The Zsh source distribution includes a number of items contributed by the user community. These are not inherently a part of the shell, and some may not be available in every zsh installation. The most significant of these are documented here. For documentation on other contributed items such as shell functions, look for comments in the function source files.
The key sequence ESC h is normally bound by ZLE to execute the run-help widget (see zshzle(1)). This invokes the run-help command with the command word from the current input line as its argument. By default, run-help is an alias for the man command, so this often fails when the command word is a shell builtin or a user-defined function. By redefining the run-help alias, one can improve the on-line help provided by the shell.
The helpfiles utility, found in the Util directory of the distribution, is a Perl program that can be used to process the zsh manual to produce a separate help file for each shell builtin and for many other shell features as well. The autoloadable run-help function, found in Functions/Misc, searches for these helpfiles and performs several other tests to produce the most complete help possible for the command.
There may already be a directory of help files on your system; look in /usr/share/zsh or /usr/local/share/zsh and subdirectories below those, or ask your system administrator.
To create your own help files with helpfiles, choose or create a directory where the individual command help files will reside. For example, you might choose ~/zsh_help. If you unpacked the zsh distribution in your home directory, you would use the commands:
mkdir ~/zsh_help cd ~/zsh_help man zshall | colcrt - | \ perl ~/zsh-4.3.10/Util/helpfiles
Next, to use the run-help function, you need to add lines something like the following to your .zshrc or equivalent startup file:
unalias run-help autoload run-help HELPDIR=~/zsh_help
The HELPDIR parameter tells run-help where to look for the help files. If your system already has a help file directory installed, set HELPDIR to the path of that directory instead.
Note that in order for `autoload run-help' to work, the run-help file must be in one of the directories named in your fpath array (see zshparam(1)). This should already be the case if you have a standard zsh installation; if it is not, copy Functions/Misc/run-help to an appropriate directory.
If you frequently edit your zsh functions, or periodically update your zsh installation to track the latest developments, you may find that function digests compiled with the zcompile builtin are frequently out of date with respect to the function source files. This is not usually a problem, because zsh always looks for the newest file when loading a function, but it may cause slower shell startup and function loading. Also, if a digest file is explicitly used as an element of fpath, zsh won't check whether any of its source files has changed.
The zrecompile autoloadable function, found in Functions/Misc, can be used to keep function digests up to date.
In the first form, each name is the name of a compiled file or a directory containing *.zwc files that should be checked. If no arguments are given, the directories and *.zwc files in fpath are used.
When -t is given, no compilation is performed, but a return status of zero (true) is set if there are files that need to be re-compiled and non-zero (false) otherwise. The -q option quiets the chatty output that describes what zrecompile is doing.
Without the -t option, the return status is zero if all files that needed re-compilation could be compiled and non-zero if compilation for at least one of the files failed.
If the -p option is given, the args are interpreted as one or more sets of arguments for zcompile, separated by `--'. For example:
zrecompile -p \ -R ~/.zshrc -- \ -M ~/.zcompdump -- \ ~/zsh/comp.zwc ~/zsh/Completion/*/_*
This compiles ~/.zshrc into ~/.zshrc.zwc if that doesn't exist or if it is older than ~/.zshrc. The compiled file will be marked for reading instead of mapping. The same is done for ~/.zcompdump and ~/.zcompdump.zwc, but this compiled file is marked for mapping. The last line re-creates the file ~/zsh/comp.zwc if any of the files matching the given pattern is newer than it.
Without the -p option, zrecompile does not create function digests that do not already exist, nor does it add new functions to the digest.
The following shell loop is an example of a method for creating function digests for all functions in your fpath, assuming that you have write permission to the directories:
for ((i=1; i <= $#fpath; ++i)); do dir=$fpath[i] zwc=${dir:t}.zwc if [[ $dir == (.|..) || $dir == (.|..)/* ]]; then continue fi files=($dir/*(N-.)) if [[ -w $dir:h && -n $files ]]; then files=(${${(M)files%/*/*}#/}) if ( cd $dir:h && zrecompile -p -U -z $zwc $files ); then fpath[i]=$fpath[i].zwc fi fi done
The -U and -z options are appropriate for functions in the default zsh installation fpath; you may need to use different options for your personal function directories.
Once the digests have been created and your fpath modified to refer to them, you can keep them up to date by running zrecompile with no arguments.
The large number of possible combinations of keyboards, workstations, terminals, emulators, and window systems makes it impossible for zsh to have built-in key bindings for every situation. The zkbd utility, found in Functions/Misc, can help you quickly create key bindings for your configuration.
Run zkbd either as an autoloaded function, or as a shell script:
zsh -f ~/zsh-4.3.10/Functions/Misc/zkbd
When you run zkbd, it first asks you to enter your terminal type; if the default it offers is correct, just press return. It then asks you to press a number of different keys to determine characteristics of your keyboard and terminal; zkbd warns you if it finds anything out of the ordinary, such as a Delete key that sends neither ^H nor ^?.
The keystrokes read by zkbd are recorded as a definition for an associative array named key, written to a file in the subdirectory .zkbd within either your HOME or ZDOTDIR directory. The name of the file is composed from the TERM, VENDOR and OSTYPE parameters, joined by hyphens.
You may read this file into your .zshrc or another startup file with the `source' or `.' commands, then reference the key parameter in bindkey commands, like this:
source ${ZDOTDIR:-$HOME}/.zkbd/$TERM-$VENDOR-$OSTYPE [[ -n ${key[Left]} ]] && bindkey "${key[Left]}" backward-char [[ -n ${key[Right]} ]] && bindkey "${key[Right]}" forward-char # etc.
Note that in order for `autoload zkbd' to work, the zkdb file must be in one of the directories named in your fpath array (see zshparam(1)). This should already be the case if you have a standard zsh installation; if it is not, copy Functions/Misc/zkbd to an appropriate directory.
Occasionally you may encounter what appears to be a bug in the shell, particularly if you are using a beta version of zsh or a development release. Usually it is sufficient to send a description of the problem to one of the zsh mailing lists (see zsh(1)), but sometimes one of the zsh developers will need to recreate your environment in order to track the problem down.
The script named reporter, found in the Util directory of the distribution, is provided for this purpose. (It is also possible to autoload reporter, but reporter is not installed in fpath by default.) This script outputs a detailed dump of the shell state, in the form of another script that can be read with `zsh -f' to recreate that state.
To use reporter, read the script into your shell with the `.' command and redirect the output into a file:
. ~/zsh-4.3.10/Util/reporter > zsh.report
You should check the zsh.report file for any sensitive information such as passwords and delete them by hand before sending the script to the developers. Also, as the output can be voluminous, it's best to wait for the developers to ask for this information before sending it.
You can also use reporter to dump only a subset of the shell state. This is sometimes useful for creating startup files for the first time. Most of the output from reporter is far more detailed than usually is necessary for a startup file, but the aliases, options, and zstyles states may be useful because they include only changes from the defaults. The bindings state may be useful if you have created any of your own keymaps, because reporter arranges to dump the keymap creation commands as well as the bindings for every keymap.
As is usual with automated tools, if you create a startup file with reporter, you should edit the results to remove unnecessary commands. Note that if you're using the new completion system, you should not dump the functions state to your startup files with reporter; use the compdump function instead (see zshcompsys(1)).
If the state is omitted, all is assumed.
With the exception of `all', every state can be abbreviated by any prefix, even a single letter; thus a is the same as aliases, z is the same as zstyles, etc.
hook is one of chpwd, periodic, precmd or preexec, the special functions in question.
functions is name of an ordinary shell function. If no options are given this will be added to the array of functions to be executed. in the given context.
If the option -d is given, the function is removed from the array of functions to be executed.
If the option -D is given, the function is treated as a pattern and any matching names of functions are removed from the array of functions to be executed.
In a lot of cases, it is nice to automatically retrieve information from version control systems (VCSs), such as subversion, CVS or git, to be able to provide it to the user; possibly in the user's prompt. So that you can instantly tell on which branch you are currently on, for example.
In order to do that, you may use the vcs_info function.
The following VCSs are supported, showing the abbreviated name by which they are referred to within the system:
To load vcs_info:
autoload -Uz vcs_info
It can be used in any existing prompt, because it does not require any $psvar entries to be left available.
To get this feature working quickly (including colors), you can do the following (assuming, you loaded vcs_info properly - see above):
zstyle ':vcs_info:*' actionformats \ '%F{5}(%f%s%F{5})%F{3}-%F{5}[%F{2}%b%F{3}|%F{1}%a%F{5}]%f ' zstyle ':vcs_info:*' formats \ '%F{5}(%f%s%F{5})%F{3}-%F{5}[%F{2}%b%F{5}]%f ' zstyle ':vcs_info:(sv[nk]|bzr):*' branchformat '%b%F{1}:%F{3}%r' precmd () { vcs_info } PS1='%F{5}[%F{2}%n%F{5}] %F{3}%3~ ${vcs_info_msg_0_}%f%# '
Obviously, the last two lines are there for demonstration: You need to call vcs_info from your precmd function. Once that is done you need a single quoted '${vcs_info_msg_0_}' in your prompt.
To be able to use '${vcs_info_msg_0_}' directly in your prompt like this, you will need to have the PROMPT_SUBST option enabled.
Now call the vcs_info_printsys utility from the command line:
% vcs_info_printsys ## list of supported version control backends: ## disabled systems are prefixed by a hash sign (#) bzr cdv cvs darcs git hg mtn p4 svk svn tla ## flavours (cannot be used in the enable or disable styles; they ## are enabled and disabled with their master [git-svn -> git]) ## they *can* be used contexts: ':vcs_info:git-svn:*'. git-p4 git-svn
You may not want all of these because there is no point in running the code to detect systems you do not use. So there is a way to disable some backends altogether:
zstyle ':vcs_info:*' disable bzr cdv darcs mtn svk tla
You may also pick a few from that list and enable only those:
zstyle ':vcs_info:*' enable git cvs svn
If you rerun vcs_info_printsys after one of these commands, you will see the backends listed in the disable style (or backends not in the enable style - if you used that) marked as disabled by a hash sign. That means the detection of these systems is skipped completely. No wasted time there.
The vcs_info feature can be configured via zstyle.
First, the context in which we are working:
:vcs_info:<vcs-string>:<user-context>:<repo-root-name>
There are three special values for <vcs-string>: The first is named -init-, that is in effect as long as there was no decision what vcs backend to use. The second is -preinit-; it is used before vcs_info is run, when initializing the data exporting variables. The third special value is formats and is used by the vcs_info_lastmsg for looking up its styles.
The initial value of <repo-root-name> is -all- and it is replaced with the actual name, as soon as it is known. Only use this part of the context for defining the formats, actionformats or branchformat styles. As it is guaranteed that <repo-root-name> is set up correctly for these only. For all other styles, just use '*' instead.
There are two pre-defined values for <user-context>:
You can of course use ':vcs_info:*' to match all VCSs in all user-contexts at once.
This is a description of all styles that are looked up.
Say, ~/.zsh is a directory under version control, in which you do not want vcs_info to be active, do:
zstyle ':vcs_info:*' disable-patterns "$HOME/.zsh(|/*)"
Note, that the actions taken if this style is enabled are potentially expensive (read: they take time, depending on how big the current repository is). Therefore, it is disabled by default.
If the value of this style is empty (which is the default), the used binary name is the name of the backend in use (e.g. svn is used in a svn repository).
The repo-root-name part in the context is always the default -all- when this style is looked up.
For example, this style can be used to use binaries from non-default installation directories. Assume, git is installed in /usr/bin, but your sysadmin installed a newer version in /usr/bin/local. Now, instead of changing the order of your $PATH parameter, you can do this:
zstyle ':vcs_info:git:*:-all-' command /usr/local/bin/git
The default values for these styles in all contexts are:
In normal formats and actionformats, the following replacements are done:
In branchformat these replacements are done:
In stgitformat these replacements are done:
Not all vcs backends have to support all replacements. For nvcsformats no replacements are performed at all. It is just a string.
If you want to use the %b (bold off) prompt expansion in formats, which expands %b itself, use %%b. That will cause the vcs_info expansion to replace %%b with %b. So zsh's prompt expansion mechanism can handle it. Similarly, to hand down %b from branchformat, use %%%%b. Sorry for this inconvenience, but it cannot be easily avoided. Luckily we do not clash with a lot of prompt expansions and this only needs to be done for those.
All functions named VCS_INFO_* are for internal use only.
All variables named VCS_INFO_* are for internal use only.
Don't use vcs_info at all (even though it's in your prompt):
zstyle ':vcs_info:*' enable NONE
Disable the backends for bzr and svk:
zstyle ':vcs_info:*' disable bzr svk
Disable everything but bzr and svk:
zstyle ':vcs_info:*' enable bzr svk
Provide a special formats for git:
zstyle ':vcs_info:git:*' formats ' GIT, BABY! [%b]' zstyle ':vcs_info:git:*' actionformats ' GIT ACTION! [%b|%a]'
Use the quicker bzr backend
zstyle ':vcs_info:bzr:*' use-simple true
If you do use use-simple, please report if it does `the-right-thing[tm]'.
Display the revision number in yellow for bzr and svn:
zstyle ':vcs_info:(svn|bzr):*' branchformat '%b%{'${fg[yellow]}'%}:%r'
If you want colors, make sure you enclose the color codes in %{...%}, if you want to use the string provided by vcs_info in prompts.
Here is how to print the vcs information as a command (not in a prompt):
alias vcsi='vcs_info command; vcs_info_lastmsg'
This way, you can even define different formats for output via vcs_info_lastmsg in the ':vcs_info:formats:command:*' namespace.
You should make sure all the functions from the Functions/Prompts directory of the source distribution are available; they all begin with the string `prompt_' except for the special function`promptinit'. You also need the `colors' function from Functions/Misc. All of these functions may already have been installed on your system; if not, you will need to find them and copy them. The directory should appear as one of the elements of the fpath array (this should already be the case if they were installed), and at least the function promptinit should be autoloaded; it will autoload the rest. Finally, to initialize the use of the system you need to call the promptinit function. The following code in your .zshrc will arrange for this; assume the functions are stored in the directory ~/myfns:
fpath=(~/myfns $fpath) autoload -U promptinit promptinit
Use the prompt command to select your preferred theme. This command may be added to your .zshrc following the call to promptinit in order to start zsh with a theme already selected.
In some cases the theme may be modified by one or more arguments, which should be given after the theme name. See the help for each theme for descriptions of these arguments.
Options are:
These functions all implement user-defined ZLE widgets (see zshzle(1)) which can be bound to keystrokes in interactive shells. To use them, your .zshrc should contain lines of the form
autoload function zle -N function
followed by an appropriate bindkey command to associate the function with a key sequence. Suggested bindings are described below.
autoload -U select-word-style select-word-style bash
The simplest way of configuring the functions is to use select-word-style, which can either be called as a normal function with the appropriate argument, or invoked as a user-defined widget that will prompt for the first character of the word style to be used. The first time it is invoked, the eight -match functions will automatically replace the builtin versions, so they do not need to be loaded explicitly.
The word styles available are as follows. Only the first character is examined.
All but `default' can be input as an upper case character, which has the same effect but with subword matching turned on. In this case, words with upper case characters are treated specially: each separate run of upper case characters, or an upper case character followed by any number of other characters, is considered a word. The style subword-range can supply an alternative character range to the default `[:upper:]'; the value of the style is treated as the contents of a `[...]' pattern (note that the outer brackets should not be supplied, only those surrounding named ranges).
More control can be obtained using the zstyle command, as described in zshmodules(1). Each style is looked up in the context :zle:widget where widget is the name of the user-defined widget, not the name of the function implementing it, so in the case of the definitions supplied by select-word-style the appropriate contexts are :zle:forward-word, and so on. The function select-word-style itself always defines styles for the context `:zle:*' which can be overridden by more specific (longer) patterns as well as explicit contexts.
The style word-style specifies the rules to use. This may have the following values.
The first three of those rules usually use $WORDCHARS, but the value in the parameter can be overridden by the style word-chars, which works in exactly the same way as $WORDCHARS. In addition, the style word-class uses character class syntax to group characters and takes precedence over word-chars if both are set. The word-class style does not include the surrounding brackets of the character class; for example, `-:[:alnum:]' is a valid word-class to include all alphanumerics plus the characters `-' and `:'. Be careful including `]', `^' and `-' as these are special inside character classes.
word-style may also have `-subword' appended to its value to turn on subword matching, as described above.
The style skip-chars is mostly useful for transpose-words and similar functions. If set, it gives a count of characters starting at the cursor position which will not be considered part of the word and are treated as space, regardless of what they actually are. For example, if
zstyle ':zle:transpose-words' skip-chars 1
has been set, and transpose-words-match is called with the cursor on the X of fooXbar, where X can be any character, then the resulting expression is barXfoo.
Finer grained control can be obtained by setting the style word-context to an array of pairs of entries. Each pair of entries consists of a pattern and a subcontext. The shell argument the cursor is on is matched against each pattern in turn until one matches; if it does, the context is extended by a colon and the corresponding subcontext. Note that the test is made against the original word on the line, with no stripping of quotes. Special handling is done between words: the current context is examined and if it contains the string back, the word before the cursor is considered, else the word after cursor is considered. Some examples are given below.
Here are some examples of use of the styles, actually taken from the simplified interface in select-word-style:
zstyle ':zle:*' word-style standard zstyle ':zle:*' word-chars ''
Implements bash-style word handling for all widgets, i.e. only alphanumerics are word characters; equivalent to setting the parameter WORDCHARS empty for the given context.
style ':zle:*kill*' word-style space
Uses space-delimited words for widgets with the word `kill' in the name. Neither of the styles word-chars nor word-class is used in this case.
Here are some examples of use of the word-context style to extend the context.
zstyle ':zle:*' word-context "*/*" file "[[:space:]]" whitespace zstyle ':zle:transpose-words:whitespace' word-style shell zstyle ':zle:transpose-words:filename' word-style normal zstyle ':zle:transpose-words:filename' word-chars ''
This provides two different ways of using transpose-words depending on whether the cursor is on whitespace between words or on a filename, here any word containing a /. On whitespace, complete arguments as defined by standard shell rules will be transposed. In a filename, only alphanumerics will be transposed. Elsewhere, words will be transposed using the default style for :zle:transpose-words.
The word matching and all the handling of zstyle settings is actually implemented by the function match-words-by-style. This can be used to create new user-defined widgets. The calling function should set the local parameter curcontext to :zle:widget, create the local parameter matched_words and call match-words-by-style with no arguments. On return, matched_words will be set to an array with the elements: (1) the start of the line (2) the word before the cursor (3) any non-word characters between that word and the cursor (4) any non-word character at the cursor position plus any remaining non-word characters before the next word, including all characters specified by the skip-chars style, (5) the word at or following the cursor (6) any non-word characters following that word (7) the remainder of the line. Any of the elements may be an empty string; the calling function should test for this to decide whether it can perform its function.
It is possible to pass options with arguments to match-words-by-style to override the use of styles. The options are:
For example, match-words-by-style -w shell -c 0 may be used to extract the command argument around the cursor.
The word-context style is implemented by the function match-word-context. This should not usually need to be called directly.
The basic behaviour is to delete the word around the cursor. There is no numeric prefix handling; only the single word around the cursor is considered. If the widget contains the string kill, the removed text will be placed in the cutbuffer for future yanking. This can be obtained by defining kill-whole-word-match as follows:
zle -N kill-whole-word-match delete-whole-word-match
and then binding the widget kill-whole-word-match.
If insert-last-word has been used to retrieve the last word on a previous history line, repeated invocations will replace that word with earlier words from the same line.
Otherwise, the widget applies to words on the line currently being edited. The widget style can be set to the name of another widget that should be called to retrieve words. This widget must accept the same three arguments as insert-last-word.
This widget allows the cursor to be easily moved to the other interesting spots. It can be invoked repeatedly to cycle between all positions reported by the completion system.
bindkey -M vicmd v edit-command-line
Although you autoload only one function, the commands to use it are slightly different because it implements two widgets.
zle -N history-beginning-search-backward-end \ history-search-end zle -N history-beginning-search-forward-end \ history-search-end bindkey '\e^P' history-beginning-search-backward-end bindkey '\e^N' history-beginning-search-forward-end
With a prefix argument, the search is not anchored to the start of the line; the string typed by the use may appear anywhere in the line in the history.
If the widget name contains `-end' the cursor is moved to the end of the line inserted. If the widget name contains `-space' any space in the text typed is treated as a wildcard and can match anything (hence a leading space is equivalent to giving a prefix argument). Both forms can be combined, for example:
zle -N history-beginning-search-menu-space-end \ history-beginning-search-menu
The commands to create bindable widgets are similar to those in the example immediately above:
autoload -U history-pattern-search zle -N history-pattern-search-backward history-pattern-search zle -N history-pattern-search-forward history-pattern-search
bindkey '^X+' incarg
This works only with the new function based completion system.
bindkey '^Xi' incremental-complete-word
The function may optionally be followed by up to two characters which replace one or both of the characters read from the keyboard; if both characters are supplied, no input is read. For example, insert-composed-char a: can be used within a widget to insert an a with umlaut into the command line. This has the advantages over use of a literal character that it is more portable.
For best results zsh should have been built with support for multibyte characters (configured with --enable-multibyte); however, the function works for the limited range of characters available in single-byte character sets such as ISO-8859-1.
The character is converted into the local representation and inserted into the command line at the cursor position. (The conversion is done within the shell, using whatever facilities the C library provides.) With a numeric argument, the character and its code are previewed in the status line
The function may be run outside zle in which case it prints the character (together with a newline) to standard output. Input is still read from keystrokes.
See insert-unicode-char for an alternative way of inserting Unicode characters using their hexadecimal character number.
The set of accented characters is reasonably complete up to Unicode character U+0180, the set of special characters less so. However, it it is very sporadic from that point. Adding new characters is easy, however; see the function define-composed-chars. Please send any additions to zsh-workers@sunsite.dk.
The codes for the second character when used to accent the first are as follows. Note that not every character can take every accent.
The most common characters from the Arabic, Cyrillic, Greek and Hebrew alphabets are available; consult RFC 1345 for the appropriate sequences. In addition, a set of two letter codes not in RFC 1345 are available for the double-width characters corresponding to ASCII characters from ! to ~ (0x21 to 0x7e) by preceding the character with ^, for example ^A for a double-width A.
The following other two-character sequences are understood.
bindkey '^Xf' insert-files
narrow-to-region may be used as a widget or called as a function from a user-defined widget; by default, the text outside the editable area remains visible. A recursive-edit is performed and the original widening status is then restored. Various options and arguments are available when it is called as a function.
The options -p pretext and -P posttext may be used to replace the text before and after the display for the duration of the function; either or both may be an empty string.
If the option -n is also given, pretext or posttext will only be inserted if there is text before or after the region respectively which will be made invisible.
Two numeric arguments may be given which will be used instead of the cursor and mark positions.
The option -S statepm is used to narrow according to the other options while saving the original state in the parameter with name statepm, while the option -R statepm is used to restore the state from the parameter; note in both cases the name of the parameter is required. In the second case, other options and arguments are irrelevant. When this method is used, no recursive-edit is performed; the calling widget should call this function with the option -S, perform its own editing on the command line or pass control to the user via `zle recursive-edit', then call this function with the option -R. The argument statepm must be a suitable name for an ordinary parameter, except that parameters beginning with the prefix _ntr_ are reserved for use within narrow-to-region. Typically the parameter will be local to the calling function.
narrow-to-region-invisible is a simple widget which calls narrow-to-region with arguments which replace any text outside the region with `...'.
The display is restored (and the widget returns) upon any zle command which would usually cause the line to be accepted or aborted. Hence an additional such command is required to accept or abort the current line.
The return status of both widgets is zero if the line was accepted, else non-zero.
Here is a trivial example of a widget using this feature.
local state narrow-to-region -p $'Editing restricted region\n' \ -P '' -S state zle recursive-edit narrow-to-region -R state
See insert-composed-char for a way of inserting characters using a two-character mnemonic.
With the function based completion system (which is needed for this), you should be able to type TAB at almost any point to advance the cursor to the next ``interesting'' character position (usually the end of the current word, but sometimes somewhere in the middle of the word). And of course as soon as the entire line is what you want, you can accept with return, without needing to move the cursor to the end first.
The first time predict-on is used, it creates several additional widget functions:
Although you autoload only the predict-on function, it is necessary to create a keybinding for predict-off as well.
zle -N predict-on zle -N predict-off bindkey '^X^Z' predict-on bindkey '^Z' predict-off
If one argument is supplied to the function it is taken as a prompt, otherwise `? ' is used. If two arguments are supplied, they are the prompt and the initial value of $LBUFFER, and if a third argument is given it is the initial value of $RBUFFER. This provides a default value and starting cursor placement. Upon return the entire buffer is the value of $REPLY.
One option is available: `-k num' specifies that num characters are to be read instead of a whole line. The line editor is not invoked recursively in this case, so depending on the terminal settings the input may not be visible, and only the input keys are placed in $REPLY, not the entire buffer. Note that unlike the read builtin num must be given; there is no default.
The name is a slight misnomer, as in fact the shell's own minibuffer is not used. Hence it is still possible to call executed-named-cmd and similar functions while reading a value.
If the widget name contains the word `pattern', for example by defining the widget using the command `zle -N replace-pattern replace-string', then the replacement is done by pattern matching. All zsh extended globbing patterns can be used in the source string; note that unlike filename generation the pattern does not need to match an entire word, nor do glob qualifiers have any effect. In addition, the replacement string can contain parameter or command substitutions. Furthermore, a `&' in the replacement string will be replaced with the matched source string, and a backquoted digit `\N' will be replaced by the Nth parenthesised expression matched. The form `\{N}' may be used to protect the digit from following digits.
By default the previous source or replacement string will not be offered for editing. However, this feature can be activated by setting the style edit-previous in the context :zle:widget (for example, :zle:replace-string) to true. In addition, a positive numeric argument forces the previous values to be offered, a negative or zero argument forces them not to be.
The function replace-string-again can be used to repeat the previous replacement; no prompting is done. As with replace-string, if the name of the widget contains the word `pattern', pattern matching is performed, else a literal string replacement. Note that the previous source and replacement text are the same whether pattern or string matching is used.
For example, starting from the line:
print This line contains fan and fond
and invoking replace-pattern with the source string `f(?)n' and the replacement string `c\1r' produces the not very useful line:
print This line contains car and cord
The range of the replacement string can be limited by using the narrow-to-region-invisible widget. One limitation of the current version is that undo will cycle through changes to the replacement and source strings before undoing the replacement itself.
zle -N insert-last-word smart-insert-last-word
With a numeric prefix, or when passed command line arguments in a call from another widget, it behaves like insert-last-word, except that words in comments are ignored when INTERACTIVE_COMMENTS is set.
Otherwise, the rightmost ``interesting'' word from the previous command is found and inserted. The default definition of ``interesting'' is that the word contains at least one alphabetic character, slash, or backslash. This definition may be overridden by use of the match style. The context used to look up the style is the widget name, so usually the context is :insert-last-word. However, you can bind this function to different widgets to use different patterns:
zle -N insert-last-assignment smart-insert-last-word zstyle :insert-last-assignment match '[[:alpha:]][][[:alnum:]]#=*' bindkey '\e=' insert-last-assignment
If no interesting word is found and the auto-previous style is set to a true value, the search continues upward through the history. When auto-previous is unset or false (the default), the widget must be invoked repeatedly in order to search earlier history lines.
The style whence is available in the context :zle:$WIDGET; this may be set to an array to give the command and options that will be used to investigate the command word found. The default is whence -c.
These functions are useful in constructing widgets. They should be loaded with `autoload -U function' and called as indicated from user-defined widgets.
The parameter REPLY is set to the index of the word in reply which contains the character after the cursor, where the first element has index 1. The parameter REPLY2 is set to the index of the character under the cursor in that word, where the first character has index 1.
Hence reply, REPLY and REPLY2 should all be made local to the enclosing function.
See the function modify-current-argument, described below, for an example of how to call this function.
For example, a user-defined widget containing the following code converts the characters in the argument under the cursor into all upper case:
modify-current-argument '${(U)ARG}'
The following strips any quoting from the current word (whether backslashes or one of the styles of quotes), and replaces it with single quoting throughout:
modify-current-argument '${(qq)${(Q)ARG}}'
The behavior of several of the above widgets can be controlled by the use of the zstyle mechanism. In particular, widgets that interact with the completion system pass along their context to any completions that they invoke.
zstyle ':completion:*' completer \ _complete _correct _approximate zstyle ':completion:incremental:*' completer \ _complete _correct zstyle ':completion:predict:*' completer \ _complete
It is a good idea to restrict the completers used in prediction, because they may be automatically invoked as you type. The _list and _menu completers should never be used with prediction. The _approximate, _correct, _expand, and _match completers may be used, but be aware that they may change characters anywhere in the word behind the cursor, so you need to watch carefully that the result is what you intended.
Any other value for this style unconditionally leaves the cursor at the position where the completion code left it.
The insert-and-predict widget uses this style to decide if the completion should be shown even if there is only one possible completion. This is done if the value of this style is the string always. In this case the context is `:predict' (not `:completion:predict').
zstyle :insert-last-word match '*[[:alpha:]/\\]*'
However, you might want to include words that contain spaces:
zstyle :insert-last-word match '*[[:alpha:][:space:]/\\]*'
Or include numbers as long as the word is at least two characters long:
zstyle :insert-last-word match '*([[:digit:]]?|[[:alpha:]/\\])*'
The above example causes redirections like "2>" to be included.
Like `break-keys', this uses the `:incremental' context.
zstyle :copy-earlier-word widget smart-insert-last-word
Check the documentation for the calling widget or function to determine whether the widget style is used.
Two functions are provided to enable zsh to provide exception handling in a form that should be familiar from other languages.
The functions are designed to be used together with the always construct described in zshmisc(1). This is important as only this construct provides the required support for exceptions. A typical example is as follows.
{ # "try" block # ... nested code here calls "throw MyExcept" } always { # "always" block if catch MyExcept; then print "Caught exception MyExcept" elif catch ''; then print "Caught a shell error. Propagating..." throw '' fi # Other exceptions are not handled but may be caught further # up the call stack. }
If all exceptions should be caught, the following idiom might be preferable.
{ # ... nested code here throws an exception } always { if catch *; then case $CAUGHT in (MyExcept) print "Caught my own exception" ;; (*) print "Caught some other exception" ;; esac fi }
In common with exception handling in other languages, the exception may be thrown by code deeply nested inside the `try' block. However, note that it must be thrown inside the current shell, not in a subshell forked for a pipeline, parenthesised current-shell construct, or some form of command or process substitution.
The system internally uses the shell variable EXCEPTION to record the name of the exception between throwing and catching. One drawback of this scheme is that if the exception is not handled the variable EXCEPTION remains set and may be incorrectly recognised as the name of an exception if a shell error subsequently occurs. Adding unset EXCEPTION at the start of the outermost layer of any code that uses exception handling will eliminate this problem.
Three functions are available to provide handling of files recognised by extension, for example to dispatch a file text.ps when executed as a command to an appropriate viewer.
To enable the system, the function zsh-mime-setup should be autoloaded and run. This allows files with extensions to be treated as executable; such files be completed by the function completion system. The function zsh-mime-handler should not need to be called by the user.
The system works by setting up suffix aliases with `alias -s'. Suffix aliases already installed by the user will not be overwritten.
For suffixes defined in lower case, upper case variants will also automatically be handled (e.g. PDF is automatically handled if handling for the suffix pdf is defined), but not vice versa.
Repeated calls to zsh-mime-setup do not override the existing mapping between suffixes and executable files unless the option -f is given. Note, however, that this does not override existing suffix aliases assigned to handlers other than zsh-mime-handler.
Calling zsh-mime-setup with the option -l lists the existing mappings without altering them. Suffixes to list (which may contain pattern characters that should be quoted from immediate interpretation on the command line) may be given as additional arguments, otherwise all suffixes are listed.
Calling zsh-mime-setup with the option -v causes verbose output to be shown during the setup operation.
The system respects the mailcap flags needsterminal and copiousoutput, see mailcap(4).
The functions use the following styles, which are defined with the zstyle builtin command (see zshmodules(1)). They should be defined before zsh-mime-setup is run. The contexts used all start with :mime:, with additional components in some cases. It is recommended that a trailing * (suitably quoted) be appended to style patterns in case the system is extended in future. Some examples are given below.
Note that as this style is handled during initialisation, the context is always :mime:, with no discrimination by suffix.
Examples:
zstyle ':mime:*' mailcap ~/.mailcap /usr/local/etc/mailcap zstyle ':mime:.txt:' handler less %s zstyle ':mime:.txt:' flags needsterminal
When zsh-mime-setup is subsequently run, it will look for mailcap entries in the two files given. Files of suffix .txt will be handled by running `less file.txt'. The flag needsterminal is set to show that this program must run attached to a terminal.
As there are several steps to dispatching a command, the following should be checked if attempting to execute a file by extension .ext does not have the expected effect.
The command `alias -s ext' should show `ps=zsh-mime-handler'. If it shows something else, another suffix alias was already installed and was not overwritten. If it shows nothing, no handler was installed: this is most likely because no handler was found in the .mime.types and mailcap combination for .ext files. In that case, appropriate handling should be added to ~/.mime.types and mailcap.
If the extension is handled by zsh-mime-handler but the file is not opened correctly, either the handler defined for the type is incorrect, or the flags associated with it are in appropriate. Running zsh-mime-setup -l will show the handler and, if there are any, the flags. A %s in the handler is replaced by the file (suitably quoted if necessary). Check that the handler program listed lists and can be run in the way shown. Also check that the flags needsterminal or copiousoutput are set if the handler needs to be run under a terminal; the second flag is used if the output should be sent to a pager. An example of a suitable mailcap entry for such a program is:
text/html; /usr/bin/lynx '%s'; needsterminal
autoload -U pick-web-browser alias -s html=pick-web-browser
It is provided as an intelligent front end to dispatch a web browser. It may be run as either a function or a shell script. The status 255 is returned if no browser could be started.
Various styles are available to customize the choice of browsers:
If the style is not set the default running x tty is used.
zstyle ':mime:*' x-browsers opera konqueror firefox
specifies that pick-web-browser should first look for a running instance of Opera, Konqueror or Firefox, in that order, and if it fails to find any should attempt to start Opera. The default is firefox mozilla netscape opera konqueror.
Each line typed is evaluated as an expression. The prompt shows a number, which corresponds to a positional parameter where the result of that calculation is stored. For example, the result of the calculation on the line preceded by `4> ' is available as $4. The last value calculated is available as ans. Full command line editing, including the history of previous calculations, is available; the history is saved in the file ~/.zcalc_history. To exit, enter a blank line or type `:q' on its own (`q' is allowed for historical compatibility).
If arguments are given to zcalc on start up, they are used to prime the first few positional parameters. A visual indication of this is given when the calculator starts.
The constants PI (3.14159...) and E (2.71828...) are provided. Parameter assignment is possible, but note that all parameters will be put into the global namespace.
The output base can be initialised by passing the option `-#base', for example `zcalc -#16' (the `#' may have to be quoted, depending on the globbing options set).
The prompt is configurable via the parameter ZCALCPROMPT, which undergoes standard prompt expansion. The index of the current entry is stored locally in the first element of the array psvar, which can be referred to in ZCALCPROMPT as `%1v'. The default prompt is `%1v> '.
A few special commands are available; these are introduced by a colon. For backward compatibility, the colon may be omitted for certain commands. Completion is available if compinit has been run.
The output precision may be specified within zcalc by special commands familiar from many calculators.
Other special commands:
Note that zcalc takes care of all quoting. Hence for example:
function cube $1 * $1 * $1
defines a function to cube the sole argument.
See the comments in the function for a few extra tips.
With two arguments, define a mathematical function named mathfunc which can be used in any form of arithmetic evaluation. body is a mathematical expression to implement the function. It may contain references to position parameters $1, $2, ... to refer to mandatory parameters and ${1:-defvalue} ... to refer to optional parameters. Note that the forms must be strictly adhered to for the function to calculate the correct number of arguments. The implementation is held in a shell function named zsh_math_func_mathfunc; usually the user will not need to refer to the shell function directly. Any existing function of the same name is silently replaced.
With one argument, remove the mathematical function mathfunc as well as the shell function implementation.
With no arguments, list all mathfunc functions in a form suitable for restoring the definition. The functions have not necessarily been defined by zmathfuncdef.
The zsh/newuser module comes with a function to aid in configuring shell options for new users. If the module is installed, this function can also be run by hand. It is available even if the module's default behaviour, namely running the function for a new user logging in without startup files, is inhibited.
By default the function exits immediately if it finds any of the files .zshenv, .zprofile, .zshrc, or .zlogin in the appropriate directory. The option -f is required in order to force the function to continue. Note this may happen even if .zshrc itself does not exist.
As currently configured, the function will exit immediately if the user has root privileges; this behaviour cannot be overridden.
Once activated, the function's behaviour is supposed to be self-explanatory. Menus are present allowing the user to alter the value of options and parameters. Suggestions for improvements are always welcome.
When the script exits, the user is given the opportunity to save the new file or not; changes are not irreversible until this point. However, the script is careful to restrict changes to the file only to a group marked by the lines `# Lines configured by zsh-newuser-install' and `# End of lines configured by zsh-newuser-install'. In addition, the old version of .zshrc is saved to a file with the suffix .zni appended.
If the function edits an existing .zshrc, it is up to the user to ensure that the changes made will take effect. For example, if control usually returns early from the existing .zshrc the lines will not be executed; or a later initialization file may override options or parameters, and so on. The function itself does not attempt to detect any such conflicts.
There are a large number of helpful functions in the Functions/Misc directory of the zsh distribution. Most are very simple and do not require documentation here, but a few are worthy of special mention.
The eight base colors are: black, red, green, yellow, blue, magenta, cyan, and white. Each of these has codes for foreground and background. In addition there are eight intensity attributes: bold, faint, standout, underline, blink, reverse, and conceal. Finally, there are six codes used to negate attributes: none (reset all attributes to the defaults), normal (neither bold nor faint), no-standout, no-underline, no-blink, and no-reverse.
Some terminals do not support all combinations of colors and intensities.
The associative arrays are:
Although it is a misnomer to call them `colors', these arrays also map the other fourteen attributes from names to codes and codes to names.
In addition, the scalar parameters reset_color and bold_color are set to the ANSI terminal escapes that turn off all attributes and turn on bold intensity, respectively.
This is useful in startup files to set options and other state that are not available in all versions of zsh.
is-at-least 3.1.6-15 && setopt NO_GLOBAL_RCS is-at-least 3.1.0 && setopt HIST_REDUCE_BLANKS is-at-least 2.6-17 || print "You can't use is-at-least here."
See also the pager, prompt and rprompt styles below.
In the discussion which follows, if cmd is a filesystem path, it is first reduced to its rightmost component (the file name).
Help is first sought by looking for a file named cmd in the directory named by the HELPDIR parameter. If no file is found, an assistant function, alias, or command named run-help-cmd is sought. If found, the assistant is executed with the rest of the current command line (everything after the command name cmd) as its arguments. When neither file nor assistant is found, the external command `man cmd' is run.
An example assistant for the "ssh" command:
run-help-ssh() { emulate -LR zsh local -a args # Delete the "-l username" option zparseopts -D -E -a args l: # Delete other options, leaving: host command args=(${@:#-*}) if [[ ${#args} -lt 2 ]]; then man ssh else run-help $args[2] fi }
Several of these assistants are provided in the Functions/Misc directory. These must be autoloaded, or placed as executable scripts in your search path, in order to be found and used by run-help.
This function must be used as a ZLE widget:
autoload -U tetris zle -N tetris bindkey keys tetris
To start a game, execute the widget by typing the keys. Whatever command line you were editing disappears temporarily, and your keymap is also temporarily replaced by the Tetris control keys. The previous editor state is restored when you quit the game (by pressing `q') or when you lose.
If you quit in the middle of a game, the next invocation of the tetris widget will continue where you left off. If you lost, it will start a new game.
The option list represents options of the zargs command itself, which are the same as those of xargs. The input list is the collection of strings (often file names) that become the arguments of the command, analogous to the standard input of xargs. Finally, the arg list consists of those arguments (usually options) that are passed to the command each time it runs. The arg list precedes the elements from the input list in each run. If no command is provided, then no arg list may be provided, and in that event the default command is `print' with arguments `-r --'.
For example, to get a long ls listing of all plain files in the current directory or its subdirectories:
autoload -U zargs zargs -- **/*(.) -- ls -l
Note that `--' is used both to mark the end of the option list and to mark the end of the input list, so it must appear twice whenever the input list may be empty. If there is guaranteed to be at least one input and the first input does not begin with a `-', then the first `--' may be omitted.
In the event that the string `--' is or may be an input, the -e option may be used to change the end-of-inputs marker. Note that this does not change the end-of-options marker. For example, to use `..' as the marker:
zargs -e.. -- **/*(.) .. ls -l
This is a good choice in that example because no plain file can be named `..', but the best end-marker depends on the circumstances.
For details of the other zargs options, see xargs(1) or run zargs with the --help option.
Only one name argument is allowed. If the -f option is given, the name is taken to be that of a function; if the function is marked for autoloading, zed searches for it in the fpath and loads it. Note that functions edited this way are installed into the current shell, but not written back to the autoload file.
Without -f, name is the path name of the file to edit, which need not exist; it is created on write, if necessary.
While editing, the function sets the main keymap to zed and the vi command keymap to zed-vicmd. These will be copied from the existing main and vicmd keymaps if they do not exist the first time zed is run. They can be used to provide special key bindings used only in zed.
If it creates the keymap, zed rebinds the return key to insert a line break and `^X^W' to accept the edit in the zed keymap, and binds `ZZ' to accept the edit in the zed-vicmd keymap.
The bindings alone can be installed by running `zed -b'. This is suitable for putting into a startup file. Note that, if rerun, this will overwrite the existing zed and zed-vicmd keymaps.
Completion is available, and styles may be set with the context prefix `:completion:zed'.
A zle widget zed-set-file-name is available. This can be called by name from within zed using `\ex zed-set-file-name' (note, however, that because of zed's rebindings you will have to type ^j at the end instead of the return key), or can be bound to a key in either of the zed or zed-vicmd keymaps after `zed -b' has been run. When the widget is called, it prompts for a new name for the file being edited. When zed exits the file will be written under that name and the original file will be left alone. The widget has no effect with `zed -f'.
While zed-set-file-name is running, zed uses the keymap zed-normal-keymap, which is linked from the main keymap in effect at the time zed initialised its bindings. (This is to make the return key operate normally.) The result is that if the main keymap has been changed, the widget won't notice. This is not a concern for most users.
zmv '(*).lis' '$1.txt'
renames `foo.lis' to `foo.txt', `my.old.stuff.lis' to `my.old.stuff.txt', and so on.
The pattern is always treated as an EXTENDED_GLOB pattern. Any file whose name is not changed by the substitution is simply ignored. Any error (a substitution resulted in an empty string, two substitutions gave the same result, the destination was an existing regular file and -f was not given) causes the entire function to abort without doing anything.
Options:
Further examples:
zmv -v '(* *)' '${1// /_}'
For any file in the current directory with at least one space in the name, replace every space by an underscore and display the commands executed.
For more complete examples and other implementation details, see the zmv source file, usually located in one of the directories named in your fpath, or in Functions/Misc/zmv in the zsh distribution.
zstyle+ ':foo:bar' style1 value1 \ + ':baz' style2 value2 \ + ':frob' style3 value3
This defines `style1' with `value1' for the context :foo:bar as usual, but it also defines `style2' with `value2' for the context :foo:bar:baz and `style3' with `value3' for :foo:bar:frob. Any subcontext may be the empty string to re-use the first context unchanged.