texinode(Completion Widgets)(Completion System)(Zsh Line Editor)(Top)
chapter(Completion Widgets)
cindex(completion, widgets)
cindex(completion, programmable)
cindex(completion, controlling)
sect(Description)
The shell's programmable completion mechanism can be manipulated in two
ways; here the low-level features supporting the newer, function-based
mechanism are defined.  A complete set of shell functions based on these
features is described in
ifzman(zmanref(zshcompsys))\
ifnzman(the next chapter, noderef(Completion System)),
and users with no interest in adding to that system (or, potentially,
writing their own DASH()- see dictionary entry for `hubris') should skip
the current section.  The older system based on the tt(compctl) builtin
command is described in
ifzman(zmanref(zshcompctl))\
ifnzman(noderef(Completion Using compctl)).

Completion widgets are defined by the tt(-C) option to the tt(zle)
builtin command provided by the tt(zsh/zle) module (see
ifzman(zmanref(zshzle))\
ifnzman(noderef(The zsh/zle Module))\
). For example,

example(zle -C complete expand-or-complete completer)

defines a widget named `tt(complete)'.  The second argument is the name
of any of the builtin widgets that handle completions:
tt(complete-word), tt(expand-or-complete),
tt(expand-or-complete-prefix), tt(menu-complete),
tt(menu-expand-or-complete), tt(reverse-menu-complete),
tt(list-choices), or tt(delete-char-or-list).  Note that this will still
work even if the widget in question has been re-bound.

When this newly defined widget is bound to a key
using the tt(bindkey) builtin command defined in the tt(zsh/zle) module
(\
ifzman(see zmanref(zshzle))\
ifnzman(noderef(Zsh Line Editor))\
), typing that key will call the shell function `tt(completer)'. This
function is responsible for generating completion matches using the
builtins described below.  As with other ZLE widgets, the function is
called with its standard input closed.

Once the function returns, the completion code takes over control again
and treats the matches in the same manner as the specified builtin
widget, in this case tt(expand-or-complete).

startmenu()
menu(Completion Special Parameters)
menu(Completion Builtin Commands)
menu(Completion Condition Codes)
menu(Completion Matching Control)
menu(Completion Widget Example)
endmenu()

texinode(Completion Special Parameters)(Completion Builtin Commands)()(Completion Widgets)
sect(Completion Special Parameters)

The parameters tt(ZLE_REMOVE_SUFFIX_CHARS) and tt(ZLE_SPACE_SUFFIX_CHARS)
are used by the completion mechanism, but are not special. See
ifzman(em(Parameters Used By The Shell) in zmanref(zshparam))\
ifnzman(noderef(Parameters Used By The Shell)).

Inside completion widgets, and any functions called from them, some
parameters have special meaning; outside these functions they are not
special to the shell in any way.  These parameters are used to pass
information between the completion code and the completion widget. Some of
the builtin commands and the condition codes use or change the current
values of these parameters.  Any existing values will be hidden during
execution of completion widgets; except for tt(compstate), the parameters
are reset on each function exit (including nested function calls from
within the completion widget) to the values they had when the function was
entered.

startitem()
vindex(CURRENT)
item(tt(CURRENT))(
This is the number of the current word, i.e. the word the cursor is
currently on in the tt(words) array.  Note that this value is only
correct if the tt(ksharrays) option is not set.
)
vindex(IPREFIX)
item(tt(IPREFIX))(
Initially this will be set to the empty string.  This parameter functions
like tt(PREFIX); it contains a string which precedes the one in tt(PREFIX)
and is not considered part of the list of matches.  Typically, a string is
transferred from the beginning of tt(PREFIX) to the end of tt(IPREFIX), for
example:

example(IPREFIX=${PREFIX%%\=*}=
PREFIX=${PREFIX#*=})

causes the part of the prefix up to and including the first equal sign not
to be treated as part of a matched string.  This can be done automatically
by the tt(compset) builtin, see below.
)
vindex(ISUFFIX)
item(tt(ISUFFIX))(
As tt(IPREFIX), but for a suffix that should not be considered part
of the matches; note that the tt(ISUFFIX) string follows the tt(SUFFIX)
string.
)
vindex(PREFIX)
item(tt(PREFIX))(
Initially this will be set to the part of the current word from the
beginning of the word up to the position of the cursor; it may be altered
to give a common prefix for all matches.
)
vindex(QIPREFIX)
item(tt(QIPREFIX))(
This parameter is read-only and contains the quoted string up to the
word being completed. E.g. when completing `tt("foo)', this parameter
contains the double quote. If the tt(-q) option of tt(compset) is used
(see below), and the original string was `tt("foo bar)' with the
cursor on the `tt(bar)', this parameter contains `tt("foo )'.
)
vindex(QISUFFIX)
item(tt(QISUFFIX))(
Like tt(QIPREFIX), but containing the suffix.
)
vindex(SUFFIX)
item(tt(SUFFIX))(
Initially this will be set to the part of the current word from the
cursor position to the end; it may be altered to give a common suffix for
all matches.  It is most useful when the option tt(COMPLETE_IN_WORD) is
set, as otherwise the whole word on the command line is treated as a
prefix.
)
vindex(compstate)
cindex(completion widgets, examining and setting state in)
item(tt(compstate))(
This is an associative array with various keys and values that the
completion code uses to exchange information with the completion widget.
The keys are:

startitem()
vindex(all_quotes, compstate)
item(tt(all_quotes))(
The tt(-q) option of the tt(compset) builtin command (see below)
allows a quoted string to be broken into separate words; if the cursor is
on one of those words, that word will be completed, possibly invoking
`tt(compset -q)' recursively.  With this key it is possible to test the
types of quoted strings which are currently broken into parts in this
fashion.  Its value contains one character for each quoting level.  The
characters are a single quote or a double quote for strings quoted with
these characters, a dollars sign for strings quoted with
tt($')var(...)tt(') and a backslash for strings not starting with a
quote character.  The first character in the value always corresponds to the
innermost quoting level.
)
vindex(context, compstate)
item(tt(context))(
This will be set by the completion code to the overall context
in which completion is attempted. Possible values are:

startitem()
item(tt(array_value))(
when completing inside the value of an array parameter assignment; in
this case the tt(words) array contains the words inside the parentheses.
)
item(tt(brace_parameter))(
when completing the name of a parameter in a parameter expansion beginning
with tt(${).  This context will also be set when completing parameter
flags following tt(${LPAR()); the full command line argument is presented
and the handler must test the value to be completed to ascertain that
this is the case.
)
item(tt(assign_parameter))(
when completing the name of a parameter in a parameter assignment.
)
item(tt(command))(
when completing for a normal command (either in command position or for
an argument of the command).
)
item(tt(condition))(
when completing inside a `tt([[)...tt(]])' conditional expression; in
this case the tt(words) array contains only the words inside the
conditional expression.
)
item(tt(math))(
when completing in a mathematical environment such as a
`tt(LPAR()LPAR())...tt(RPAR()RPAR())' construct.
)
item(tt(parameter))(
when completing the name of a parameter in a parameter expansion beginning
with tt($) but not tt(${).
)
item(tt(redirect))(
when completing after a redirection operator.
)
item(tt(subscript))(
when completing inside a parameter subscript.
)
item(tt(value))(
when completing the value of a parameter assignment.
)
enditem()
)
vindex(exact, compstate)
item(tt(exact))(
Controls the behaviour when the tt(REC_EXACT) option is set.  It will be
set to tt(accept) if an exact match would be accepted, and will be unset
otherwise.

If it was set when at least one match equal to the string on the line
was generated, the match is accepted.
)
vindex(exact_string, compstate)
item(tt(exact_string))(
The string of an exact match if one was found, otherwise unset.
)
vindex(ignored, compstate)
item(tt(ignored))(
The number of completions that were ignored because they matched one of the
patterns given with the tt(-F) option to the tt(compadd) builtin
command.
)
vindex(insert, compstate)
item(tt(insert))(
This controls the manner in which a match is inserted into the command
line.  On entry to the widget function, if it is unset the command line is
not to be changed; if set to tt(unambiguous), any prefix common to all
matches is to be inserted; if set to tt(automenu-unambiguous), the
common prefix is to be inserted and the next invocation of the
completion code may start menu completion (due to the tt(AUTO_MENU)
option being set); if set to tt(menu) or tt(automenu) menu completion
will be started for the matches currently generated (in the
latter case this will happen because the tt(AUTO_MENU) is set). The
value may also contain the string `tt(tab)' when the completion code
would normally not really do completion, but only insert the TAB
character.

On exit it may be set to any of the values above (where setting it to
the empty string is the same as unsetting it), or to a number, in which
case the match whose number is given will be inserted into the command line.
Negative numbers count backward from the last match (with `tt(-1)'
selecting the last match) and out-of-range values are wrapped
around, so that a value of zero selects the last match and a value
one more than the maximum selects the first. Unless the value of this
key ends in a space, the match is inserted as in a menu completion,
i.e. without automatically appending a space.

Both tt(menu) and tt(automenu) may also specify the number of the
match to insert, given after a colon.  For example, `tt(menu:2)' says
to start menu completion, beginning with the second match.

Note that a value containing the substring `tt(tab)' makes the
matches generated be ignored and only the TAB be inserted.

Finally, it may also be set to tt(all), which makes all matches
generated be inserted into the line.
)
vindex(insert_positions, compstate)
item(tt(insert_positions))(
When the completion system inserts an unambiguous string into the
line, there may be multiple places where characters are missing or
where the character inserted differs from at least one match.  The
value of this key contains a colon separated list of all these
positions, as indexes into the command line.
)
vindex(last_prompt, compstate)
item(tt(last_prompt))(
If this is set to a non-empty string for every match added, the
completion code will move the cursor back to the previous prompt after
the list of completions has been displayed.  Initially this is set or
unset according to the tt(ALWAYS_LAST_PROMPT) option.
)
vindex(list, compstate)
item(tt(list))(
This controls whether or how the list of matches will be displayed.  If it
is unset or empty they will never be listed; if its value begins with
tt(list), they will always be listed; if it begins with tt(autolist)
or tt(ambiguous), they will be listed when the tt(AUTO_LIST) or
tt(LIST_AMBIGUOUS) options respectively would normally cause them to
be.

If the substring tt(force) appears in the value, this makes the
list be shown even if there is only one match. Normally, the list
would be shown only if there are at least two matches.

The value contains the substring tt(packed) if the tt(LIST_PACKED)
option is set. If this substring is given for all matches added to a
group, this group will show the tt(LIST_PACKED) behavior. The same is
done for the tt(LIST_ROWS_FIRST) option with the substring tt(rows).

Finally, if the value contains the string tt(explanations), only the
explanation strings, if any, will be listed and if it contains
tt(messages), only the messages (added with the tt(-x) option of
tt(compadd)) will be listed.  If it contains both tt(explanations) and
tt(messages) both kinds of explanation strings will be listed.  It
will be set appropriately on entry to a completion widget and may be
changed there.
)
vindex(list_lines, compstate)
item(tt(list_lines))(
This gives the number of lines that are needed to display the full
list of completions.  Note that to calculate the total number of lines
to display you need to add the number of lines needed for the command
line to this value, this is available as the value of the tt(BUFFERLINES)
special parameter.
)
vindex(list_max, compstate)
item(tt(list_max))(
Initially this is set to the value of the tt(LISTMAX) parameter.
It may be set to any other value; when the widget exits this value
will be used in the same way as the value of tt(LISTMAX).
)
vindex(nmatches, compstate)
item(tt(nmatches))(
The number of matches added by the completion code so far.
)
vindex(old_insert, compstate)
item(tt(old_insert))(
On entry to the widget this will be set to the number of the match of
an old list of completions that is currently inserted into the command
line. If no match has been inserted, this is unset.

As with tt(old_list), the value of this key will only be used if it is the
string tt(keep). If it was set to this value by the widget and there was an
old match inserted into the command line, this match will be kept and if
the value of the tt(insert) key specifies that another match should be
inserted, this will be inserted after the old one.
)
vindex(old_list, compstate)
item(tt(old_list))(
This is set to tt(yes) if there is still a valid list of completions
from a previous completion at the time the widget is invoked.  This will
usually be the case if and only if the previous editing operation was a
completion widget or one of the builtin completion functions.  If there is a
valid list and it is also currently shown on the screen, the value of this
key is tt(shown).

After the widget has exited the value of this key is only used if it
was set to tt(keep).  In this case the completion code will continue
to use this old list.  If the widget generated new matches, they will
not be used.
)
vindex(parameter, compstate)
item(tt(parameter))(
The name of the parameter when completing in a subscript or in the
value of a parameter assignment.
)
vindex(pattern_insert, compstate)
item(tt(pattern_insert))(
Normally this is set to tt(menu), which specifies that menu completion will
be used whenever a set of matches was generated using tt(pattern_match)
(see below).  If
it is set to any other non-empty string by the user and menu completion is
not selected by other option settings, the code will instead insert any
common prefix for the generated matches as with normal completion.
)
vindex(pattern_match, compstate)
item(tt(pattern_match))(
Locally controls the behaviour given by the tt(GLOB_COMPLETE) option.
Initially it is set to `tt(*)' if and only if the option is set.
The completion widget may set it to this value, to an empty string
(which has the same effect as unsetting it), or to any
other non-empty string.  If it is non-empty, unquoted metacharacters on the
command line will be treated as patterns; if it is `tt(*)', then
additionally a wildcard `tt(*)' is assumed at the cursor position; if
it is empty or unset, metacharacters will be treated literally.

Note that the match specifications given to the tt(compadd) builtin
command are not used if this is set to a non-empty string.
)
vindex(quote, compstate)
item(tt(quote))(
When completing inside quotes, this contains the quotation character
(i.e. either a single quote, a double quote, or a backtick).  Otherwise it
is unset.
)
vindex(quoting, compstate)
item(tt(quoting))(
When completing inside single quotes, this is set to the string
tt(single); inside double quotes, the string
tt(double); inside backticks, the string tt(backtick).
Otherwise it is unset.
)
vindex(redirect, compstate)
item(tt(redirect))(
The redirection operator when completing in a redirection position,
i.e. one of tt(<), tt(>), etc.
)
vindex(restore, compstate)
item(tt(restore))(
This is set to tt(auto) before a function is entered, which forces the
special parameters mentioned above (tt(words), tt(CURRENT), tt(PREFIX),
tt(IPREFIX), tt(SUFFIX), and tt(ISUFFIX)) to be restored to their
previous values when the function exits.   If a function unsets it or
sets it to any other string, they will not be restored.
)
vindex(to_end, compstate)
item(tt(to_end))(
Specifies the occasions on which the cursor is moved to the end of a string
when a match is inserted.  On entry to a widget function, it may be
tt(single) if this will happen when a single unambiguous match was inserted
or tt(match) if it will happen any time a match is inserted (for example,
by menu completion; this is likely to be the effect of the tt(ALWAYS_TO_END)
option).

On exit, it may be set to tt(single) as above.  It may also be set to
tt(always), or to the empty string or unset; in those cases the cursor will
be moved to the end of the string always or never respectively.  Any
other string is treated as tt(match).
)
vindex(unambiguous, compstate)
item(tt(unambiguous))(
This key is read-only and will always be set to the common (unambiguous)
prefix the completion code has generated for all matches added so far.
)
vindex(unambiguous_cursor, compstate)
item(tt(unambiguous_cursor))(
This gives the position the cursor would be placed at if the
common prefix in the tt(unambiguous) key were inserted, relative to
the value of that key. The cursor would be placed before the character
whose index is given by this key.
)
vindex(unambiguous_positions, compstate)
item(tt(unambiguous_positions))(
This contains all positions where characters in the unambiguous string
are missing or where the character inserted differs from at least one
of the matches.  The positions are given as indexes into the string
given by the value of the tt(unambiguous) key.
)
vindex(vared, compstate)
item(tt(vared))(
If completion is called while editing a line using the tt(vared)
builtin, the value of this key is set to the name of the parameter
given as an argument to tt(vared).  This key is only set while a tt(vared)
command is active.
)
enditem()
)
vindex(words)
item(tt(words))(
This array contains the words present on the command line currently being
edited.
)
enditem()

texinode(Completion Builtin Commands)(Completion Condition Codes)(Completion Special Parameters)(Completion Widgets)
sect(Completion Builtin Commands)
startitem()
findex(compadd)
cindex(completion widgets, adding specified matches)
redef(SPACES)(0)(tt(ifztexi(NOTRANS(@ @ @ @ @ @ @ @ ))ifnztexi(        )))
xitem(tt(compadd )[ tt(-akqQfenUl12C) ] [ tt(-F) var(array) ])
xitem(SPACES()[tt(-P) var(prefix) ] [ tt(-S) var(suffix) ])
xitem(SPACES()[tt(-p) var(hidden-prefix) ] [ tt(-s) var(hidden-suffix) ])
xitem(SPACES()[tt(-i) var(ignored-prefix) ] [ tt(-I) var(ignored-suffix) ])
xitem(SPACES()[tt(-W) var(file-prefix) ] [ tt(-d) var(array) ])
xitem(SPACES()[tt(-J) var(group-name) ] [ tt(-X) var(explanation) ] [ tt(-x) var(message) ])
xitem(SPACES()[tt(-V) var(group-name) ] [ tt(-o) [ var(order) ] ])
xitem(SPACES()[tt(-r) var(remove-chars) ] [ tt(-R) var(remove-func) ])
xitem(SPACES()[tt(-D) var(array) ] [ tt(-O) var(array) ] [ tt(-A) var(array) ])
xitem(SPACES()[tt(-E) var(number) ])
item(SPACES()[tt(-M) var(match-spec) ] [ tt(-)tt(-) ] [ var(completions) ... ])(

This builtin command can be used to add matches directly and control
all the information the completion code stores with each possible
completion. The return status is zero if at least one match was added and
non-zero if no matches were added.

The completion code breaks each match into seven fields in the order:

indent(var(<ipre><apre><hpre><body><hsuf><asuf><isuf>))

The first field
is an ignored prefix taken from the command line, the contents of the
tt(IPREFIX) parameter plus the string given with the tt(-i)
option. With the tt(-U) option, only the string from the tt(-i)
option is used. The field var(<apre>) is an optional prefix string
given with the tt(-P) option.  The var(<hpre>) field is a string
that is considered part of the match but that should not be shown when
listing completions, given with the tt(-p) option; for example,
functions that do filename generation might specify
a common path prefix this way.  var(<body>) is the part of the match that
should appear in the list of matches shown to the user.
The suffixes var(<hsuf>),
var(<asuf>) and var(<isuf>) correspond to the prefixes var(<hpre>),
var(<apre>) and var(<ipre>) and are given by the options tt(-s), tt(-S) and
tt(-I), respectively.

The supported flags are:

startitem()
item(tt(-P) var(prefix))(
This gives a string to be inserted before each match.  The
string given is not considered as part of the match and any shell
metacharacters in it will not be quoted when the string is inserted.
)
item(tt(-S) var(suffix))(
Like tt(-P), but gives a string to be inserted after each match.
)
item(tt(-p) var(hidden-prefix))(
This gives a string that should be inserted before each
match but that should not appear in the list of matches. Unless the
tt(-U) option is given, this string must be matched as part of the string
on the command line.
)
item(tt(-s) var(hidden-suffix))(
Like `tt(-p)', but gives a string to insert after each match.
)
item(tt(-i) var(ignored-prefix))(
This gives a string to insert just before any
string given with the `tt(-P)' option.  Without `tt(-P)' the string is
inserted before the string given with `tt(-p)' or directly before each
match.
)
item(tt(-I) var(ignored-suffix))(
Like tt(-i), but gives an ignored suffix.
)
item(tt(-a))(
With this flag the var(completions) are taken as names of arrays and the
actual completions are their values.  If only some elements of the
arrays are needed, the var(completions) may also contain subscripts, as in
`tt(foo[2,-1])'.
)
item(tt(-k))(
With this flag the var(completions) are taken as names of associative arrays
and the actual completions are their keys.  As for tt(-a), the
var(words) may also contain subscripts, as in `tt(foo[(R)*bar*])'.
)
item(tt(-d) var(array))(
This adds per-completion display strings. The var(array) should contain one
element per var(completion) given. The completion code will then display the
first element instead of the first var(completion), and so on. The
var(array) may be given as the name of an array parameter or directly
as a space-separated list of words in parentheses.

If there are fewer display strings than var(completions), the leftover
var(completions) will be displayed unchanged and if there are more display
strings than var(completions), the leftover display strings will be silently
ignored.
)
item(tt(-l))(
This option only has an effect if used together with the tt(-d)
option. If it is given, the display strings are listed one per line,
not arrayed in columns.
)
item(tt(-o) [ var(order) ])(
This controls the order in which matches are sorted. var(order) is a
comma-separated list comprising the following possible values.  These values
can be abbreviated to their initial two or three characters.  Note that the
order forms part of the group name space so matches with different orderings
will not be in the same group.

startitem()
item(tt(match))(
If given, the order of the output is determined by the match strings;
otherwise it is determined by the display strings (i.e. the strings given
by the tt(-d) option). This is the default if `tt(-o)' is specified but
the var(order) argument is omitted.
)
item(tt(nosort))(
This specifies that the var(completions)
are pre-sorted and their order should be
preserved.  This value only makes sense alone and cannot be combined with any
others.
)
item(tt(numeric))(
If the matches include numbers, sort them numerically rather than
lexicographically.
)
item(tt(reverse))(
Arrange the matches backwards by reversing the sort ordering.
)
enditem()
)
item(tt(-J) var(group-name))(
Gives the name of the group that the matches should be stored in.
)
item(tt(-V) var(group-name))(
Like tt(-J) but naming an unsorted group. This option is identical to
the combination of tt(-J) and tt(-o nosort).
)
item(tt(-1))(
If given together with the tt(-V) option, makes
only consecutive duplicates in the group be removed. If combined with
the tt(-J) option, this has no visible effect. Note that groups
with and without this flag are in different name spaces.
)
item(tt(-2))(
If given together with the tt(-J) or tt(-V) option, makes all
duplicates be kept. Again, groups with and without this flag are in
different name spaces.
)
item(tt(-X) var(explanation))(
The var(explanation) string will be printed with the list of matches,
above the group currently selected.

Within the var(explanation), the following sequences may be used to
specify output attributes
ifnzman((see noderef(Prompt Expansion)))\
ifzman(as described in the section EXPANSION OF PROMPT SEQUENCES in
zmanref(zshmisc)):
`tt(%B)', `tt(%S)', `tt(%U)', `tt(%F)', `tt(%K)' and their lower case
counterparts, as well as `tt(%{)...tt(%})'.  `tt(%F)', `tt(%K)' and
`tt(%{)...tt(%})' take arguments in the same form as prompt
expansion.  (Note that the sequence `tt(%G)' is not available; an
argument to `tt(%{)' should be used instead.)  The sequence `tt(%%)'
produces a literal `tt(%)'.

These sequences are most often employed by users when customising the
tt(format) style
(see
ifzman(zmanref(zshcompsys))\
ifnzman(noderef(Completion System))\
),
but they must also be taken into account when writing completion
functions, as passing descriptions with unescaped `tt(%)' characters
to utility functions such as tt(_arguments) and tt(_message) may
produce unexpected results. If arbitrary text is to be passed in a
description, it can be escaped using e.g. tt(${my_str//\%/%%}).
)
item(tt(-x) var(message))(
Like tt(-X), but the var(message) will be printed even if there are no
matches in the group.
)
item(tt(-q))(
The suffix given with tt(-S) will be automatically removed if
the next character typed is a blank or does not insert anything, or if
the suffix consists of only one character and the next character typed
is the same character.
)
item(tt(-r) var(remove-chars))(
This is a more versatile form of the tt(-q) option.
The suffix given with tt(-S) or the slash automatically added after
completing directories will be automatically removed if
the next character typed inserts one of the characters given in the
var(remove-chars).  This string is parsed as a characters class and
understands the backslash sequences used by the tt(print) command.  For
example, `tt(-r "a-z\t")' removes the suffix if the next character typed
inserts a lower case character or a TAB, and `tt(-r "^0-9")' removes the
suffix if the next character typed inserts anything but a digit. One extra
backslash sequence is understood in this string: `tt(\-)' stands for
all characters that insert nothing. Thus `tt(-S "=" -q)' is the same
as `tt(-S "=" -r "= \t\n\-")'.

This option may also be used without the tt(-S) option; then any
automatically added space will be removed when one of the characters in the
list is typed.
)
item(tt(-R) var(remove-func))(
This is another form of the tt(-r) option. When a match
has been accepted and a suffix has been inserted, the function
var(remove-func) will be called after the next character typed.  It is
passed the length of the suffix as an argument and can use the special
parameters available in ordinary (non-completion) zle widgets (see
ifzman(zmanref(zshzle))\
ifnzman(noderef(Zsh Line Editor))\
) to analyse and modify the command line.
)
item(tt(-f))(
If this flag is given, all of the matches built from the var(completions) are
marked as being the names of files.  They are not required to be actual
filenames, but if they are, and the option tt(LIST_TYPES) is set, the
characters describing the types of the files in the completion lists will
be shown. This also forces a slash to be added when the name of a
directory is completed.
)
item(tt(-e))(
This flag can be used to tell the completion code that the matches
added are parameter names for a parameter expansion. This will make
the tt(AUTO_PARAM_SLASH) and tt(AUTO_PARAM_KEYS) options be used for
the matches.
)
item(tt(-W) var(file-prefix))(
This string is a pathname that will be prepended to each match together
with any prefix specified by the tt(-p) option to form a complete filename
for testing.  Hence it is only useful if combined with the tt(-f) flag, as
the tests will not otherwise be performed.
)
item(tt(-F) var(array))(
Specifies an array containing patterns.  var(completions) that match one of
these patterns are ignored, that is, not considered to be matches.

The var(array) may be the name of an array parameter or a list of
literal patterns enclosed in parentheses and quoted, as in `tt(-F "(*?.o
*?.h)")'. If the name of an array is given, the elements of the array are
taken as the patterns.
)
item(tt(-Q))(
This flag instructs the completion
code not to quote any metacharacters in the matches when inserting them
into the command line.
)
item(tt(-M) var(match-spec))(
This gives local match specifications as described below in
noderef(Completion Matching Control). This option may be given more than once.
In this case all var(match-spec)s given are concatenated with spaces
between them to form the specification string to use.
Note that they will only be used if the tt(-U) option is not given.
)
item(tt(-n))(
Specifies that matching var(completions) are to be added to the set of
matches, but are not to be listed to the user.
)
item(tt(-U))(
If this flag is given, all var(completions) are added
to the set of matches and no matching
will be done by the completion code. Normally this is used in
functions that do the matching themselves.
)
item(tt(-O) var(array))(
If this option is given, the var(completions) are em(not) added to the set of
matches.  Instead, matching is done as usual and all of the
var(completions) that match
will be stored in the array parameter whose name is given as var(array).
)
item(tt(-A) var(array))(
As the tt(-O) option, except that instead of those of the var(completions)
which
match being stored in var(array), the strings generated internally by the
completion code are stored.  For example,
with a match specification of `tt(-M "L:|no=")', a current word of `tt(nof)'
and var(completions) of `tt(foo)', this
option stores the string `tt(nofoo)' in the array, whereas the tt(-O)
option stores the `tt(foo)' originally given.
)
item(tt(-D) var(array))(
As with tt(-O), the var(completions) are not added to the set of matches.
Instead, whenever the var(n)th var(completion) does not
match, the var(n)th element of the var(array) is removed.  Elements
for which the corresponding var(completion) matches are retained.
This option can be used more than once to remove elements from multiple
arrays.
)
item(tt(-C))(
This option adds a special match which expands to all other matches
when inserted into the line, even those that are added after this
option is used.  Together with the tt(-d) option it is possible to
specify a string that should be displayed in the list for this special
match.  If no string is given, it will be shown as a string containing
the strings that would be inserted for the other matches, truncated to
the width of the screen.
)
item(tt(-E) var(number))(
This option adds var(number) empty matches after matching var(completions) have
been added.  An empty match takes up space in completion listings but
will never be inserted in the line and can't be selected with menu
completion or menu selection.  This makes empty matches only useful to
format completion lists and to make explanatory string be shown in
completion lists (since empty matches can be given display strings
with the tt(-d) option).  And because all but one empty string would
otherwise be removed, this option implies the tt(-V) and tt(-2)
options (even if an explicit tt(-J) option is given).  This can be
important to note as it affects the name space into which matches are
added.
)
xitem(tt(-))
item(tt(-)tt(-))(
This flag ends the list of flags and options. All arguments after it
will be taken as the var(completions) even if they begin with
hyphens.
)
enditem()

Except for the tt(-M) flag, if any of these flags is given more than
once, the first one (and its argument) will be used.
)
findex(compset)
cindex(completion widgets, modifying special parameters)
xitem(tt(compset -p) var(number))
xitem(tt(compset -P) [ var(number) ] var(pattern))
xitem(tt(compset -s) var(number))
xitem(tt(compset -S) [ var(number) ] var(pattern))
xitem(tt(compset -n) var(begin) [ var(end) ])
xitem(tt(compset -N) var(beg-pat) [ var(end-pat) ])
item(tt(compset -q))(
This command simplifies modification of the special parameters,
while its return status allows tests on them to be carried out.

The options are:

startitem()
item(tt(-p) var(number))(
If the value of the tt(PREFIX) parameter is at least var(number)
characters long, the first var(number) characters are removed from it and
appended to the contents of the tt(IPREFIX) parameter.
)
item(tt(-P) [ var(number) ] var(pattern))(
If the value of the tt(PREFIX) parameter begins with anything that
matches the var(pattern), the matched portion is removed from
tt(PREFIX) and appended to tt(IPREFIX).

Without the optional var(number), the longest match is taken, but
if var(number) is given, anything up to the var(number)th match is
moved.  If the var(number) is negative, the var(number)th longest
match is moved. For example, if tt(PREFIX) contains the string
`tt(a=b=c)', then tt(compset -P '*\=') will move the string `tt(a=b=)'
into the tt(IPREFIX) parameter, but tt(compset -P 1 '*\=') will move only
the string `tt(a=)'.
)
item(tt(-s) var(number))(
As tt(-p), but transfer the last var(number) characters from the
value of tt(SUFFIX) to the front of the value of tt(ISUFFIX).
)
item(tt(-S) [ var(number) ] var(pattern))(
As tt(-P), but match the last portion of tt(SUFFIX) and transfer the
matched portion to the front of the value of tt(ISUFFIX).
)
item(tt(-n) var(begin) [ var(end) ])(
If the current word position as specified by the parameter tt(CURRENT)
is greater than or equal to var(begin), anything up to the
var(begin)th word is removed from the tt(words) array and the value
of the parameter tt(CURRENT) is decremented by var(begin).

If the optional var(end) is given, the modification is done only if
the current word position is also less than or equal to var(end). In
this case, the words from position var(end) onwards are also removed from
the tt(words) array.

Both var(begin) and var(end) may be negative to count backwards
from the last element of the tt(words) array.
)
item(tt(-N) var(beg-pat) [ var(end-pat) ])(
If one of the elements of the tt(words) array before the one at the
index given by the value of the parameter tt(CURRENT) matches the
pattern var(beg-pat), all elements up to and including the matching one are
removed from the tt(words) array and the value of tt(CURRENT) is changed to
point to the same word in the changed array.

If the optional pattern var(end-pat) is also given, and there is an
element in the tt(words) array matching this pattern, the parameters
are modified only if the index of this word is higher than the one
given by the tt(CURRENT) parameter (so that the matching word has
to be after the cursor). In this case, the words starting with the one
matching tt(end-pat) are also removed from the tt(words)
array. If tt(words) contains no word matching var(end-pat), the
testing and modification is performed as if it were not given.
)
item(tt(-q))(
The word
currently being completed is split on spaces into separate words,
respecting the usual shell quoting conventions.  The
resulting words are stored in the tt(words) array, and tt(CURRENT),
tt(PREFIX), tt(SUFFIX), tt(QIPREFIX), and tt(QISUFFIX) are modified to
reflect the word part that is completed.
)
enditem()

In all the above cases the return status is zero if the test succeeded
and the parameters were modified and non-zero otherwise. This allows
one to use this builtin in tests such as:

example(if compset -P '*\='; then ...)

This forces anything up to and including the last equal sign to be
ignored by the completion code.
)
item(tt(compcall) [ tt(-TD) ])(
This allows the use of completions defined with the tt(compctl) builtin
from within completion widgets.  The list of matches will be generated as
if one of the non-widget completion functions (tt(complete-word), etc.)
had been called, except that only tt(compctl)s given for specific commands
are used. To force the code to try completions defined with the tt(-T)
option of tt(compctl) and/or the default completion (whether defined by
tt(compctl -D) or the builtin default) in the appropriate places, the
tt(-T) and/or tt(-D) flags can be passed to tt(compcall).

The return status can be used to test if a matching tt(compctl)
definition was found. It is non-zero if a tt(compctl) was found and
zero otherwise.

Note that this builtin is defined by the tt(zsh/compctl) module.
)
enditem()

texinode(Completion Condition Codes)(Completion Matching Control)(Completion Builtin Commands)(Completion Widgets)
sect(Completion Condition Codes)
cindex(completion widgets, condition codes)

The following additional condition codes for use within the tt([[) var(...) tt(]])
construct are available in completion widgets.  These work on the special
parameters.  All of these tests can also be performed by the tt(compset)
builtin, but in the case of the condition codes the contents of the special
parameters are not modified.

startitem()
item(tt(-prefix) [ var(number) ] var(pattern))(
true if the test for the tt(-P) option of tt(compset) would succeed.
)
item(tt(-suffix) [ var(number) ] var(pattern))(
true if the test for the tt(-S) option of tt(compset) would succeed.
)
item(tt(-after) var(beg-pat))(
true if the test of the tt(-N) option with only the var(beg-pat) given
would succeed.
)
item(tt(-between) var(beg-pat end-pat))(
true if the test for the tt(-N) option with both patterns would succeed.
)
enditem()

texinode(Completion Matching Control)(Completion Widget Example)(Completion Condition Codes)(Completion Widgets)
sect(Completion Matching Control)

When the user invokes completion, the current em(word) on the command line
(that is, the word the cursor is currently on) is used to generate a em(match
pattern).  Only those em(completions) that match the pattern are offered to the
user as em(matches).

The default match pattern is generated from the current word by either

startitemize()
itemiz(\
appending a `tt(*)' (matching any number of characters in a completion)
em(or,)\
)
itemiz(\
if the shell option tt(COMPLETE_IN_WORD) is set, inserting a `tt(*)' at the
cursor position.\
)
enditemize()

This narrow pattern can be broadened selectively by passing a em(match
specification) to the tt(compadd) builtin command through its tt(-M) option
(see
ifzman(`Completion Builtin Commands' above)\
ifnzman(noderef(Completion Builtin Commands))\
).  A match specification consists of one or more var(matchers) separated by
whitespace.  Matchers in a match specification are applied one at a time, from
left to right.  Once all matchers have been applied, completions are compared
to the final match pattern and non-matching ones are discarded.

startitemize()
itemiz(\
Note that the tt(-M) option is ignored if the current word contains a glob
pattern and the shell option tt(GLOB_COMPLETE) is set or if the
tt(pattern_match) key of the special associative array tt(compstate) is set to
a non-empty value (see
ifzman(`Completion Special Parameters' above)\
ifnzman(noderef(Completion Special Parameters))\
).\
)
itemiz(\
Users of the \
ifzman(completion system (see zmanref(zshcompsys))) \
ifnzman(noderef(Completion System)) \
should generally not use the tt(-M) option directly, but rather use the
tt(matcher-list) and tt(matcher) styles (see the subsection em(Standard Styles)
in
ifzman(\
the documentation for COMPLETION SYSTEM CONFIGURATION in zmanref(zshcompsys))\
ifnzman(noderef(Completion System Configuration))\
).\
)
enditemize()

Each matcher consists of

startitemize()
itemiz(a case-sensitive letter)
itemiz(a `tt(:)',)
itemiz(one or more patterns separated by pipes (`tt(|)'),)
itemiz(an equals sign (`tt(=)'), and)
itemiz(another pattern.)
enditemize()

The patterns before the `tt(=)' are used to match substrings of the current
word.  For each matched substring, the corresponding part of the match pattern
is broadened with the pattern after the `tt(=)', by means of a logical tt(OR).

Each pattern in a matcher cosists of either

startitemize()
itemiz(the empty string or)
itemiz(a sequence of

startitemize()
itemiz(literal characters (which may be quoted with a `tt(\)'),)
itemiz(question marks (`tt(?)'),)
itemiz(\
bracket expressions (`tt([...])'; see the subsection em(Glob Operators) in
ifnzman(noderef(Filename Generation))\
ifzman(the documentation for GLOB OPERATORS in zmanref(zshexpn))\
), and/or\
)
itemiz(brace expressions (see below).)
enditemize()
)
enditemize()

Other shell patterns are not allowed.

A brace expression, like a bracket expression, consists of a list of

startitemize()
itemiz(literal characters,)
itemiz(ranges (`tt(0-9)'), and/or)
itemiz(character classes (`tt([:)var(name)tt(:])').)
enditemize()

However, they differ from each other as follows:

startitemize()
itemiz(\
A brace expression is delimited by a pair of braces (`tt({...})').\
)
itemiz(\
Brace expressions do not support negations.  That is, an initial
`tt(!)' or `tt(^)' has no special meaning and will be interpreted as a literal
character.\
)
itemiz(\
When a character in the current word matches the var(n)th pattern in a brace
expression, the corresponding part of the match pattern is broadened only with
the var(n)th pattern of the brace expression on the other side of the `tt(=)',
if there is one; if there is no brace expression on the other side, then this
pattern is the empty string.  However, if either brace expression has more
elements than the other, then the excess entries are simply ignored.  When
comparing indexes, each literal character or character class counts as one
element, but each range is instead expanded to the full list of literal
characters it represents.  Additionally, if on em(both) sides of the
`tt(=)', the var(n)th pattern is `tt([:upper:])' or `tt([:lower:])', then these
are expanded as ranges, too.\
)
enditemize()

Note that, although the matching system does not yet handle multibyte
characters, this is likely to be a future extension.  Hence, using
`tt([:upper:])' and `tt([:lower:])' is recommended over
`tt(A-Z)' and `tt(a-z)'.

Below are the different forms of matchers supported.  Each em(uppercase) form
behaves exactly like its lowercase counterpart, but adds an additional step
em(after) the match pattern has filtered out non-matching completions:  Each of
a match's substrings that was matched by a subpattern from an uppercase matcher
is replaced with the corresponding substring of the current word.  However,
patterns from em(lowercase) matchers have higher weight:  If a substring of the
current word was matched by patterns from both a lowercase and an uppercase
matcher, then the lowercase matcher's pattern wins and the corresponding part
of the match is not modified.

Unless indicated otherwise, each example listed assumes tt(COMPLETE_IN_WORD) to
be unset (as it is by default).

startitem()
xitem(tt(m:)var(word-pat)tt(=)var(match-pat))
item(tt(M:)var(word-pat)tt(=)var(match-pat))(

For each substring of the current word that matches var(word-pat), broaden the
corresponding part of the match pattern to additionally match var(match-pat).

startitem()
item(Examples:)(

tt(m:{[:lower:]}={[:upper:]}) lets any lower case character in the current word
be completed to itself or its uppercase counterpart.  So, the completions
`tt(foo)', `tt(FOO)' and `tt(Foo)' will are be considered matches for the word
`tt(fo)'.

tt(M:_=) inserts every underscore from the current word into each match, in the
same relative position, determined by matching the substrings around it.  So,
given a completion `tt(foo)', the word `tt(f_o)' will be completed to the match
`tt(f_oo)', even though the latter was not present as a completion.
)
enditem()
)
xitem(tt(b:)var(word-pat)tt(=)var(match-pat))
xitem(tt(B:)var(word-pat)tt(=)var(match-pat))
xitem(tt(e:)var(word-pat)tt(=)var(match-pat))
item(tt(E:)var(word-pat)tt(=)var(match-pat))(

For each consecutive substring at the tt(b:)eginning or tt(e:)nd of the current
word that matches var(word-pat), broaden the corresponding part of the match
pattern to additionally match var(match-pat).

startitem()
item(Examples:)(

`tt(b:-=+)' lets any number of minuses at the start of the current word be
completed to a minus or a plus.

`tt(B:0=)' adds all zeroes at the beginning of the current word to the
beginning of each match.
)
enditem()
)
xitem(tt(l:)tt(|)var(word-pat)tt(=)var(match-pat))
xitem(tt(L:)tt(|)var(word-pat)tt(=)var(match-pat))
xitem(tt(r:)var(word-pat)tt(|)tt(=)var(match-pat))
item(tt(R:)var(word-pat)tt(|)tt(=)var(match-pat))(

If there is a substring at the tt(l:)eft or tt(r:)ight edge of the current word
that matches var(word-pat), then broaden the corresponding part of the match
pattern to additionally match var(match-pat).

For each tt(l:), tt(L:), tt(r:) and tt(R:) matcher (including the ones below),
the pattern var(match-pat) may also be a `tt(*)'.  This matches any number of
characters in a completion.

startitem()
item(Examples:)(

`tt(r:|=*)' appends a `tt(*)' to the match pattern, even when
tt(COMPLETE_IN_WORD) is set and the cursor is not at the end of the current
word.

If the current word starts with a minus, then `tt(L:|-=)' will prepend it to
each match.
)
enditem()
)
xitem(tt(l:)var(anchor)tt(|)var(word-pat)tt(=)var(match-pat))
xitem(tt(L:)var(anchor)tt(|)var(word-pat)tt(=)var(match-pat))
xitem(tt(r:)var(word-pat)tt(|)var(anchor)tt(=)var(match-pat))
item(tt(R:)var(word-pat)tt(|)var(anchor)tt(=)var(match-pat))(

For each substring of the current word that matches var(word-pat) and has on
its tt(l:)eft or tt(r:)ight another substring matching var(anchor), broaden the
corresponding part of the match pattern to additionally match var(match-pat).

Note that these matchers (and the ones below) modify only what is matched by
var(word-pat); they do not change the matching behavior of what is matched by
var(anchor) (or var(coanchor); see the matchers below).  Thus, unless its
corresponding part of the match pattern has been modified, the anchor in the
current word has to match literally in each completion, just like any other
substring of the current word.

If a matcher includes at least one anchor (which includes the matchers with two
anchors, below), then var(match-pat) may also be `tt(*)' or `tt(**)'.  `tt(*)'
can match any part of a completion that does not contain any substrings
matching var(anchor), whereas a `tt(**)' can match any part of a completion,
period.  (Note that this is different from the behavior of `tt(*)' in the
anchorless forms of `tt(l:)' and `tt(r:)' and and also different from `tt(*)'
and `tt(**)' in glob expressions.)

startitem()
item(Examples:)(

`tt(r:|.=*)' makes the completion `tt(comp.sources.unix)' a match for the word
`tt(..u)' DASH()- but em(not) for the word `tt(.u)'.

Given a completion `tt(-)tt(-foo)', the matcher `tt(L:--|no-=)' will complete
the word `tt(-)tt(-no-)' to the match `tt(-)tt(-no-foo)'.
)
enditem()
)
xitem(tt(l:)var(anchor)tt(||)var(coanchor)tt(=)var(match-pat))
xitem(tt(L:)var(anchor)tt(||)var(coanchor)tt(=)var(match-pat))
xitem(tt(r:)var(coanchor)tt(||)var(anchor)tt(=)var(match-pat))
item(tt(R:)var(coanchor)tt(||)var(anchor)tt(=)var(match-pat))(

For any two consecutive substrings of the current word that match var(anchor)
and var(coanchor), in the order given, insert the pattern var(match-pat)
between their corresponding parts in the match pattern.

Note that, unlike var(anchor), the pattern var(coanchor) does not change what
`tt(*)' can match.

startitem()
item(Examples:)(

`tt(r:?||[[:upper:]]=*)' will complete the current word `tt(fB)' to
`tt(fooBar)', but it will not complete it to `tt(fooHooBar)' (because `tt(*)'
here cannot match anything that includes a match for `tt([[:upper:]])), nor
will it complete `tt(B)' to `tt(fooBar)' (because there is no character in the
current word to match var(coanchor)).

Given the current word `tt(pass.n)' and a completion `tt(pass.byname)', the
matcher `tt(L:.||[[:alpha:]]=by)' will produce the match `tt(pass.name)'.
)
enditem()
)
item(tt(x:))(

Ignore this matcher and all matchers to its right.

This matcher is used to mark the end of a match specification.  In a single
standalone list of matchers, this has no use, but where match specifications
are concatenated, as is often the case when using the
ifzman(completion system (see zmanref(zshcompsys)))\
ifnzman(noderef(Completion System))\
, it can allow one match specification to override another.
)
enditem()

texinode(Completion Widget Example)()(Completion Matching Control)(Completion Widgets)
sect(Completion Widget Example)
cindex(completion widgets, example)

The first step is to define the widget:

example(zle -C complete complete-word complete-files)

Then the widget can be bound to a key using the tt(bindkey) builtin
command:

example(bindkey '^X\t' complete)

After that the shell function tt(complete-files) will be invoked
after typing control-X and TAB. The function should then generate the
matches, e.g.:

example(complete-files LPAR()RPAR() { compadd - * })

This function will complete files in the current directory matching the
current word.