; L1
(define-key input-decode-map "\e[195z" [f14])    ; L4, Undo

The above illustrates how to do this for Xterm. On VT100, you would have to replace "xterm" with "vt100" and also change the key sequences (the same key may emit different sequences on different types of terminals).

The above keys are global, so they are overwritten by the local maps defined by the major modes and by Viper itself. Therefore, if you wish to change a binding set by a major mode or by Viper, read this.

Viper users who wish to specify their own key bindings should be concerned only with the following three keymaps: viper-vi-global-user-map for Vi state commands, viper-insert-global-user-map for Insert state commands, and viper-emacs-global-user-map for Emacs state commands (note: customized bindings for Emacs state made to viper-emacs-global-user-map are not inherited by Insert state).

For more information on Viper keymaps, see the header of the file `viper.el'. If you wish to change a Viper binding, you can use the define-key command, to modify viper-vi-global-user-map, viper-insert-global-user-map, and viper-emacs-global-user-map, as explained below. Each of these key maps affects the corresponding Viper state. The keymap viper-insert-global-user-map also affects Viper's Replace state.

If you want to bind a key, say C-v, to the function that scrolls page down and to make 0 display information on the current buffer, putting this in `.viper' will do the trick in Vi state:

 
(define-key viper-vi-global-user-map "\C-v" 'scroll-down)
To set a key globally,
 
(define-key viper-emacs-global-user-map "\C-c m" 'smail)
(define-key viper-vi-global-user-map "0" 'viper-info-on-file)
Note, however, that this binding may be overwritten by other keymaps, since the global keymap has the lowest priority. To make sure that nothing will override a binding in Emacs state, you can write this:
 
(define-key viper-emacs-global-user-map "\C-c m" 'smail)
To customize the binding for C-h in Insert state:
 
(define-key viper-insert-global-user-map "\C-h" 'my-del-backwards-function)

Each Emacs command key calls some Lisp function. If you have enabled the Help, (see section 3.1 Rudimentary Changes) C-h k will show you the function for each specific key; C-h b will show all bindings, and C-h m will provide information on the major mode in effect. If Help is not enabled, you can still get help in Vi state by prefixing the above commands with \, e.g., \ C-h k (or you can use the Help menu in the menu bar, if Emacs runs under X).

Viper users can also change bindings on a per major mode basis. As with global bindings, this can be done separately for each of the three main Viper states. To this end, Viper provides the function viper-modify-major-mode.

To modify keys in Emacs state for my-favorite-major-mode, the user needs to create a sparse keymap, say, my-fancy-map, bind whatever keys necessary in that keymap, and put

 
(viper-modify-major-mode 'dired-mode 'emacs-state my-fancy-map)

in `~/.viper'. To do the same in Vi and Insert states, you should use vi-state and insert-state. Changes in Insert state are also in effect in Replace state. For instance, suppose that the user wants to use dd in Vi state under Dired mode to delete files, u to unmark files, etc. The following code in `~/.viper' will then do the job:

 
(setq my-dired-modifier-map (make-sparse-keymap))
(define-key my-dired-modifier-map "dd" 'dired-flag-file-deletion)
(define-key my-dired-modifier-map "u" 'dired-unmark)
(viper-modify-major-mode 'dired-mode 'vi-state my-dired-modifier-map)

A Vi purist may want to modify Emacs state under Dired mode so that k, l, etc., will move around in directory buffers, as in Vi. Although this is not recommended, as these keys are bound to useful Dired functions, the trick can be accomplished via the following code:

 
(setq my-dired-vi-purist-map (make-sparse-keymap))
(define-key my-dired-vi-purist-map "k" 'viper-previous-line)
(define-key my-dired-vi-purist-map "l" 'viper-forward-char)
(viper-modify-major-mode 'dired-mode 'emacs-state my-dired-vi-purist-map)

Yet another way to customize key bindings in a major mode is to edit the list viper-major-mode-modifier-list using the customization widget. (This variable is in the Viper-misc customization group.) The elements of this list are triples of the form: (major-mode viper-state keymap), where the keymap contains bindings that are supposed to be active in the given major mode and the given viper-state.

Effects similar to key binding changes can be achieved by defining Vi keyboard macros using the Ex commands :map and :map!. The difference is that multi-key Vi macros do not override the keys they are bound to, unless these keys are typed in quick succession. So, with macros, one can use the normal keys alongside with the macros. If per-mode modifications are needed, the user can try both ways and see which one is more convenient. See section 3.4 Vi Macros, for details.

Note: in major modes that come up in Emacs state by default, the aforesaid modifications may not take place immediately (but only after the buffer switches to some other Viper state and then back to Emacs state). To avoid this, one should add viper-change-state-to-emacs to an appropriate hook of that major mode. (Check the function viper-set-hooks in `viper.el' for examples.) However, if you did not set viper-always to nil, chances are that you won't need to perform the above procedure, because Viper will take care of most useful defaults.

Finally, Viper has a facility that lets the user define per-buffer bindings, i.e., bindings that are in effect in some specific buffers only. Unlike per-mode bindings described above, per-buffer bindings can be defined based on considerations other than the major mode. This is done via the function viper-add-local-keys, which lets one specify bindings that should be in effect in the current buffer only and for a specific Viper state. For instance,

 
(viper-add-local-keys 'vi-state '(("ZZ" . TeX-command-master)
                                 ("ZQ" . viper-save-kill-buffer)))
redefines ZZ to invoke TeX-command-master in vi-state and ZQ to save-then-kill the current buffer. These bindings take effect only in the buffer where this command is executed. The typical use of this function is to execute the above expression from within a function that is included in a hook to some major mode. For instance, the above expression could be called from a function, my-tex-init, which may be added to tex-mode-hook as follows:
 
(add-hook 'tex-mode-hook 'my-tex-init)
When TeX mode starts, the hook is executed and the above Lisp expression is evaluated. Then, the bindings for ZZ and ZQ are changed in Vi command mode for all buffers in TeX mode.

Another useful application is to bind ZZ to send-mail in the Mail mode buffers (the specifics of this depend on which mail package you are using, rmail, mh-e, vm, etc. For instance, here is how to do this for mh-e, the Emacs interface to MH:

 
(defun mh-add-vi-keys ()
  "Set up ZZ for MH-e and XMH."
  (viper-add-local-keys 'vi-state '(("ZZ" . mh-send-letter))))
(add-hook 'mh-letter-mode-hook 'mh-add-vi-keys)

You can also use viper-add-local-keys to set per buffer bindings in Insert state and Emacs state by passing as a parameter the symbols insert-state and emacs-state, respectively. As with global bindings, customized local bindings done to Emacs state are not inherited by Insert state.

On rare occasions, local keys may be added by mistake. Usually this is done indirectly, by invoking a major mode that adds local keys (e.g., shell-mode redefines RET). In such a case, exiting the wrong major mode won't rid you from unwanted local keys, since these keys are local to Viper state and the current buffer, not to the major mode. In such situations, the remedy is to type M-x viper-zap-local-keys.

So much about Viper-specific bindings. See section `Customization' in The GNU Emacs Manual, and the Emacs quick reference card for the general info on key bindings in Emacs.


[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.2.1 Packages that Change Keymaps

Viper is designed to coexist with all major and minor modes of Emacs. This means that bindings set by those modes are generally available with Viper (unless you explicitly prohibit them by setting viper-want-emacs-keys-in-vi and viper-want-emacs-keys-in-insert to nil). If viper-always is set to t (which is the default), Viper will try to bring each buffer in the Viper state that is most appropriate for that buffer. Usually, this would be the Vi state, but sometimes it could be the Insert state or the Emacs state.

Some major mode bindings will necessarily be overwritten by Viper. Indeed, in Vi state, most of the 1-character keys are used for Vi-style editing. This usually causes no problems because most packages designed for editing files typically do not bind such keys. Instead, they use key sequences that start with C-x and C-c. This is why it was so important for us to free up C-x and C-c. It is common for language-specific major modes to bind TAB and C-j (the line feed) keys to various formatting functions. This is extremely useful, but may require some getting used to for a Vi user. If you decide that this feature is not for you, you can re-bind these keys as explained earlier (see section 3. Customization).

Binding for TAB is one of the most unusual aspects of Viper for many novice users. In Emacs, TAB is used to format text and programs, and is extremely useful. For instance, hitting TAB causes the current line to be re-indented in accordance with the context. In programming, this is very important, since improper automatic indentation would immediately alert the programmer to a possible error. For instance, if a ) or a " is missing somewhere above the current line, TAB is likely to mis-indent the line.

For this reason, Viper doesn't change the standard Emacs binding of TAB, thereby sacrificing Vi compatibility (except for users at level 1). Instead, in Viper, the key S-tab (shift+ tab) is chosen to emulate Vi's TAB.

We should note that on some non-windowing terminals, Shift doesn't modify the TAB key, so S-tab behaves as if it were TAB. In such a case, you will have to bind viper-insert-tab to some other convenient key.

Some packages, notably Dired, Gnus, Info, etc., attach special meaning to common keys like SPC, x, d, v, and others. This means that Vi command state is inappropriate for working with these packages. Fortunately, these modes operate on read-only buffers and are designed not for editing files, but for special-purpose browsing, reading news, mail, etc., and Vi commands are meaningless in these situations. For this reason, Viper doesn't force Vi state on such major modes--it brings them in Emacs state. You can switch to Vi state by typing C-z if, for instance, you want to do Vi-style search in a buffer (although, usually, incremental search, which is bound to C-s, is sufficient in these situations). But you should then switch back to Emacs state if you plan to continue using these major modes productively. You can also switch to Vi temporarily, to execute just one command. This is done by typing C-c \. (In some of these modes, / and : are bound Vi-style, unless these keys perform essential duties.)

If you would like certain major modes to come up in Emacs state rather than Vi state (but Viper thinks otherwise), you should put these major modes on the viper-emacs-state-mode-list list and delete them from viper-vi-state-mode-list. Likewise, you can force Viper's Insert state on a major mode by putting it in viper-insert-state-mode-list.

It is also possible to impose Vi on some major modes, even though they may bind common keys to specialized commands. This might make sense for modes that bind only a small number of common keys. For instance, Viper subverts the Shell mode by changing the bindings for C-m and C-d using viper-add-local-keys described in the section on customization (see section 3. Customization).

In some cases, some minor modes might override certain essential bindings in Vi command state. This is not a big problem because this can happen only in the beginning, when the minor mode kicks in. Typing M-x viper-mode will correct the situation. Viper knows about several such minor modes and takes care of them, so the above trick is usually not necessary. If you find that some minor mode, e.g., nasty-mode interferes with Viper, putting the following in `.viper' should fix the problem:

 
(viper-harness-minor-mode "nasty-mode")
The argument to viper-harness-minor-mode is the name of the file for the offending minor mode with the suffixes `.el' and `.elc' removed.

It may not be always obvious which minor mode is at fault. The only guidance here is to look into the file that defines the minor mode you are suspecting, say `nasty-mode.el', and see if it has a variable called nasty-mode-map. Then check if there is a statement of the form

 
(define-key nasty-mode-map key function)
that binds the misbehaving keys. If so, use the above line to harness nasty-mode. If your suspicion is wrong, no harm is done if you harness a minor mode that doesn't need to be harnessed.

It is recommended to harness even those minor modes that don't override Viper keys, but still have their own keymaps. A general way to make a minor mode, my-mode, compatible with Viper is to have the file `my-mode.el' include the following code:

 
(when (fboundp 'viper-harness-minor-mode)
  (let ((lib (file-name-sans-extension
               (file-name-nondirectory load-file-name))))
    (viper-harness-minor-mode lib)))


[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.3 Viper Specials

Viper extends Vi with a number of useful features. This includes various search functions, histories of search strings, Ex commands, insertions, and Vi's destructive commands. In addition, Viper supports file name completion and history, completion of Ex commands and variables, and many other features. Some of these features are explained in detail elsewhere in this document. Other features are explained here.

(viper-buffer-search-enable)
viper-buffer-search-char nil
Enable buffer search. Explicit call to viper-buffer-search-enable sets viper-buffer-search-char to g. Alternatively, the user can set viper-buffer-search-char in `.viper' to a key sequence to be used for buffer search. There is no need to call viper-buffer-search-enable in that case.
viper-toggle-search-style
This function, bound to C-c /, lets one toggle case-sensitive and case-insensitive search, and also switch between plain vanilla search and search via regular expressions. Without the prefix argument, the user is asked which mode to toggle. With prefix argument 1, this toggles case-sensitivity. With prefix argument 2, regular expression/vanilla search will be toggled.

However, we found that the most convenient way to toggle these options is to bind a Vi macro to bind // to toggles case sensitivity and to /// to toggles vanilla search. Thus, quickly hitting / twice will switch Viper from case sensitive search to case-insensitive. Repeating this once again will restore the original state. Likewise, quickly hitting / three times will switch you from vanilla-style search to search via regular expressions. If you hit something other than / after the first / or if the second / doesn't follow quickly enough, then Viper will issue the usual prompt / and will wait for input, as usual in Vi. If you don't like this behavior, you can "unrecord" these macros in your `~/.viper' file. For instance, if you don't like the above feature, put this in `~/.viper':

 
(viper-set-searchstyle-toggling-macros 'undefine)

If you don't like this feature as a default, but would still like to have it in some major modes, you can do so by first unsetting it globally, as shown above, and then setting it in the desired major modes as follows:

 
(viper-set-searchstyle-toggling-macros nil 'c-mode)
(viper-set-searchstyle-toggling-macros nil 'lisp-mode)

Vi-isms in Emacs state
Some people find it useful to use the Vi-style search key, `/', to invoke search in modes which Viper leaves in emacs-state. These modes are: dired-mode, mh-folder-mode, Info-mode, and Buffer-menu-mode (more may be added in the future). So, in the above modes, Viper binds `/' so that it will behave Vi-style. Furthermore, in those major modes, Viper binds `:' to invoke ex-style commands, like in vi-state. And, as described above, `//' and `///' get bound to Vi-style macros that toggle case-insensitivity and regexp-search.

If you don't like these features--which I don't really understand--you can unbind `/' and `:' in viper-dired-modifier-map (for Dired) or in viper-slash-and-colon-map, for other modes.

To unbind the macros `//' and `///' for a major mode where you feel they are undesirable, execute viper-set-emacs-state-searchstyle-macros with a non-nil argument. This can be done either interactively, by supplying a prefix argument, or by placing

 
(viper-set-emacs-state-searchstyle-macros 'undefine)
in the hook to the major mode (e.g., dired-mode-hook). See section 3.4 Vi Macros, for more information on Vi macros.

viper-heading-start
viper-heading-end
Regular Expressions for [[ and ]]. Note that Emacs defines Regexps for paragraphs and sentences. See section `Paragraphs and Sentences' in The GNU Emacs Manual, for details.
M-x viper-set-expert-level
Change your user level interactively.
viper-smart-suffix-list '("" "tex" "c" "cc" "el" "p")
Viper supports Emacs-style file completion when it prompts the user for a file name. However, in many cases, the same directory may contain files with identical prefix but different suffixes, e.g., prog.c, prog.o, paper.tex, paper.dvi. In such cases, completion will stop at the `.'. If the above variable is a list of strings representing suffixes, Viper will try these suffixes in the order listed and will check if the corresponding file exists.

For instance, if completion stopped at `paper.' and the user typed RET, then Viper will check if the files `paper.', `paper.tex', `paper.c', etc., exist. It will take the first such file. If no file exists, Viper will give a chance to complete the file name by typing the appropriate suffix. If `paper.' was the intended file name, hitting return will accept it.

To turn this feature off, set the above variable to nil.

viper-insertion-ring-size 14
Viper remembers what was previously inserted in Insert and Replace states. Several such recent insertions are kept in a special ring of strings of size viper-insertion-ring-size. If you enter Insert or Replace state you can reinsert strings from this ring by typing C-c M-p or C-c M-n. The former will search the ring in the direction of older insertions, and the latter will search in the direction of newer insertions. Hitting C-c M-p or C-c M-n in succession will undo the previous insertion from the ring and insert the next item on the ring. If a larger ring size is needed, change the value of the above variable in the `~/.viper' file.

Since typing these sequences of keys may be tedious, it is suggested that the user should bind a function key, such as f31, as follows:

 
(define-key viper-insert-global-user-map [f31]
            'viper-insert-prev-from-insertion-ring)
This binds f31 (which is usually R11 on a Sun workstation) to the function that inserts the previous string in the insertion history. To rotate the history in the opposite direction, you can either bind an unused key to viper-insert-next-from-insertion-ring or hit any digit (1 to 9) then f31.

One should not bind the above functions to M-p or M-n, since this will interfere with the minibuffer histories and, possibly, other major modes.

viper-command-ring-size 14
Viper keeps track of the recent history of destructive commands, such as dw, i, etc. In Vi state, the most recent command can be re-executed by hitting `.', as in Vi. However, repeated typing C-c M-p will cause Viper to show the previous destructive commands in the minibuffer. Subsequent hitting `.' will execute the command that was displayed last. The key C-c M-n will cycle through the command history in the opposite direction. Since typing C-c M-p may be tedious, it is more convenient to bind an appropriate function to an unused function key on the keyboard and use that key. For instance, the following
 
(define-key viper-vi-global-user-map [f31]
            'viper-prev-destructive-command)
binds the key f31 (which is usually R11 on a Sun workstation) to the function that searches the command history in the direction of older commands. To search in the opposite direction, you can either bind an unused key to viper-next-destructive-command or hit any digit (1 to 9) then f31.

One should not bind the above functions to M-p or M-n, since this will interfere with the minibuffer histories and, possibly, other major modes.

viper-minibuffer-vi-face 'viper-minibuffer-vi-face
viper-minibuffer-insert-face 'viper-minibuffer-insert-face
viper-minibuffer-emacs-face 'viper-minibuffer-emacs-face
These faces control the appearance of the minibuffer text in the corresponding Viper states. You can change the appearance of these faces through Emacs' customization widget, which is accessible through the menubar.

Viper is located in this widget under the Emulations customization subgroup of the Editing group. All Viper faces are grouped together in Viper's Highlighting customization subgroup.

Note that only the text you type in is affected by the above faces. Prompts and minibuffer messages are not affected.

Purists who do not like adornments in the minibuffer can always zap them by putting

 
(copy-face 'default 'viper-minibuffer-vi-face)
(copy-face 'default 'viper-minibuffer-insert-face)
(copy-face 'default 'viper-minibuffer-emacs-face)
in the `~/.viper' file or through the customization widget, as described above. However, in that case, the user will not have any indication of the current Viper state in the minibuffer. (This is important if the user accidentally switches to another Viper state by typing ESC or C-z).
M-x viper-go-away
Make Viper disappear from the face of your running Emacs instance. If your fingers start aching again, M-x viper-mode might save your day.
M-x toggle-viper-mode
Toggle Viperization of Emacs on and off.

Viper provides some support for multi-file documents and programs. If a document consists of several files we can designate one of them as a master and put the following at the end of that file:

 
;; Local Variables:
;; eval: (viper-setup-master-buffer "file1" "file2" "file3" "file4")
;; End:
where file1 to file4 are names of files related to the master file. Next time, when the master file is visited, the command viper-setup-master-buffer will be evaluated and the above files will be associated with the master file. Then, the new Ex command :RelatedFile (abbr. :R) will display files 1 to 4 one after another, so you can edit them. If a file is not in any Emacs buffer, it will be visited. The command PreviousRelatedFile (abbr., :P) goes through the file list in the opposite direction.

These commands are akin to :n and :N, but they allow the user to focus on relevant files only.

Note that only the master file needs to have the aforementioned block of commands. Also, ";;" above can be replaced by some other markers. Semicolon is good for Lisp programs, since it is considered a comment designator there. For LaTeX, this could be "%%%", and for C the above block should be commented out.

Even though these commands are sometimes useful, they are no substitute for the powerful tag table facility of Emacs. Viper's :tag command in a primitive interface to Emacs tags. See section `Tags' in The GNU Emacs Manual, for more information on tags.

The following two commands are normally bound to a mouse click and are part of Viper. They work only if Emacs runs as an application under X Windows (or under some other window system for which a port of GNU Emacs 20 is available). Clicking the mouse when Emacs is invoked in an Xterm window (using emacs -nw) will do no good.

viper-mouse-search-key (meta shift 1)
This variable controls the mouse-search feature of Viper. The default value states that holding Meta and Shift keys while clicking mouse button 1 should initiate search for a region under the mouse pointer (defined below). This command can take a prefix argument, which indicates the occurrence of the pattern to search for.

Note: while loading initially, Viper binds this mouse action only if it is not already bound to something else. If you want to use the mouse-search feature, and the Meta-Shift-Mouse-1 mouse action is already bound to something else, you can rebind the mouse-search feature by setting viper-mouse-search-key to something else in your ~/.viper file:

 
(setq viper-mouse-search-key '(meta 1))
This would bind mouse search to the action invoked by pressing the Meta key and clicking mouse button 1. The allowed values of viper-mouse-search-key are lists that contain a mouse-button number (1,2, or 3) and any combination of the words `control', `meta', and `shift'.

If the requested mouse action (e.g., (meta 1)) is already taken for other purposes then you have to confirm your intention by placing the following command in ~/.viper after setting viper-mouse-search-key:

 
(viper-bind-mouse-search-key 'force)

You can also change this setting interactively, through the customization widget of Emacs (type :customize).

The region that is chosen as a pattern to search for is determined as follows. If search is invoked via a single click, Viper chooses the region that lies between the beginning of the "word" under the pointer ("word" is understood in Vi sense) and the end of that word. The only difference with Vi's words is that in Lisp major modes `-' is considered an alphanumeric symbol. This is done for the convenience of working with Lisp symbols, which often have an `-' in them. Also, if you click on a non-alphanumeric character that is not a word separator (in Vi sense) then this character will also be considered alphanumeric, provided that it is adjacent (from either side) to an alphanumeric character. This useful feature gives added control over the patterns selected by the mouse click.

On a double-click, the region is determined by the beginning of the current Vi's "Word" (i.e., the largest non-separator chunk of text) and the End of that "Word" (as determined by the E command).

On a triple-click, the region consists of the entire line where the click occurred with all leading and trailing spaces and tabs removed.

viper-mouse-insert-key (meta shift 2)
This variable controls the mouse-insert feature of Viper. The above default value states that holding Meta and Shift keys while clicking mouse button 2 should insert the region surrounding the mouse pointer. The rules defining this region are the same as for mouse-search. This command takes an optional prefix argument, which indicates how many such regions to snarf from the buffer and insert. (In case of a triple-click, the prefix argument is ignored.)

Note: while loading initially, Viper binds this mouse action only if it not already bound to something else. If you want to use this feature and the default mouse action is already bound, you can rebind mouse-insert by placing this command in ~/.viper:

 
(setq viper-mouse-insert-key '(meta 2))
If you want to bind mouse-insert to an action even if this action is already taken for other purposes in Emacs, then you should add this command to ~/.viper, after setting viper-mouse-insert-key:
 
(viper-bind-mouse-insert-key 'force)

This value can also be changed via the Emacs customization widget at the menubar.

viper-multiclick-timeout
This variable controls the rate at which double-clicking must occur for the purpose of mouse search and mouse insert. By default, this is set to double-click-time in Emacs and to mouse-track-multi-click-time milliseconds in XEmacs.

Note: The above functions search and insert in the selected window of the latest active frame. This means that you can click in another window or another frame and have search or insertion done in the frame and window you just left. This lets one use these functions in a multi-frame configuration. However, this may require some getting used to. For instance, if you are typing in a frame, A, and then move the mouse to frame B and click to invoke mouse search, search (or insertion) will be performed in frame A. To perform search/insertion in frame B, you will first have to shift focus there, which doesn't happen until you type a character or perform some other action in frame B--mouse search doesn't shift focus.

If you decide that you don't like the above feature and always want search/insertion be performed in the frame where the click occurs, don't bind (and unbind, if necessary) viper-mouse-catch-frame-switch from the mouse event it is bound to.

Mouse search is integrated with Vi-style search, so you can repeat it with n and N. It should be also noted that, while case-sensitivity of search in Viper is controlled by the variable viper-case-fold-search, the case of mouse search is controlled by the Emacs variable case-fold-search, which may be set differently from viper-case-fold-search. Therefore, case-sensitivity of mouse search may be different from that of the usual Vi-style search.

Finally, if the way Viper determines the word to be searched for or to be inserted is not what you want, there is a variable, viper-surrounding-word-function, which can be changed to indicate another function for snarfing words out of the buffer. The catch is that you will then have to write such a function and make it known to your Emacs. The function viper-surrounding-word in `viper.el' can be used as a guiding example.


[ < ] [ > ]   [ << ] [ Up ] [ >> ]         [Top] [Contents] [Index] [ ? ]

3.4 Vi Macros

Viper supports much enhanced Vi-style macros and also facilitates the use of Emacs-style macros. To define a temporary macro, it is generally more convenient to use Emacs keyboard macro facility. Emacs keyboard macros are usually defined anonymously, and the latest macro can be executed by typing C-x e (or *, if Viper is in Vi state). If you need to use several temporary macros, Viper lets you save them to a register (a lowercase letter); such macros can then be executed by typing @a in Vi state (if a macro was previously saved in register a). See section 2.4 Macros and Registers, for details.

If, however, you need to use a macro regularly, it must be given a permanent name and saved. Emacs manual explains how to do this, but invocation of named Emacs macros is quite different from Vi's. First, invocation of permanent Emacs macros takes time because it requires typing too many keys (to a Vi user's taste, anyway). Second, binding such macros to function keys, for fast access, hogs valuable real estate on the keyboard.

Vi-style macros are better in that respect, since Vi lets the user overload the meaning of key sequences: keys typed in fast succession are treated specially, if this key sequence is bound to a macro.

Viper provides Vi-style keyboard macros through the usual Ex commands, :map and :map!. These macros are much more powerful in Viper than they are in the original Vi and in other emulators. This is because Viper implements an enhanced vi-style interface to the powerful Emacs keyboard macro facility.

First, any Emacs command can be executed while defining a macro, not just the Vi commands. In particular, the user can invoke Emacs commands via M-x command-name or by pressing various function keys on the keyboard. One can even use the mouse, although this is usually not useful and is not recommended (and macros defined with the use of the mouse cannot be saved in command history and in the startup file, for future use).

Macros defined by mixing Vi and Emacs commands are represented as vectors. So, don't be confused when you see one (usually through the history of Ex commands). For instance, if gg is defined by typing l, the up-arrow key and M-x next-line, its definition will look as follows in Emacs (in XEmacs, it looks slightly different, see below):

 
[l up (meta x) n e x t - l i n e return]

Second, Viper macros are defined in a WYSIWYG style. This means that commands are executed as you type them, so you can see precisely what is being defined. Third, macros can be bound to arbitrary sequences of keys, not just to printable keys. For instance, one can define a macro that will be invoked by hitting f3 then f2 function keys. (The keys delete and backspace are excluded; also, a macro invocation sequence can't start with ESC. Some other keys, such as f1 and help, can't be bound to macros under Emacs, since they are bound in key-translation-map, which overrides any other binding the user gives to keys. In general, keys that have a binding in key-translation-map can't be bound to a macro.)

Fourth, in Viper, one can define macros that are specific to a given buffer, a given major mode, or macros that are defined for all buffers. In fact, the same macro name can have several different definitions: one global, several definitions for various major modes, and definitions for various specific buffers. Buffer-specific definitions override mode-specific definitions, which, in turn, override global definitions.

As if all that is not enough, Viper (through its interface to Emacs macros) lets the user define keyboard macros that ask for confirmation or even prompt the user for input and then continue. To do this, one should type C-x q (for confirmation) or C-u C-x q (for prompt). For details, see section `Customization' in The GNU Emacs Manual

When the user finishes defining a macro (which is done by typing C-x) --- a departure from Vi), you will be asked whether you want this macro to be global, mode-specific, or buffer-specific. You will also be given a chance to save the macro in your `~/.viper' file. This is the easiest way to save a macro and make it permanently available. If you work your startup files with bare hands, here is how Viper saves the above macro so that it will be available in Viper's Insert state (and Replace state) in buffer my-buf only:

 
(viper-record-kbd-macro "gg" 'insert-state
       [l up (meta x) n e x t - l i n e return]
       "my-buf")

To do the same for Vi state and all buffers with the major mode cc-mode, use:

 
(viper-record-kbd-macro "gg" 'vi-state
       [l up (meta x) n e x t - l i n e return]
       'cc-mode)

Both macro names and macro definitions are vectors of symbols that denote keys on the keyboard. Some keys, like \, , or digit-keys must be escaped with a backslash. Modified keys are represented as lists. For instance, holding Meta and Control and pressing f4 is represented as (control meta f4). If all members of a vectors are printable characters (or sequences, such as \e, \t, for ESC and TAB), then they can also be represented as strings:

 
(viper-record-kbd-macro "aa" 'vi-state  "aaa\e"  "my-buffer")

Thus, typing aa fast in Vi state will switch Viper to Insert state (due to the first a), insert aa, and then it will switch back to Vi state. All this will take effect only in the buffer named my-buffer.

Note that the last argument to viper-record-kbd-macro must be either a string (a buffer name), a symbol representing a major mode, or t; the latter says that the macro is to be defined for all buffers (which is how macros are defined in original Vi).

For convenience, Viper also lets you define Vi-style macros in its Emacs state. There is no Ex command, like :map and :map! for doing this, but the user can include such a macro in the `~/.viper' file. The only thing is that the viper-record-kbd-macro command should specify emacs-state instead of vi-state or insert-state.

The user can get rid of a macro either by using the Ex commands :unmap and :unmap! or by issuing a call to viper-unrecord-kbd-macro. The latter is more powerful, since it can delete macros even in emacs-state. However, viper-unrecord-kbd-macro is usually needed only when the user needs to get rid of the macros that are already predefined in Viper. The syntax is:

 
(viper-unrecord-kbd-macro macro state)
The second argument must be vi-state, insert-state, or emacs-state. The first argument is a name of a macro. To avoid mistakes in specifying names of existing macros, type M-x viper-describe-kbd-macros and use a name from the list displayed by this command.

If an error occurs during macro definition, Emacs aborts the process, and it must be repeated. This is analogous to Vi, except that in Vi the user doesn't know there is an error until the macro is actually run. All that means that in order for a definition to be successful, the user must do some simple planning of the process in advance, to avoid errors. For instance, if you want to map gg to llll in Vi state, you must make sure that there is enough room on the current line. Since l moves the cursor forward, it may signal an error on reaching the end of line, which will abort the definition.

These precautions are necessary only when defining macros; they will help avoid the need to redo the job. When macros are actually run, an error during the execution will simply terminate the current execution (but the macro will remain mapped).

A macro name can be a string of characters or a vector of keys. The latter makes it possible to define macros bound to, say, double-hits on a function key, such as up or f13. This is very useful if you run out of function keys on your keyboard; it makes Viper macro facility a keyboard doubler, so to speak.

Elsewhere (See section 3.2 Key Bindings, for details), we review the standard Emacs mechanism for binding function keys to commands. For instance,

 
(global-set-key [f13] 'repeat-complex-command)

binds the key f13 to the Emacs function that repeats the last minibuffer command. Under Viper, however, you may still use this key for additional purposes, if you bind, say, a double-hitting action for that key to some other function. Emacs doesn't allow the user to do that, but Viper does this through its keyboard macro facility. To do this, type :map first. When you are asked to enter a macro name, hit f13 twice, followed by RET or SPC.

Emacs will now start the mapping process by actually executing Vi and Emacs commands, so that you could see what will happen each time the macro is executed. Suppose now we wanted to bind the key sequence f13 f13 to the command eval-last-sexp. To accomplish this, we can type M-x eval-last-sexp followed by C-x ). If you answer positively to Viper's offer to save this macro in `~/.viper' for future uses, the following will be inserted in that file:

 
(viper-record-kbd-macro [f16 f16] 'vi-state
         [(meta x) e v a l - l a s t - s e x p]
         'lisp-interaction-mode)

To illustrate the above point, Viper provides two canned macros, which, by default, are bound to [f12 \1] and [f12 \2] (invoked by typing f12 then 1 and 2, respectively). These macros are useful shortcuts to Viper's command ring history. The first macro will execute the second-last destructive command (the last one is executed by ., as usual). The second macro executes the third-last command.

If you need to go deeper into the command history, you will have to use other commands, as described earlier in this section; or you can bind, say, f12 \3 like this:

 
(viper-record-kbd-macro [f12 \3] 'vi-state
                      [(meta x) r e p e a t - f r o m - h i s t o r y]
                      t)

Note that even though the macro uses the function key f12, the key is actually free and can still be bound to some Emacs function via define-key or global-set-key.

Viper allows the user to define macro names that are prefixes of other macros. For instance, one can define [[ and [[[[ to be macros. If you type the exact sequence of such keys and then pause, Viper will execute the right macro. However, if you don't pause and, say, type [[[[text then the conflict is resolved as follows. If only one of the key sequences, [[ or [[[[ has a definition applicable to the current buffer, then, in fact, there is no conflict and the right macro will be chosen. If both have applicable definitions, then the first one found will be executed. Usually this is the macro with a shorter name. So, in our case, [[[[text will cause the macro [[ to be executed twice and then the remaining keys, t e x t, will be processed.

When defining macros using :map or :map!, the user enters the actually keys to be used to invoke the macro. For instance, you should hit the actual key f6 if it is to be part of a macro name; you do not write f 6. When entering keys, Viper displays them as strings or vectors (e.g., "abc" or [f6 f7 a]). The same holds for unmapping. Hitting TAB while typing a macro name in the :unmap or :unmap! command will cause name completion. Completions are displayed as strings or vectors. However, as before, you don't actually type `"', `[', or `]' that appear in the completions. These are meta-symbols that indicate whether the corresponding macro name is a vector or a string.

One last difference from Vi: Vi-style keyboard macros cannot be defined in terms of other Vi-style keyboard macros (but named Emacs macros are OK). More precisely, while defining or executing a macro, the special meaning of key sequences (as Vi macros) is ignored. This is because it is all too easy to create an infinite loop in this way. Since Viper macros are much more powerful than Vi's it is impossible to detect such loops. In practice, this is not really a limitation but, rather, a feature.

We should also note that Vi macros are disabled in the minibuffer, which helps keep some potential troubles away.

The rate at which the user must type keys in order for them to be recognized as a timeout macro is controlled by the variable viper-fast-keyseq-timeout, which defaults to 200 milliseconds.

For the most part, Viper macros defined in `~/.viper' can be shared between Emacs, XEmacs, and X and TTY modes. However, macros defined via function keys may need separate definitions when XEmacs and Emacs have different names for the same keyboard key. For instance, the `Page Up' key may be known in Emacs as prior and in XEmacs as pgup. The problem with TTY may be that the function keys there generate sequences of events instead of a single event (as under a window system). Emacs maps some of these sequences back to the logical keys (e.g., the sequences generated by the arrow keys are mapped to up, left, etc.). However, not all function keys are mapped in this way. Macros that are bound to key sequences that contain such unmapped function keys have to be redefined for TTY's (and possibly for every type of TTY you may be using). To do this, start Emacs on an appropriate TTY device and define the macro using :map, as usual.

Finally, Viper provides a function that conveniently displays all macros currently defined. To see all macros along with their definitions, type M-x viper-describe-kbd-macros.


[ << ] [ >> ]           [Top] [Contents] [Index] [ ? ]

This document was generated by XEmacs shared group account on December, 19 2009 using texi2html 1.65.