1ZSHEXPN(1) General Commands Manual ZSHEXPN(1)
2
6 zshexpn - zsh expansion and substitution
7
9 The following types of expansions are performed in the indicated order
10 in five steps:
11 History Expansion
13 This is performed only in interactive shells.
14 Alias Expansion
16 Aliases are expanded immediately before the command line is
17 parsed as explained under Aliasing in zshmisc(1).
18 Process Substitution
20 Parameter Expansion
21 Command Substitution
22 Arithmetic Expansion
23 Brace Expansion
24 These five are performed in left-to-right fashion. On each ar‐
25 gument, any of the five steps that are needed are performed one
26 after the other. Hence, for example, all the parts of parameter
27 expansion are completed before command substitution is started.
28 After these expansions, all unquoted occurrences of the charac‐
29 ters `\',`'' and `"' are removed.
30 Filename Expansion
32 If the SH_FILE_EXPANSION option is set, the order of expansion
33 is modified for compatibility with sh and ksh. In that case
34 filename expansion is performed immediately after alias expan‐
35 sion, preceding the set of five expansions mentioned above.
36 Filename Generation
38 This expansion, commonly referred to as globbing, is always done
39 last.
40 The following sections explain the types of expansion in detail.
42
44 History expansion allows you to use words from previous command lines
45 in the command line you are typing. This simplifies spelling correc‐
46 tions and the repetition of complicated commands or arguments.
47 Immediately before execution, each command is saved in the history
49 list, the size of which is controlled by the HISTSIZE parameter. The
50 one most recent command is always retained in any case. Each saved
51 command in the history list is called a history event and is assigned a
52 number, beginning with 1 (one) when the shell starts up. The history
53 number that you may see in your prompt (see EXPANSION OF PROMPT SE‐
54 QUENCES in zshmisc(1)) is the number that is to be assigned to the next
55 command.
56 Overview
58 A history expansion begins with the first character of the histchars
59 parameter, which is `!' by default, and may occur anywhere on the com‐
60 mand line, including inside double quotes (but not inside single quotes
61 '...' or C-style quotes $'...' nor when escaped with a backslash).
62 The first character is followed by an optional event designator (see
64 the section `Event Designators') and then an optional word designator
65 (the section `Word Designators'); if neither of these designators is
66 present, no history expansion occurs.
67 Input lines containing history expansions are echoed after being ex‐
69 panded, but before any other expansions take place and before the com‐
70 mand is executed. It is this expanded form that is recorded as the
71 history event for later references.
72 History expansions do not nest.
74 By default, a history reference with no event designator refers to the
76 same event as any preceding history reference on that command line; if
77 it is the only history reference in a command, it refers to the previ‐
78 ous command. However, if the option CSH_JUNKIE_HISTORY is set, then
79 every history reference with no event specification always refers to
80 the previous command.
81 For example, `!' is the event designator for the previous command, so
83 `!!:1' always refers to the first word of the previous command, and
84 `!!$' always refers to the last word of the previous command. With
85 CSH_JUNKIE_HISTORY set, then `!:1' and `!$' function in the same manner
86 as `!!:1' and `!!$', respectively. Conversely, if CSH_JUNKIE_HISTORY
87 is unset, then `!:1' and `!$' refer to the first and last words, re‐
88 spectively, of the same event referenced by the nearest other history
89 reference preceding them on the current command line, or to the previ‐
90 ous command if there is no preceding reference.
91 The character sequence `^foo^bar' (where `^' is actually the second
93 character of the histchars parameter) repeats the last command, replac‐
94 ing the string foo with bar. More precisely, the sequence `^foo^bar^'
95 is synonymous with `!!:s^foo^bar^', hence other modifiers (see the sec‐
96 tion `Modifiers') may follow the final `^'. In particular,
97 `^foo^bar^:G' performs a global substitution.
98 If the shell encounters the character sequence `!"' in the input, the
100 history mechanism is temporarily disabled until the current list (see
101 zshmisc(1)) is fully parsed. The `!"' is removed from the input, and
102 any subsequent `!' characters have no special significance.
103 A less convenient but more comprehensible form of command history sup‐
105 port is provided by the fc builtin.
106 Event Designators
108 An event designator is a reference to a command-line entry in the his‐
109 tory list. In the list below, remember that the initial `!' in each
110 item may be changed to another character by setting the histchars pa‐
111 rameter.
112 ! Start a history expansion, except when followed by a blank, new‐
114 line, `=' or `('. If followed immediately by a word designator
115 (see the section `Word Designators'), this forms a history ref‐
116 erence with no event designator (see the section `Overview').
117 !! Refer to the previous command. By itself, this expansion re‐
119 peats the previous command.
120 !n Refer to command-line n.
122 !-n Refer to the current command-line minus n.
124 !str Refer to the most recent command starting with str.
126 !?str[?]
128 Refer to the most recent command containing str. The trailing
129 `?' is necessary if this reference is to be followed by a modi‐
130 fier or followed by any text that is not to be considered part
131 of str.
132 !# Refer to the current command line typed in so far. The line is
134 treated as if it were complete up to and including the word be‐
135 fore the one with the `!#' reference.
136 !{...} Insulate a history reference from adjacent characters (if neces‐
138 sary).
139 Word Designators
141 A word designator indicates which word or words of a given command line
142 are to be included in a history reference. A `:' usually separates the
143 event specification from the word designator. It may be omitted only
144 if the word designator begins with a `^', `$', `*', `-' or `%'. Word
145 designators include:
146 0 The first input word (command).
148 n The nth argument.
149 ^ The first argument. That is, 1.
150 $ The last argument.
151 % The word matched by (the most recent) ?str search.
152 x-y A range of words; x defaults to 0.
153 * All the arguments, or a null value if there are none.
154 x* Abbreviates `x-$'.
155 x- Like `x*' but omitting word $.
156 Note that a `%' word designator works only when used in one of `!%',
158 `!:%' or `!?str?:%', and only when used after a !? expansion (possibly
159 in an earlier command). Anything else results in an error, although
160 the error may not be the most obvious one.
161 Modifiers
163 After the optional word designator, you can add a sequence of one or
164 more of the following modifiers, each preceded by a `:'. These modi‐
165 fiers also work on the result of filename generation and parameter ex‐
166 pansion, except where noted.
167 a Turn a file name into an absolute path: prepends the current
169 directory, if necessary; remove `.' path segments; and remove
170 `..' path segments and the segments that immediately precede
171 them.
172 This transformation is agnostic about what is in the filesystem,
174 i.e. is on the logical, not the physical directory. It takes
175 place in the same manner as when changing directories when nei‐
176 ther of the options CHASE_DOTS or CHASE_LINKS is set. For exam‐
177 ple, `/before/here/../after' is always transformed to `/be‐
178 fore/after', regardless of whether `/before/here' exists or what
179 kind of object (dir, file, symlink, etc.) it is.
180 A Turn a file name into an absolute path as the `a' modifier does,
182 and then pass the result through the realpath(3) library func‐
183 tion to resolve symbolic links.
184 Note: on systems that do not have a realpath(3) library func‐
186 tion, symbolic links are not resolved, so on those systems `a'
187 and `A' are equivalent.
188 Note: foo:A and realpath(foo) are different on some inputs. For
190 realpath(foo) semantics, see the `P` modifier.
191 c Resolve a command name into an absolute path by searching the
193 command path given by the PATH variable. This does not work for
194 commands containing directory parts. Note also that this does
195 not usually work as a glob qualifier unless a file of the same
196 name is found in the current directory.
197 e Remove all but the part of the filename extension following the
199 `.'; see the definition of the filename extension in the de‐
200 scription of the r modifier below. Note that according to that
201 definition the result will be empty if the string ends with a
202 `.'.
203 h [ digits ]
205 Remove a trailing pathname component, shortening the path by one
206 directory level: this is the `head' of the pathname. This works
207 like `dirname'. If the h is followed immediately (with no spa‐
208 ces or other separator) by any number of decimal digits, and the
209 value of the resulting number is non-zero, that number of lead‐
210 ing components is preserved instead of the final component being
211 removed. In an absolute path the leading `/' is the first com‐
212 ponent, so, for example, if var=/my/path/to/something, then
213 ${var:h3} substitutes /my/path. Consecutive `/'s are treated
214 the same as a single `/'. In parameter substitution, digits may
215 only be used if the expression is in braces, so for example the
216 short form substitution $var:h2 is treated as ${var:h}2, not as
217 ${var:h2}. No restriction applies to the use of digits in his‐
218 tory substitution or globbing qualifiers. If more components
219 are requested than are present, the entire path is substituted
220 (so this does not trigger a `failed modifier' error in history
221 expansion).
222 l Convert the words to all lowercase.
224 p Print the new command but do not execute it. Only works with
226 history expansion.
227 P Turn a file name into an absolute path, like realpath(3). The
229 resulting path will be absolute, will refer to the same direc‐
230 tory entry as the input filename, and none of its components
231 will be symbolic links or equal to `.' or `..'.
232 Unlike realpath(3), non-existent trailing components are permit‐
234 ted and preserved.
235 q Quote the substituted words, escaping further substitutions.
237 Works with history expansion and parameter expansion, though for
238 parameters it is only useful if the resulting text is to be
239 re-evaluated such as by eval.
240 Q Remove one level of quotes from the substituted words.
242 r Remove a filename extension leaving the root name. Strings with
244 no filename extension are not altered. A filename extension is
245 a `.' followed by any number of characters (including zero) that
246 are neither `.' nor `/' and that continue to the end of the
247 string. For example, the extension of `foo.orig.c' is `.c', and
248 `dir.c/foo' has no extension.
249 s/l/r[/]
251 Substitute r for l as described below. The substitution is done
252 only for the first string that matches l. For arrays and for
253 filename generation, this applies to each word of the expanded
254 text. See below for further notes on substitutions.
255 The forms `gs/l/r' and `s/l/r/:G' perform global substitution,
257 i.e. substitute every occurrence of r for l. Note that the g or
258 :G must appear in exactly the position shown.
259 See further notes on this form of substitution below.
261 & Repeat the previous s substitution. Like s, may be preceded im‐
263 mediately by a g. In parameter expansion the & must appear in‐
264 side braces, and in filename generation it must be quoted with a
265 backslash.
266 t [ digits ]
268 Remove all leading pathname components, leaving the final compo‐
269 nent (tail). This works like `basename'. Any trailing slashes
270 are first removed. Decimal digits are handled as described
271 above for (h), but in this case that number of trailing compo‐
272 nents is preserved instead of the default 1; 0 is treated the
273 same as 1.
274 u Convert the words to all uppercase.
276 x Like q, but break into words at whitespace. Does not work with
278 parameter expansion.
279 The s/l/r/ substitution works as follows. By default the left-hand
281 side of substitutions are not patterns, but character strings. Any
282 character can be used as the delimiter in place of `/'. A backslash
283 quotes the delimiter character. The character `&', in the
284 right-hand-side r, is replaced by the text from the left-hand-side l.
285 The `&' can be quoted with a backslash. A null l uses the previous
286 string either from the previous l or from the contextual scan string s
287 from `!?s'. You can omit the rightmost delimiter if a newline immedi‐
288 ately follows r; the rightmost `?' in a context scan can similarly be
289 omitted. Note the same record of the last l and r is maintained across
290 all forms of expansion.
291 Note that if a `&' is used within glob qualifiers an extra backslash is
293 needed as a & is a special character in this case.
294 Also note that the order of expansions affects the interpretation of l
296 and r. When used in a history expansion, which occurs before any other
297 expansions, l and r are treated as literal strings (except as explained
298 for HIST_SUBST_PATTERN below). When used in parameter expansion, the
299 replacement of r into the parameter's value is done first, and then any
300 additional process, parameter, command, arithmetic, or brace references
301 are applied, which may evaluate those substitutions and expansions more
302 than once if l appears more than once in the starting value. When used
303 in a glob qualifier, any substitutions or expansions are performed once
304 at the time the qualifier is parsed, even before the `:s' expression
305 itself is divided into l and r sides.
306 If the option HIST_SUBST_PATTERN is set, l is treated as a pattern of
308 the usual form described in the section FILENAME GENERATION below.
309 This can be used in all the places where modifiers are available; note,
310 however, that in globbing qualifiers parameter substitution has already
311 taken place, so parameters in the replacement string should be quoted
312 to ensure they are replaced at the correct time. Note also that com‐
313 plicated patterns used in globbing qualifiers may need the extended
314 glob qualifier notation (#q:s/.../.../) in order for the shell to rec‐
315 ognize the expression as a glob qualifier. Further, note that bad pat‐
316 terns in the substitution are not subject to the NO_BAD_PATTERN option
317 so will cause an error.
318 When HIST_SUBST_PATTERN is set, l may start with a # to indicate that
320 the pattern must match at the start of the string to be substituted,
321 and a % may appear at the start or after an # to indicate that the pat‐
322 tern must match at the end of the string to be substituted. The % or #
323 may be quoted with two backslashes.
324 For example, the following piece of filename generation code with the
326 EXTENDED_GLOB option:
327 print -r -- *.c(#q:s/#%(#b)s(*).c/'S${match[1]}.C'/)
329 takes the expansion of *.c and applies the glob qualifiers in the
331 (#q...) expression, which consists of a substitution modifier anchored
332 to the start and end of each word (#%). This turns on backreferences
333 ((#b)), so that the parenthesised subexpression is available in the re‐
334 placement string as ${match[1]}. The replacement string is quoted so
335 that the parameter is not substituted before the start of filename gen‐
336 eration.
337 The following f, F, w and W modifiers work only with parameter expan‐
339 sion and filename generation. They are listed here to provide a single
340 point of reference for all modifiers.
341 f Repeats the immediately (without a colon) following modifier un‐
343 til the resulting word doesn't change any more.
344 F:expr:
346 Like f, but repeats only n times if the expression expr evalu‐
347 ates to n. Any character can be used instead of the `:'; if
348 `(', `[', or `{' is used as the opening delimiter, the closing
349 delimiter should be ')', `]', or `}', respectively.
350 w Makes the immediately following modifier work on each word in
352 the string.
353 W:sep: Like w but words are considered to be the parts of the string
355 that are separated by sep. Any character can be used instead of
356 the `:'; opening parentheses are handled specially, see above.
357
359 Each part of a command argument that takes the form `<(list)',
360 `>(list)' or `=(list)' is subject to process substitution. The expres‐
361 sion may be preceded or followed by other strings except that, to pre‐
362 vent clashes with commonly occurring strings and patterns, the last
363 form must occur at the start of a command argument, and the forms are
364 only expanded when first parsing command or assignment arguments.
365 Process substitutions may be used following redirection operators; in
366 this case, the substitution must appear with no trailing string.
367 Note that `<<(list)' is not a special syntax; it is equivalent to `<
369 <(list)', redirecting standard input from the result of process substi‐
370 tution. Hence all the following documentation applies. The second
371 form (with the space) is recommended for clarity.
372 In the case of the < or > forms, the shell runs the commands in list as
374 a subprocess of the job executing the shell command line. If the sys‐
375 tem supports the /dev/fd mechanism, the command argument is the name of
376 the device file corresponding to a file descriptor; otherwise, if the
377 system supports named pipes (FIFOs), the command argument will be a
378 named pipe. If the form with > is selected then writing on this spe‐
379 cial file will provide input for list. If < is used, then the file
380 passed as an argument will be connected to the output of the list
381 process. For example,
382 paste <(cut -f1 file1) <(cut -f3 file2) |
384 tee >(process1) >(process2) >/dev/null
385 cuts fields 1 and 3 from the files file1 and file2 respectively, pastes
387 the results together, and sends it to the processes process1 and
388 process2.
389 If =(...) is used instead of <(...), then the file passed as an argu‐
391 ment will be the name of a temporary file containing the output of the
392 list process. This may be used instead of the < form for a program
393 that expects to lseek (see lseek(2)) on the input file.
394 There is an optimisation for substitutions of the form =(<<<arg), where
396 arg is a single-word argument to the here-string redirection <<<. This
397 form produces a file name containing the value of arg after any substi‐
398 tutions have been performed. This is handled entirely within the cur‐
399 rent shell. This is effectively the reverse of the special form
400 $(<arg) which treats arg as a file name and replaces it with the file's
401 contents.
402 The = form is useful as both the /dev/fd and the named pipe implementa‐
404 tion of <(...) have drawbacks. In the former case, some programmes may
405 automatically close the file descriptor in question before examining
406 the file on the command line, particularly if this is necessary for se‐
407 curity reasons such as when the programme is running setuid. In the
408 second case, if the programme does not actually open the file, the sub‐
409 shell attempting to read from or write to the pipe will (in a typical
410 implementation, different operating systems may have different behav‐
411 iour) block for ever and have to be killed explicitly. In both cases,
412 the shell actually supplies the information using a pipe, so that pro‐
413 grammes that expect to lseek (see lseek(2)) on the file will not work.
414 Also note that the previous example can be more compactly and effi‐
416 ciently written (provided the MULTIOS option is set) as:
417 paste <(cut -f1 file1) <(cut -f3 file2) > >(process1) > >(process2)
419 The shell uses pipes instead of FIFOs to implement the latter two
421 process substitutions in the above example.
422 There is an additional problem with >(process); when this is attached
424 to an external command, the parent shell does not wait for process to
425 finish and hence an immediately following command cannot rely on the
426 results being complete. The problem and solution are the same as de‐
427 scribed in the section MULTIOS in zshmisc(1). Hence in a simplified
428 version of the example above:
429 paste <(cut -f1 file1) <(cut -f3 file2) > >(process)
431 (note that no MULTIOS are involved), process will be run asynchronously
433 as far as the parent shell is concerned. The workaround is:
434 { paste <(cut -f1 file1) <(cut -f3 file2) } > >(process)
436 The extra processes here are spawned from the parent shell which will
438 wait for their completion.
439 Another problem arises any time a job with a substitution that requires
441 a temporary file is disowned by the shell, including the case where
442 `&!' or `&|' appears at the end of a command containing a substitution.
443 In that case the temporary file will not be cleaned up as the shell no
444 longer has any memory of the job. A workaround is to use a subshell,
445 for example,
446 (mycmd =(myoutput)) &!
448 as the forked subshell will wait for the command to finish then remove
450 the temporary file.
451 A general workaround to ensure a process substitution endures for an
453 appropriate length of time is to pass it as a parameter to an anonymous
454 shell function (a piece of shell code that is run immediately with
455 function scope). For example, this code:
456 () {
458 print File $1:
459 cat $1
460 } =(print This be the verse)
461 outputs something resembling the following
463 File /tmp/zsh6nU0kS:
465 This be the verse
466 The temporary file created by the process substitution will be deleted
468 when the function exits.
469
471 The character `$' is used to introduce parameter expansions. See zsh‐
472 param(1) for a description of parameters, including arrays, associative
473 arrays, and subscript notation to access individual array elements.
474 Note in particular the fact that words of unquoted parameters are not
476 automatically split on whitespace unless the option SH_WORD_SPLIT is
477 set; see references to this option below for more details. This is an
478 important difference from other shells. However, as in other shells,
479 null words are elided from unquoted parameters' expansions.
480 With default options, after the assignments:
482 array=("first word" "" "third word")
484 scalar="only word"
485 then $array substitutes two words, `first word' and `third word', and
487 $scalar substitutes a single word `only word'. Note that second ele‐
488 ment of array was elided. Scalar parameters can be elided too if their
489 value is null (empty). To avoid elision, use quoting as follows:
490 "$scalar" for scalars and "${array[@]}" or "${(@)array}" for arrays.
491 (The last two forms are equivalent.)
492 Parameter expansions can involve flags, as in `${(@kv)aliases}', and
494 other operators, such as `${PREFIX:-"/usr/local"}'. Parameter expan‐
495 sions can also be nested. These topics will be introduced below. The
496 full rules are complicated and are noted at the end.
497 In the expansions discussed below that require a pattern, the form of
499 the pattern is the same as that used for filename generation; see the
500 section `Filename Generation'. Note that these patterns, along with
501 the replacement text of any substitutions, are themselves subject to
502 parameter expansion, command substitution, and arithmetic expansion.
503 In addition to the following operations, the colon modifiers described
504 in the section `Modifiers' in the section `History Expansion' can be
505 applied: for example, ${i:s/foo/bar/} performs string substitution on
506 the expansion of parameter $i.
507 In the following descriptions, `word' refers to a single word substi‐
509 tuted on the command line, not necessarily a space delimited word.
510 ${name}
512 The value, if any, of the parameter name is substituted. The
513 braces are required if the expansion is to be followed by a let‐
514 ter, digit, or underscore that is not to be interpreted as part
515 of name. In addition, more complicated forms of substitution
516 usually require the braces to be present; exceptions, which only
517 apply if the option KSH_ARRAYS is not set, are a single sub‐
518 script or any colon modifiers appearing after the name, or any
519 of the characters `^', `=', `~', `#' or `+' appearing before the
520 name, all of which work with or without braces.
521 If name is an array parameter, and the KSH_ARRAYS option is not
523 set, then the value of each element of name is substituted, one
524 element per word. Otherwise, the expansion results in one word
525 only; with KSH_ARRAYS, this is the first element of an array.
526 No field splitting is done on the result unless the
527 SH_WORD_SPLIT option is set. See also the flags = and
528 s:string:.
529 ${+name}
531 If name is the name of a set parameter `1' is substituted, oth‐
532 erwise `0' is substituted.
533 ${name-word}
535 ${name:-word}
536 If name is set, or in the second form is non-null, then substi‐
537 tute its value; otherwise substitute word. In the second form
538 name may be omitted, in which case word is always substituted.
539 ${name+word}
541 ${name:+word}
542 If name is set, or in the second form is non-null, then substi‐
543 tute word; otherwise substitute nothing.
544 ${name=word}
546 ${name:=word}
547 ${name::=word}
548 In the first form, if name is unset then set it to word; in the
549 second form, if name is unset or null then set it to word; and
550 in the third form, unconditionally set name to word. In all
551 forms, the value of the parameter is then substituted.
552 ${name?word}
554 ${name:?word}
555 In the first form, if name is set, or in the second form if name
556 is both set and non-null, then substitute its value; otherwise,
557 print word and exit from the shell. Interactive shells instead
558 return to the prompt. If word is omitted, then a standard mes‐
559 sage is printed.
560 In any of the above expressions that test a variable and substitute an
562 alternate word, note that you can use standard shell quoting in the
563 word value to selectively override the splitting done by the
564 SH_WORD_SPLIT option and the = flag, but not splitting by the s:string:
565 flag.
566 In the following expressions, when name is an array and the substitu‐
568 tion is not quoted, or if the `(@)' flag or the name[@] syntax is used,
569 matching and replacement is performed on each array element separately.
570 ${name#pattern}
572 ${name##pattern}
573 If the pattern matches the beginning of the value of name, then
574 substitute the value of name with the matched portion deleted;
575 otherwise, just substitute the value of name. In the first
576 form, the smallest matching pattern is preferred; in the second
577 form, the largest matching pattern is preferred.
578 ${name%pattern}
580 ${name%%pattern}
581 If the pattern matches the end of the value of name, then sub‐
582 stitute the value of name with the matched portion deleted; oth‐
583 erwise, just substitute the value of name. In the first form,
584 the smallest matching pattern is preferred; in the second form,
585 the largest matching pattern is preferred.
586 ${name:#pattern}
588 If the pattern matches the value of name, then substitute the
589 empty string; otherwise, just substitute the value of name. If
590 name is an array the matching array elements are removed (use
591 the `(M)' flag to remove the non-matched elements).
592 ${name:|arrayname}
594 If arrayname is the name (N.B., not contents) of an array vari‐
595 able, then any elements contained in arrayname are removed from
596 the substitution of name. If the substitution is scalar, either
597 because name is a scalar variable or the expression is quoted,
598 the elements of arrayname are instead tested against the entire
599 expression.
600 ${name:*arrayname}
602 Similar to the preceding substitution, but in the opposite
603 sense, so that entries present in both the original substitution
604 and as elements of arrayname are retained and others removed.
605 ${name:^arrayname}
607 ${name:^^arrayname}
608 Zips two arrays, such that the output array is twice as long as
609 the shortest (longest for `:^^') of name and arrayname, with the
610 elements alternatingly being picked from them. For `:^', if one
611 of the input arrays is longer, the output will stop when the end
612 of the shorter array is reached. Thus,
613 a=(1 2 3 4); b=(a b); print ${a:^b}
615 will output `1 a 2 b'. For `:^^', then the input is repeated
617 until all of the longer array has been used up and the above
618 will output `1 a 2 b 3 a 4 b'.
619 Either or both inputs may be a scalar, they will be treated as
621 an array of length 1 with the scalar as the only element. If ei‐
622 ther array is empty, the other array is output with no extra el‐
623 ements inserted.
624 Currently the following code will output `a b' and `1' as two
626 separate elements, which can be unexpected. The second print
627 provides a workaround which should continue to work if this is
628 changed.
629 a=(a b); b=(1 2); print -l "${a:^b}"; print -l "${${a:^b}}"
631 ${name:offset}
633 ${name:offset:length}
634 This syntax gives effects similar to parameter subscripting in
635 the form $name[start,end], but is compatible with other shells;
636 note that both offset and length are interpreted differently
637 from the components of a subscript.
638 If offset is non-negative, then if the variable name is a scalar
640 substitute the contents starting offset characters from the
641 first character of the string, and if name is an array substi‐
642 tute elements starting offset elements from the first element.
643 If length is given, substitute that many characters or elements,
644 otherwise the entire rest of the scalar or array.
645 A positive offset is always treated as the offset of a character
647 or element in name from the first character or element of the
648 array (this is different from native zsh subscript notation).
649 Hence 0 refers to the first character or element regardless of
650 the setting of the option KSH_ARRAYS.
651 A negative offset counts backwards from the end of the scalar or
653 array, so that -1 corresponds to the last character or element,
654 and so on.
655 When positive, length counts from the offset position toward the
657 end of the scalar or array. When negative, length counts back
658 from the end. If this results in a position smaller than off‐
659 set, a diagnostic is printed and nothing is substituted.
660 The option MULTIBYTE is obeyed, i.e. the offset and length count
662 multibyte characters where appropriate.
663 offset and length undergo the same set of shell substitutions as
665 for scalar assignment; in addition, they are then subject to
666 arithmetic evaluation. Hence, for example
667 print ${foo:3}
669 print ${foo: 1 + 2}
670 print ${foo:$(( 1 + 2))}
671 print ${foo:$(echo 1 + 2)}
672 all have the same effect, extracting the string starting at the
674 fourth character of $foo if the substitution would otherwise re‐
675 turn a scalar, or the array starting at the fourth element if
676 $foo would return an array. Note that with the option KSH_AR‐
677 RAYS $foo always returns a scalar (regardless of the use of the
678 offset syntax) and a form such as ${foo[*]:3} is required to ex‐
679 tract elements of an array named foo.
680 If offset is negative, the - may not appear immediately after
682 the : as this indicates the ${name:-word} form of substitution.
683 Instead, a space may be inserted before the -. Furthermore,
684 neither offset nor length may begin with an alphabetic character
685 or & as these are used to indicate history-style modifiers. To
686 substitute a value from a variable, the recommended approach is
687 to precede it with a $ as this signifies the intention (parame‐
688 ter substitution can easily be rendered unreadable); however, as
689 arithmetic substitution is performed, the expression ${var:
690 offs} does work, retrieving the offset from $offs.
691 For further compatibility with other shells there is a special
693 case for array offset 0. This usually accesses the first ele‐
694 ment of the array. However, if the substitution refers to the
695 positional parameter array, e.g. $@ or $*, then offset 0 instead
696 refers to $0, offset 1 refers to $1, and so on. In other words,
697 the positional parameter array is effectively extended by
698 prepending $0. Hence ${*:0:1} substitutes $0 and ${*:1:1} sub‐
699 stitutes $1.
700 ${name/pattern/repl}
702 ${name//pattern/repl}
703 ${name:/pattern/repl}
704 Replace the longest possible match of pattern in the expansion
705 of parameter name by string repl. The first form replaces just
706 the first occurrence, the second form all occurrences, and the
707 third form replaces only if pattern matches the entire string.
708 Both pattern and repl are subject to double-quoted substitution,
709 so that expressions like ${name/$opat/$npat} will work, but obey
710 the usual rule that pattern characters in $opat are not treated
711 specially unless either the option GLOB_SUBST is set, or $opat
712 is instead substituted as ${~opat}.
713 The pattern may begin with a `#', in which case the pattern must
715 match at the start of the string, or `%', in which case it must
716 match at the end of the string, or `#%' in which case the pat‐
717 tern must match the entire string. The repl may be an empty
718 string, in which case the final `/' may also be omitted. To
719 quote the final `/' in other cases it should be preceded by a
720 single backslash; this is not necessary if the `/' occurs inside
721 a substituted parameter. Note also that the `#', `%' and `#%
722 are not active if they occur inside a substituted parameter,
723 even at the start.
724 If, after quoting rules apply, ${name} expands to an array, the
726 replacements act on each element individually. Note also the
727 effect of the I and S parameter expansion flags below; however,
728 the flags M, R, B, E and N are not useful.
729 For example,
731 foo="twinkle twinkle little star" sub="t*e" rep="spy"
733 print ${foo//${~sub}/$rep}
734 print ${(S)foo//${~sub}/$rep}
735 Here, the `~' ensures that the text of $sub is treated as a pat‐
737 tern rather than a plain string. In the first case, the longest
738 match for t*e is substituted and the result is `spy star', while
739 in the second case, the shortest matches are taken and the re‐
740 sult is `spy spy lispy star'.
741 ${#spec}
743 If spec is one of the above substitutions, substitute the length
744 in characters of the result instead of the result itself. If
745 spec is an array expression, substitute the number of elements
746 of the result. This has the side-effect that joining is skipped
747 even in quoted forms, which may affect other sub-expressions in
748 spec. Note that `^', `=', and `~', below, must appear to the
749 left of `#' when these forms are combined.
750 If the option POSIX_IDENTIFIERS is not set, and spec is a simple
752 name, then the braces are optional; this is true even for spe‐
753 cial parameters so e.g. $#- and $#* take the length of the
754 string $- and the array $* respectively. If POSIX_IDENTIFIERS
755 is set, then braces are required for the # to be treated in this
756 fashion.
757 ${^spec}
759 ${^^spec}
760 Turn on the RC_EXPAND_PARAM option for the evaluation of spec;
761 if the `^' is doubled, turn it off. When this option is set,
762 array expansions of the form foo${xx}bar, where the parameter xx
763 is set to (a b c), are substituted with `fooabar foobbar
764 foocbar' instead of the default `fooa b cbar'. Note that an
765 empty array will therefore cause all arguments to be removed.
766 Internally, each such expansion is converted into the equivalent
768 list for brace expansion. E.g., ${^var} becomes
769 {$var[1],$var[2],...}, and is processed as described in the sec‐
770 tion `Brace Expansion' below: note, however, the expansion hap‐
771 pens immediately, with any explicit brace expansion happening
772 later. If word splitting is also in effect the $var[N] may
773 themselves be split into different list elements.
774 ${=spec}
776 ${==spec}
777 Perform word splitting using the rules for SH_WORD_SPLIT during
778 the evaluation of spec, but regardless of whether the parameter
779 appears in double quotes; if the `=' is doubled, turn it off.
780 This forces parameter expansions to be split into separate words
781 before substitution, using IFS as a delimiter. This is done by
782 default in most other shells.
783 Note that splitting is applied to word in the assignment forms
785 of spec before the assignment to name is performed. This af‐
786 fects the result of array assignments with the A flag.
787 ${~spec}
789 ${~~spec}
790 Turn on the GLOB_SUBST option for the evaluation of spec; if the
791 `~' is doubled, turn it off. When this option is set, the
792 string resulting from the expansion will be interpreted as a
793 pattern anywhere that is possible, such as in filename expansion
794 and filename generation and pattern-matching contexts like the
795 right hand side of the `=' and `!=' operators in conditions.
796 In nested substitutions, note that the effect of the ~ applies
798 to the result of the current level of substitution. A surround‐
799 ing pattern operation on the result may cancel it. Hence, for
800 example, if the parameter foo is set to *, ${~foo//\*/*.c} is
801 substituted by the pattern *.c, which may be expanded by file‐
802 name generation, but ${${~foo}//\*/*.c} substitutes to the
803 string *.c, which will not be further expanded.
804 If a ${...} type parameter expression or a $(...) type command substi‐
806 tution is used in place of name above, it is expanded first and the re‐
807 sult is used as if it were the value of name. Thus it is possible to
808 perform nested operations: ${${foo#head}%tail} substitutes the value
809 of $foo with both `head' and `tail' deleted. The form with $(...) is
810 often useful in combination with the flags described next; see the ex‐
811 amples below. Each name or nested ${...} in a parameter expansion may
812 also be followed by a subscript expression as described in Array Param‐
813 eters in zshparam(1).
814 Note that double quotes may appear around nested expressions, in which
816 case only the part inside is treated as quoted; for example,
817 ${(f)"$(foo)"} quotes the result of $(foo), but the flag `(f)' (see be‐
818 low) is applied using the rules for unquoted expansions. Note further
819 that quotes are themselves nested in this context; for example, in
820 "${(@f)"$(foo)"}", there are two sets of quotes, one surrounding the
821 whole expression, the other (redundant) surrounding the $(foo) as be‐
822 fore.
823 Parameter Expansion Flags
825 If the opening brace is directly followed by an opening parenthesis,
826 the string up to the matching closing parenthesis will be taken as a
827 list of flags. In cases where repeating a flag is meaningful, the rep‐
828 etitions need not be consecutive; for example, `(q%q%q)' means the same
829 thing as the more readable `(%%qqq)'. The following flags are sup‐
830 ported:
831 # Evaluate the resulting words as numeric expressions and inter‐
833 pret these as character codes. Output the corresponding charac‐
834 ters. Note that this form is entirely distinct from use of the
835 # without parentheses.
836 If the MULTIBYTE option is set and the number is greater than
838 127 (i.e. not an ASCII character) it is treated as a Unicode
839 character.
840 % Expand all % escapes in the resulting words in the same way as
842 in prompts (see EXPANSION OF PROMPT SEQUENCES in zshmisc(1)). If
843 this flag is given twice, full prompt expansion is done on the
844 resulting words, depending on the setting of the PROMPT_PERCENT,
845 PROMPT_SUBST and PROMPT_BANG options.
846 @ In double quotes, array elements are put into separate words.
848 E.g., `"${(@)foo}"' is equivalent to `"${foo[@]}"' and
849 `"${(@)foo[1,2]}"' is the same as `"$foo[1]" "$foo[2]"'. This
850 is distinct from field splitting by the f, s or z flags, which
851 still applies within each array element.
852 A Convert the substitution into an array expression, even if it
854 otherwise would be scalar. This has lower precedence than sub‐
855 scripting, so one level of nested expansion is required in order
856 that subscripts apply to array elements. Thus ${${(A)name}[1]}
857 yields the full value of name when name is scalar.
858 This assigns an array parameter with `${...=...}', `${...:=...}'
860 or `${...::=...}'. If this flag is repeated (as in `AA'), as‐
861 signs an associative array parameter. Assignment is made before
862 sorting or padding; if field splitting is active, the word part
863 is split before assignment. The name part may be a subscripted
864 range for ordinary arrays; when assigning an associative array,
865 the word part must be converted to an array, for example by us‐
866 ing `${(AA)=name=...}' to activate field splitting.
867 Surrounding context such as additional nesting or use of the
869 value in a scalar assignment may cause the array to be joined
870 back into a single string again.
871 a Sort in array index order; when combined with `O' sort in re‐
873 verse array index order. Note that `a' is therefore equivalent
874 to the default but `Oa' is useful for obtaining an array's ele‐
875 ments in reverse order.
876 b Quote with backslashes only characters that are special to pat‐
878 tern matching. This is useful when the contents of the variable
879 are to be tested using GLOB_SUBST, including the ${~...} switch.
880 Quoting using one of the q family of flags does not work for
882 this purpose since quotes are not stripped from non-pattern
883 characters by GLOB_SUBST. In other words,
884 pattern=${(q)str}
886 [[ $str = ${~pattern} ]]
887 works if $str is `a*b' but not if it is `a b', whereas
889 pattern=${(b)str}
891 [[ $str = ${~pattern} ]]
892 is always true for any possible value of $str.
894 c With ${#name}, count the total number of characters in an array,
896 as if the elements were concatenated with spaces between them.
897 This is not a true join of the array, so other expressions used
898 with this flag may have an effect on the elements of the array
899 before it is counted.
900 C Capitalize the resulting words. `Words' in this case refers to
902 sequences of alphanumeric characters separated by non-alphanu‐
903 merics, not to words that result from field splitting.
904 D Assume the string or array elements contain directories and at‐
906 tempt to substitute the leading part of these by names. The re‐
907 mainder of the path (the whole of it if the leading part was not
908 substituted) is then quoted so that the whole string can be used
909 as a shell argument. This is the reverse of `~' substitution:
910 see the section FILENAME EXPANSION below.
911 e Perform single word shell expansions, namely parameter expan‐
913 sion, command substitution and arithmetic expansion, on the re‐
914 sult. Such expansions can be nested but too deep recursion may
915 have unpredictable effects.
916 f Split the result of the expansion at newlines. This is a short‐
918 hand for `ps:\n:'.
919 F Join the words of arrays together using newline as a separator.
921 This is a shorthand for `pj:\n:'.
922 g:opts:
924 Process escape sequences like the echo builtin when no options
925 are given (g::). With the o option, octal escapes don't take a
926 leading zero. With the c option, sequences like `^X' are also
927 processed. With the e option, processes `\M-t' and similar se‐
928 quences like the print builtin. With both of the o and e op‐
929 tions, behaves like the print builtin except that in none of
930 these modes is `\c' interpreted.
931 i Sort case-insensitively. May be combined with `n' or `O'.
933 k If name refers to an associative array, substitute the keys (el‐
935 ement names) rather than the values of the elements. Used with
936 subscripts (including ordinary arrays), force indices or keys to
937 be substituted even if the subscript form refers to values.
938 However, this flag may not be combined with subscript ranges.
939 With the KSH_ARRAYS option a subscript `[*]' or `[@]' is needed
940 to operate on the whole array, as usual.
941 L Convert all letters in the result to lower case.
943 n Sort decimal integers numerically; if the first differing char‐
945 acters of two test strings are not digits, sorting is lexical.
946 `+' and `-' are not treated specially; they are treated as any
947 other non-digit. Integers with more initial zeroes are sorted
948 before those with fewer or none. Hence the array `foo+24 foo1
949 foo02 foo2 foo3 foo20 foo23' is sorted into the order shown.
950 May be combined with `i' or `O'.
951 - As n, but a leading minus sign indicates a negative decimal in‐
953 teger. A leading minus sign not followed by an integer does not
954 trigger numeric sorting. Note that `+' signs are not handled
955 specially (this may change in the future).
956 o Sort the resulting words in ascending order; if this appears on
958 its own the sorting is lexical and case-sensitive (unless the
959 locale renders it case-insensitive). Sorting in ascending order
960 is the default for other forms of sorting, so this is ignored if
961 combined with `a', `i', `n' or `-'.
962 O Sort the resulting words in descending order; `O' without `a',
964 `i', `n' or `-' sorts in reverse lexical order. May be combined
965 with `a', `i', `n' or `-' to reverse the order of sorting.
966 P This forces the value of the parameter name to be interpreted as
968 a further parameter name, whose value will be used where appro‐
969 priate. Note that flags set with one of the typeset family of
970 commands (in particular case transformations) are not applied to
971 the value of name used in this fashion.
972 If used with a nested parameter or command substitution, the re‐
974 sult of that will be taken as a parameter name in the same way.
975 For example, if you have `foo=bar' and `bar=baz', the strings
976 ${(P)foo}, ${(P)${foo}}, and ${(P)$(echo bar)} will be expanded
977 to `baz'.
978 Likewise, if the reference is itself nested, the expression with
980 the flag is treated as if it were directly replaced by the pa‐
981 rameter name. It is an error if this nested substitution pro‐
982 duces an array with more than one word. For example, if
983 `name=assoc' where the parameter assoc is an associative array,
984 then `${${(P)name}[elt]}' refers to the element of the associa‐
985 tive subscripted `elt'.
986 q Quote characters that are special to the shell in the resulting
988 words with backslashes; unprintable or invalid characters are
989 quoted using the $'\NNN' form, with separate quotes for each
990 octet.
991 If this flag is given twice, the resulting words are quoted in
993 single quotes and if it is given three times, the words are
994 quoted in double quotes; in these forms no special handling of
995 unprintable or invalid characters is attempted. If the flag is
996 given four times, the words are quoted in single quotes preceded
997 by a $. Note that in all three of these forms quoting is done
998 unconditionally, even if this does not change the way the re‐
999 sulting string would be interpreted by the shell.
1000 If a q- is given (only a single q may appear), a minimal form of
1002 single quoting is used that only quotes the string if needed to
1003 protect special characters. Typically this form gives the most
1004 readable output.
1005 If a q+ is given, an extended form of minimal quoting is used
1007 that causes unprintable characters to be rendered using $'...'.
1008 This quoting is similar to that used by the output of values by
1009 the typeset family of commands.
1010 Q Remove one level of quotes from the resulting words.
1012 t Use a string describing the type of the parameter where the
1014 value of the parameter would usually appear. This string con‐
1015 sists of keywords separated by hyphens (`-'). The first keyword
1016 in the string describes the main type, it can be one of
1017 `scalar', `array', `integer', `float' or `association'. The
1018 other keywords describe the type in more detail:
1019 local for local parameters
1021 left for left justified parameters
1023 right_blanks
1025 for right justified parameters with leading blanks
1026 right_zeros
1028 for right justified parameters with leading zeros
1029 lower for parameters whose value is converted to all lower case
1031 when it is expanded
1032 upper for parameters whose value is converted to all upper case
1034 when it is expanded
1035 readonly
1037 for readonly parameters
1038 tag for tagged parameters
1040 tied for parameters tied to another parameter in the manner of
1042 PATH (colon-separated list) and path (array), whether
1043 these are special parameters or user-defined with `type‐
1044 set -T'
1045 export for exported parameters
1047 unique for arrays which keep only the first occurrence of dupli‐
1049 cated values
1050 hide for parameters with the `hide' flag
1052 hideval
1054 for parameters with the `hideval' flag
1055 special
1057 for special parameters defined by the shell
1058 u Expand only the first occurrence of each unique word.
1060 U Convert all letters in the result to upper case.
1062 v Used with k, substitute (as two consecutive words) both the key
1064 and the value of each associative array element. Used with sub‐
1065 scripts, force values to be substituted even if the subscript
1066 form refers to indices or keys.
1067 V Make any special characters in the resulting words visible.
1069 w With ${#name}, count words in arrays or strings; the s flag may
1071 be used to set a word delimiter.
1072 W Similar to w with the difference that empty words between re‐
1074 peated delimiters are also counted.
1075 X With this flag, parsing errors occurring with the Q, e and #
1077 flags or the pattern matching forms such as `${name#pattern}'
1078 are reported. Without the flag, errors are silently ignored.
1079 z Split the result of the expansion into words using shell parsing
1081 to find the words, i.e. taking into account any quoting in the
1082 value. Comments are not treated specially but as ordinary
1083 strings, similar to interactive shells with the INTERACTIVE_COM‐
1084 MENTS option unset (however, see the Z flag below for related
1085 options)
1086 Note that this is done very late, even later than the `(s)'
1088 flag. So to access single words in the result use nested expan‐
1089 sions as in `${${(z)foo}[2]}'. Likewise, to remove the quotes in
1090 the resulting words use `${(Q)${(z)foo}}'.
1091 0 Split the result of the expansion on null bytes. This is a
1093 shorthand for `ps:\0:'.
1094 The following flags (except p) are followed by one or more arguments as
1096 shown. Any character, or the matching pairs `(...)', `{...}', `[...]',
1097 or `<...>', may be used in place of a colon as delimiters, but note
1098 that when a flag takes more than one argument, a matched pair of delim‐
1099 iters must surround each argument.
1100 p Recognize the same escape sequences as the print builtin in
1102 string arguments to any of the flags described below that follow
1103 this argument.
1104 Alternatively, with this option string arguments may be in the
1106 form $var in which case the value of the variable is substi‐
1107 tuted. Note this form is strict; the string argument does not
1108 undergo general parameter expansion.
1109 For example,
1111 sep=:
1113 val=a:b:c
1114 print ${(ps.$sep.)val}
1115 splits the variable on a :.
1117 ~ Strings inserted into the expansion by any of the flags below
1119 are to be treated as patterns. This applies to the string argu‐
1120 ments of flags that follow ~ within the same set of parentheses.
1121 Compare with ~ outside parentheses, which forces the entire sub‐
1122 stituted string to be treated as a pattern. Hence, for example,
1123 [[ "?" = ${(~j.|.)array} ]]
1125 treats `|' as a pattern and succeeds if and only if $array con‐
1127 tains the string `?' as an element. The ~ may be repeated to
1128 toggle the behaviour; its effect only lasts to the end of the
1129 parenthesised group.
1130 j:string:
1132 Join the words of arrays together using string as a separator.
1133 Note that this occurs before field splitting by the s:string:
1134 flag or the SH_WORD_SPLIT option.
1135 l:expr::string1::string2:
1137 Pad the resulting words on the left. Each word will be trun‐
1138 cated if required and placed in a field expr characters wide.
1139 The arguments :string1: and :string2: are optional; neither, the
1141 first, or both may be given. Note that the same pairs of delim‐
1142 iters must be used for each of the three arguments. The space
1143 to the left will be filled with string1 (concatenated as often
1144 as needed) or spaces if string1 is not given. If both string1
1145 and string2 are given, string2 is inserted once directly to the
1146 left of each word, truncated if necessary, before string1 is
1147 used to produce any remaining padding.
1148 If either of string1 or string2 is present but empty, i.e. there
1150 are two delimiters together at that point, the first character
1151 of $IFS is used instead.
1152 If the MULTIBYTE option is in effect, the flag m may also be
1154 given, in which case widths will be used for the calculation of
1155 padding; otherwise individual multibyte characters are treated
1156 as occupying one unit of width.
1157 If the MULTIBYTE option is not in effect, each byte in the
1159 string is treated as occupying one unit of width.
1160 Control characters are always assumed to be one unit wide; this
1162 allows the mechanism to be used for generating repetitions of
1163 control characters.
1164 m Only useful together with one of the flags l or r or with the #
1166 length operator when the MULTIBYTE option is in effect. Use the
1167 character width reported by the system in calculating how much
1168 of the string it occupies or the overall length of the string.
1169 Most printable characters have a width of one unit, however cer‐
1170 tain Asian character sets and certain special effects use wider
1171 characters; combining characters have zero width. Non-printable
1172 characters are arbitrarily counted as zero width; how they would
1173 actually be displayed will vary.
1174 If the m is repeated, the character either counts zero (if it
1176 has zero width), else one. For printable character strings this
1177 has the effect of counting the number of glyphs (visibly sepa‐
1178 rate characters), except for the case where combining characters
1179 themselves have non-zero width (true in certain alphabets).
1180 r:expr::string1::string2:
1182 As l, but pad the words on the right and insert string2 immedi‐
1183 ately to the right of the string to be padded.
1184 Left and right padding may be used together. In this case the
1186 strategy is to apply left padding to the first half width of
1187 each of the resulting words, and right padding to the second
1188 half. If the string to be padded has odd width the extra pad‐
1189 ding is applied on the left.
1190 s:string:
1192 Force field splitting at the separator string. Note that a
1193 string of two or more characters means that all of them must
1194 match in sequence; this differs from the treatment of two or
1195 more characters in the IFS parameter. See also the = flag and
1196 the SH_WORD_SPLIT option. An empty string may also be given in
1197 which case every character will be a separate element.
1198 For historical reasons, the usual behaviour that empty array el‐
1200 ements are retained inside double quotes is disabled for arrays
1201 generated by splitting; hence the following:
1202 line="one::three"
1204 print -l "${(s.:.)line}"
1205 produces two lines of output for one and three and elides the
1207 empty field. To override this behaviour, supply the `(@)' flag
1208 as well, i.e. "${(@s.:.)line}".
1209 Z:opts:
1211 As z but takes a combination of option letters between a follow‐
1212 ing pair of delimiter characters. With no options the effect is
1213 identical to z. The following options are available:
1214 (Z+c+) causes comments to be parsed as a string and retained;
1216 any field in the resulting array beginning with an un‐
1217 quoted comment character is a comment.
1218 (Z+C+) causes comments to be parsed and removed. The rule for
1220 comments is standard: anything between a word starting
1221 with the third character of $HISTCHARS, default #, up to
1222 the next newline is a comment.
1223 (Z+n+) causes unquoted newlines to be treated as ordinary white‐
1225 space, else they are treated as if they are shell code
1226 delimiters and converted to semicolons.
1227 Options are combined within the same set of delimiters, e.g.
1229 (Z+Cn+).
1230 _:flags:
1232 The underscore (_) flag is reserved for future use. As of this
1233 revision of zsh, there are no valid flags; anything following an
1234 underscore, other than an empty pair of delimiters, is treated
1235 as an error, and the flag itself has no effect.
1236 The following flags are meaningful with the ${...#...} or ${...%...}
1238 forms. The S, I, and * flags may also be used with the ${.../...}
1239 forms.
1240 S With # or ##, search for the match that starts closest to the
1242 start of the string (a `substring match'). Of all matches at a
1243 particular position, # selects the shortest and ## the longest:
1244 % str="aXbXc"
1246 % echo ${(S)str#X*}
1247 abXc
1248 % echo ${(S)str##X*}
1249 a
1250 %
1251 With % or %%, search for the match that starts closest to the
1253 end of the string:
1254 % str="aXbXc"
1256 % echo ${(S)str%X*}
1257 aXbc
1258 % echo ${(S)str%%X*}
1259 aXb
1260 %
1261 (Note that % and %% don't search for the match that ends closest
1263 to the end of the string, as one might expect.)
1264 With substitution via ${.../...} or ${...//...}, specifies
1266 non-greedy matching, i.e. that the shortest instead of the long‐
1267 est match should be replaced:
1268 % str="abab"
1270 % echo ${str/*b/_}
1271 _
1272 % echo ${(S)str/*b/_}
1273 _ab
1274 %
1275 I:expr:
1277 Search the exprth match (where expr evaluates to a number).
1278 This only applies when searching for substrings, either with the
1279 S flag, or with ${.../...} (only the exprth match is substi‐
1280 tuted) or ${...//...} (all matches from the exprth on are sub‐
1281 stituted). The default is to take the first match.
1282 The exprth match is counted such that there is either one or
1284 zero matches from each starting position in the string, although
1285 for global substitution matches overlapping previous replace‐
1286 ments are ignored. With the ${...%...} and ${...%%...} forms,
1287 the starting position for the match moves backwards from the end
1288 as the index increases, while with the other forms it moves for‐
1289 ward from the start.
1290 Hence with the string
1292 which switch is the right switch for Ipswich?
1293 substitutions of the form ${(SI:N:)string#w*ch} as N increases
1294 from 1 will match and remove `which', `witch', `witch' and
1295 `wich'; the form using `##' will match and remove `which switch
1296 is the right switch for Ipswich', `witch is the right switch for
1297 Ipswich', `witch for Ipswich' and `wich'. The form using `%'
1298 will remove the same matches as for `#', but in reverse order,
1299 and the form using `%%' will remove the same matches as for `##'
1300 in reverse order.
1301 * Enable EXTENDED_GLOB for substitution via ${.../...} or
1303 ${...//...}. Note that `**' does not disable extendedglob.
1304 B Include the index of the beginning of the match in the result.
1306 E Include the index one character past the end of the match in the
1308 result (note this is inconsistent with other uses of parameter
1309 index).
1310 M Include the matched portion in the result.
1312 N Include the length of the match in the result.
1314 R Include the unmatched portion in the result (the Rest).
1316 Rules
1318 Here is a summary of the rules for substitution; this assumes that
1319 braces are present around the substitution, i.e. ${...}. Some particu‐
1320 lar examples are given below. Note that the Zsh Development Group ac‐
1321 cepts no responsibility for any brain damage which may occur during the
1322 reading of the following rules.
1323 1. Nested substitution
1325 If multiple nested ${...} forms are present, substitution is
1326 performed from the inside outwards. At each level, the substi‐
1327 tution takes account of whether the current value is a scalar or
1328 an array, whether the whole substitution is in double quotes,
1329 and what flags are supplied to the current level of substitu‐
1330 tion, just as if the nested substitution were the outermost.
1331 The flags are not propagated up to enclosing substitutions; the
1332 nested substitution will return either a scalar or an array as
1333 determined by the flags, possibly adjusted for quoting. All the
1334 following steps take place where applicable at all levels of
1335 substitution.
1336 Note that, unless the `(P)' flag is present, the flags and any
1338 subscripts apply directly to the value of the nested substitu‐
1339 tion; for example, the expansion ${${foo}} behaves exactly the
1340 same as ${foo}. When the `(P)' flag is present in a nested sub‐
1341 stitution, the other substitution rules are applied to the value
1342 before it is interpreted as a name, so ${${(P)foo}} may differ
1343 from ${(P)foo}.
1344 At each nested level of substitution, the substituted words un‐
1346 dergo all forms of single-word substitution (i.e. not filename
1347 generation), including command substitution, arithmetic expan‐
1348 sion and filename expansion (i.e. leading ~ and =). Thus, for
1349 example, ${${:-=cat}:h} expands to the directory where the cat
1350 program resides. (Explanation: the internal substitution has no
1351 parameter but a default value =cat, which is expanded by file‐
1352 name expansion to a full path; the outer substitution then ap‐
1353 plies the modifier :h and takes the directory part of the path.)
1354 2. Internal parameter flags
1356 Any parameter flags set by one of the typeset family of com‐
1357 mands, in particular the -L, -R, -Z, -u and -l options for pad‐
1358 ding and capitalization, are applied directly to the parameter
1359 value. Note these flags are options to the command, e.g. `type‐
1360 set -Z'; they are not the same as the flags used within parame‐
1361 ter substitutions.
1362 At the outermost level of substitution, the `(P)' flag (rule 4.)
1364 ignores these transformations and uses the unmodified value of
1365 the parameter as the name to be replaced. This is usually the
1366 desired behavior because padding may make the value syntacti‐
1367 cally illegal as a parameter name, but if capitalization changes
1368 are desired, use the ${${(P)foo}} form (rule 25.).
1369 3. Parameter subscripting
1371 If the value is a raw parameter reference with a subscript, such
1372 as ${var[3]}, the effect of subscripting is applied directly to
1373 the parameter. Subscripts are evaluated left to right; subse‐
1374 quent subscripts apply to the scalar or array value yielded by
1375 the previous subscript. Thus if var is an array, ${var[1][2]}
1376 is the second character of the first word, but ${var[2,4][2]} is
1377 the entire third word (the second word of the range of words two
1378 through four of the original array). Any number of subscripts
1379 may appear. Flags such as `(k)' and `(v)' which alter the re‐
1380 sult of subscripting are applied.
1381 4. Parameter name replacement
1383 At the outermost level of nesting only, the `(P)' flag is ap‐
1384 plied. This treats the value so far as a parameter name (which
1385 may include a subscript expression) and replaces that with the
1386 corresponding value. This replacement occurs later if the `(P)'
1387 flag appears in a nested substitution.
1388 If the value so far names a parameter that has internal flags
1390 (rule 2.), those internal flags are applied to the new value af‐
1391 ter replacement.
1392 5. Double-quoted joining
1394 If the value after this process is an array, and the substitu‐
1395 tion appears in double quotes, and neither an `(@)' flag nor a
1396 `#' length operator is present at the current level, then words
1397 of the value are joined with the first character of the parame‐
1398 ter $IFS, by default a space, between each word (single word ar‐
1399 rays are not modified). If the `(j)' flag is present, that is
1400 used for joining instead of $IFS.
1401 6. Nested subscripting
1403 Any remaining subscripts (i.e. of a nested substitution) are
1404 evaluated at this point, based on whether the value is an array
1405 or a scalar. As with 3., multiple subscripts can appear. Note
1406 that ${foo[2,4][2]} is thus equivalent to ${${foo[2,4]}[2]} and
1407 also to "${${(@)foo[2,4]}[2]}" (the nested substitution returns
1408 an array in both cases), but not to "${${foo[2,4]}[2]}" (the
1409 nested substitution returns a scalar because of the quotes).
1410 7. Modifiers
1412 Any modifiers, as specified by a trailing `#', `%', `/' (possi‐
1413 bly doubled) or by a set of modifiers of the form `:...' (see
1414 the section `Modifiers' in the section `History Expansion'), are
1415 applied to the words of the value at this level.
1416 8. Character evaluation
1418 Any `(#)' flag is applied, evaluating the result so far numeri‐
1419 cally as a character.
1420 9. Length
1422 Any initial `#' modifier, i.e. in the form ${#var}, is used to
1423 evaluate the length of the expression so far.
1424 10. Forced joining
1426 If the `(j)' flag is present, or no `(j)' flag is present but
1427 the string is to be split as given by rule 11., and joining did
1428 not take place at rule 5., any words in the value are joined to‐
1429 gether using the given string or the first character of $IFS if
1430 none. Note that the `(F)' flag implicitly supplies a string for
1431 joining in this manner.
1432 11. Simple word splitting
1434 If one of the `(s)' or `(f)' flags are present, or the `=' spec‐
1435 ifier was present (e.g. ${=var}), the word is split on occur‐
1436 rences of the specified string, or (for = with neither of the
1437 two flags present) any of the characters in $IFS.
1438 If no `(s)', `(f)' or `=' was given, but the word is not quoted
1440 and the option SH_WORD_SPLIT is set, the word is split on occur‐
1441 rences of any of the characters in $IFS. Note this step, too,
1442 takes place at all levels of a nested substitution.
1443 12. Case modification
1445 Any case modification from one of the flags `(L)', `(U)' or
1446 `(C)' is applied.
1447 13. Escape sequence replacement
1449 First any replacements from the `(g)' flag are performed, then
1450 any prompt-style formatting from the `(%)' family of flags is
1451 applied.
1452 14. Quote application
1454 Any quoting or unquoting using `(q)' and `(Q)' and related flags
1455 is applied.
1456 15. Directory naming
1458 Any directory name substitution using `(D)' flag is applied.
1459 16. Visibility enhancement
1461 Any modifications to make characters visible using the `(V)'
1462 flag are applied.
1463 17. Lexical word splitting
1465 If the '(z)' flag or one of the forms of the '(Z)' flag is
1466 present, the word is split as if it were a shell command line,
1467 so that quotation marks and other metacharacters are used to de‐
1468 cide what constitutes a word. Note this form of splitting is
1469 entirely distinct from that described by rule 11.: it does not
1470 use $IFS, and does not cause forced joining.
1471 18. Uniqueness
1473 If the result is an array and the `(u)' flag was present, dupli‐
1474 cate elements are removed from the array.
1475 19. Ordering
1477 If the result is still an array and one of the `(o)' or `(O)'
1478 flags was present, the array is reordered.
1479 20. RC_EXPAND_PARAM
1481 At this point the decision is made whether any resulting array
1482 elements are to be combined element by element with surrounding
1483 text, as given by either the RC_EXPAND_PARAM option or the `^'
1484 flag.
1485 21. Re-evaluation
1487 Any `(e)' flag is applied to the value, forcing it to be re-ex‐
1488 amined for new parameter substitutions, but also for command and
1489 arithmetic substitutions.
1490 22. Padding
1492 Any padding of the value by the `(l.fill.)' or `(r.fill.)' flags
1493 is applied.
1494 23. Semantic joining
1496 In contexts where expansion semantics requires a single word to
1497 result, all words are rejoined with the first character of IFS
1498 between. So in `${(P)${(f)lines}}' the value of ${lines} is
1499 split at newlines, but then must be joined again before the
1500 `(P)' flag can be applied.
1501 If a single word is not required, this rule is skipped.
1503 24. Empty argument removal
1505 If the substitution does not appear in double quotes, any re‐
1506 sulting zero-length argument, whether from a scalar or an ele‐
1507 ment of an array, is elided from the list of arguments inserted
1508 into the command line.
1509 Strictly speaking, the removal happens later as the same happens
1511 with other forms of substitution; the point to note here is sim‐
1512 ply that it occurs after any of the above parameter operations.
1513 25. Nested parameter name replacement
1515 If the `(P)' flag is present and rule 4. has not applied, the
1516 value so far is treated as a parameter name (which may include a
1517 subscript expression) and replaced with the corresponding value,
1518 with internal flags (rule 2.) applied to the new value.
1519 Examples
1521 The flag f is useful to split a double-quoted substitution line by
1522 line. For example, ${(f)"$(<file)"} substitutes the contents of file
1523 divided so that each line is an element of the resulting array. Com‐
1524 pare this with the effect of $(<file) alone, which divides the file up
1525 by words, or the same inside double quotes, which makes the entire con‐
1526 tent of the file a single string.
1527 The following illustrates the rules for nested parameter expansions.
1529 Suppose that $foo contains the array (bar baz):
1530 "${(@)${foo}[1]}"
1532 This produces the result b. First, the inner substitution
1533 "${foo}", which has no array (@) flag, produces a single word
1534 result "bar baz". The outer substitution "${(@)...[1]}" detects
1535 that this is a scalar, so that (despite the `(@)' flag) the sub‐
1536 script picks the first character.
1537 "${${(@)foo}[1]}"
1539 This produces the result `bar'. In this case, the inner substi‐
1540 tution "${(@)foo}" produces the array `(bar baz)'. The outer
1541 substitution "${...[1]}" detects that this is an array and picks
1542 the first word. This is similar to the simple case "${foo[1]}".
1543 As an example of the rules for word splitting and joining, suppose $foo
1545 contains the array `(ax1 bx1)'. Then
1546 ${(s/x/)foo}
1548 produces the words `a', `1 b' and `1'.
1549 ${(j/x/s/x/)foo}
1551 produces `a', `1', `b' and `1'.
1552 ${(s/x/)foo%%1*}
1554 produces `a' and ` b' (note the extra space). As substitution
1555 occurs before either joining or splitting, the operation first
1556 generates the modified array (ax bx), which is joined to give
1557 "ax bx", and then split to give `a', ` b' and `'. The final
1558 empty string will then be elided, as it is not in double quotes.
1559
1561 A command enclosed in parentheses preceded by a dollar sign, like
1562 `$(...)', or quoted with grave accents, like ``...`', is replaced with
1563 its standard output, with any trailing newlines deleted. If the sub‐
1564 stitution is not enclosed in double quotes, the output is broken into
1565 words using the IFS parameter.
1566 The substitution `$(cat foo)' may be replaced by the faster `$(<foo)'.
1568 In this case foo undergoes single word shell expansions (parameter ex‐
1569 pansion, command substitution and arithmetic expansion), but not file‐
1570 name generation.
1571 If the option GLOB_SUBST is set, the result of any unquoted command
1573 substitution, including the special form just mentioned, is eligible
1574 for filename generation.
1575
1577 A string of the form `$[exp]' or `$((exp))' is substituted with the
1578 value of the arithmetic expression exp. exp is subjected to parameter
1579 expansion, command substitution and arithmetic expansion before it is
1580 evaluated. See the section `Arithmetic Evaluation'.
1581
1583 A string of the form `foo{xx,yy,zz}bar' is expanded to the individual
1584 words `fooxxbar', `fooyybar' and `foozzbar'. Left-to-right order is
1585 preserved. This construct may be nested. Commas may be quoted in or‐
1586 der to include them literally in a word.
1587 An expression of the form `{n1..n2}', where n1 and n2 are integers, is
1589 expanded to every number between n1 and n2 inclusive. If either number
1590 begins with a zero, all the resulting numbers will be padded with lead‐
1591 ing zeroes to that minimum width, but for negative numbers the - char‐
1592 acter is also included in the width. If the numbers are in decreasing
1593 order the resulting sequence will also be in decreasing order.
1594 An expression of the form `{n1..n2..n3}', where n1, n2, and n3 are in‐
1596 tegers, is expanded as above, but only every n3th number starting from
1597 n1 is output. If n3 is negative the numbers are output in reverse or‐
1598 der, this is slightly different from simply swapping n1 and n2 in the
1599 case that the step n3 doesn't evenly divide the range. Zero padding
1600 can be specified in any of the three numbers, specifying it in the
1601 third can be useful to pad for example `{-99..100..01}' which is not
1602 possible to specify by putting a 0 on either of the first two numbers
1603 (i.e. pad to two characters).
1604 An expression of the form `{c1..c2}', where c1 and c2 are single char‐
1606 acters (which may be multibyte characters), is expanded to every char‐
1607 acter in the range from c1 to c2 in whatever character sequence is used
1608 internally. For characters with code points below 128 this is US ASCII
1609 (this is the only case most users will need). If any intervening char‐
1610 acter is not printable, appropriate quotation is used to render it
1611 printable. If the character sequence is reversed, the output is in re‐
1612 verse order, e.g. `{d..a}' is substituted as `d c b a'.
1613 If a brace expression matches none of the above forms, it is left un‐
1615 changed, unless the option BRACE_CCL (an abbreviation for `brace char‐
1616 acter class') is set. In that case, it is expanded to a list of the
1617 individual characters between the braces sorted into the order of the
1618 characters in the ASCII character set (multibyte characters are not
1619 currently handled). The syntax is similar to a [...] expression in
1620 filename generation: `-' is treated specially to denote a range of
1621 characters, but `^' or `!' as the first character is treated normally.
1622 For example, `{abcdef0-9}' expands to 16 words 0 1 2 3 4 5 6 7 8 9 a b
1623 c d e f.
1624 Note that brace expansion is not part of filename generation (glob‐
1626 bing); an expression such as */{foo,bar} is split into two separate
1627 words */foo and */bar before filename generation takes place. In par‐
1628 ticular, note that this is liable to produce a `no match' error if ei‐
1629 ther of the two expressions does not match; this is to be contrasted
1630 with */(foo|bar), which is treated as a single pattern but otherwise
1631 has similar effects.
1632 To combine brace expansion with array expansion, see the ${^spec} form
1634 described in the section `Parameter Expansion' above.
1635
1637 Each word is checked to see if it begins with an unquoted `~'. If it
1638 does, then the word up to a `/', or the end of the word if there is no
1639 `/', is checked to see if it can be substituted in one of the ways de‐
1640 scribed here. If so, then the `~' and the checked portion are replaced
1641 with the appropriate substitute value.
1642 A `~' by itself is replaced by the value of $HOME. A `~' followed by a
1644 `+' or a `-' is replaced by current or previous working directory, re‐
1645 spectively.
1646 A `~' followed by a number is replaced by the directory at that posi‐
1648 tion in the directory stack. `~0' is equivalent to `~+', and `~1' is
1649 the top of the stack. `~+' followed by a number is replaced by the di‐
1650 rectory at that position in the directory stack. `~+0' is equivalent
1651 to `~+', and `~+1' is the top of the stack. `~-' followed by a number
1652 is replaced by the directory that many positions from the bottom of the
1653 stack. `~-0' is the bottom of the stack. The PUSHD_MINUS option ex‐
1654 changes the effects of `~+' and `~-' where they are followed by a num‐
1655 ber.
1656 Dynamic named directories
1658 If the function zsh_directory_name exists, or the shell variable
1659 zsh_directory_name_functions exists and contains an array of function
1660 names, then the functions are used to implement dynamic directory nam‐
1661 ing. The functions are tried in order until one returns status zero,
1662 so it is important that functions test whether they can handle the case
1663 in question and return an appropriate status.
1664 A `~' followed by a string namstr in unquoted square brackets is
1666 treated specially as a dynamic directory name. Note that the first un‐
1667 quoted closing square bracket always terminates namstr. The shell
1668 function is passed two arguments: the string n (for name) and namstr.
1669 It should either set the array reply to a single element which is the
1670 directory corresponding to the name and return status zero (executing
1671 an assignment as the last statement is usually sufficient), or it
1672 should return status non-zero. In the former case the element of reply
1673 is used as the directory; in the latter case the substitution is deemed
1674 to have failed. If all functions fail and the option NOMATCH is set,
1675 an error results.
1676 The functions defined as above are also used to see if a directory can
1678 be turned into a name, for example when printing the directory stack or
1679 when expanding %~ in prompts. In this case each function is passed two
1680 arguments: the string d (for directory) and the candidate for dynamic
1681 naming. The function should either return non-zero status, if the di‐
1682 rectory cannot be named by the function, or it should set the array re‐
1683 ply to consist of two elements: the first is the dynamic name for the
1684 directory (as would appear within `~[...]'), and the second is the pre‐
1685 fix length of the directory to be replaced. For example, if the trial
1686 directory is /home/myname/src/zsh and the dynamic name for /home/my‐
1687 name/src (which has 16 characters) is s, then the function sets
1688 reply=(s 16)
1690 The directory name so returned is compared with possible static names
1692 for parts of the directory path, as described below; it is used if the
1693 prefix length matched (16 in the example) is longer than that matched
1694 by any static name.
1695 It is not a requirement that a function implements both n and d calls;
1697 for example, it might be appropriate for certain dynamic forms of ex‐
1698 pansion not to be contracted to names. In that case any call with the
1699 first argument d should cause a non-zero status to be returned.
1700 The completion system calls `zsh_directory_name c' followed by equiva‐
1702 lent calls to elements of the array zsh_directory_name_functions, if it
1703 exists, in order to complete dynamic names for directories. The code
1704 for this should be as for any other completion function as described in
1705 zshcompsys(1).
1706 As a working example, here is a function that expands any dynamic names
1708 beginning with the string p: to directories below /home/pws/perforce.
1709 In this simple case a static name for the directory would be just as
1710 effective.
1711 zsh_directory_name() {
1713 emulate -L zsh
1714 setopt extendedglob
1715 local -a match mbegin mend
1716 if [[ $1 = d ]]; then
1717 # turn the directory into a name
1718 if [[ $2 = (#b)(/home/pws/perforce/)([^/]##)* ]]; then
1719 typeset -ga reply
1720 reply=(p:$match[2] $(( ${#match[1]} + ${#match[2]} )) )
1721 else
1722 return 1
1723 fi
1724 elif [[ $1 = n ]]; then
1725 # turn the name into a directory
1726 [[ $2 != (#b)p:(?*) ]] && return 1
1727 typeset -ga reply
1728 reply=(/home/pws/perforce/$match[1])
1729 elif [[ $1 = c ]]; then
1730 # complete names
1731 local expl
1732 local -a dirs
1733 dirs=(/home/pws/perforce/*(/:t))
1734 dirs=(p:${^dirs})
1735 _wanted dynamic-dirs expl 'dynamic directory' compadd -S\] -a dirs
1736 return
1737 else
1738 return 1
1739 fi
1740 return 0
1741 }
1742 Static named directories
1744 A `~' followed by anything not already covered consisting of any number
1745 of alphanumeric characters or underscore (`_'), hyphen (`-'), or dot
1746 (`.') is looked up as a named directory, and replaced by the value of
1747 that named directory if found. Named directories are typically home
1748 directories for users on the system. They may also be defined if the
1749 text after the `~' is the name of a string shell parameter whose value
1750 begins with a `/'. Note that trailing slashes will be removed from the
1751 path to the directory (though the original parameter is not modified).
1752 It is also possible to define directory names using the -d option to
1754 the hash builtin.
1755 When the shell prints a path (e.g. when expanding %~ in prompts or when
1757 printing the directory stack), the path is checked to see if it has a
1758 named directory as its prefix. If so, then the prefix portion is re‐
1759 placed with a `~' followed by the name of the directory. The shorter
1760 of the two ways of referring to the directory is used, i.e. either the
1761 directory name or the full path; the name is used if they are the same
1762 length. The parameters $PWD and $OLDPWD are never abbreviated in this
1763 fashion.
1764 `=' expansion
1766 If a word begins with an unquoted `=' and the EQUALS option is set, the
1767 remainder of the word is taken as the name of a command. If a command
1768 exists by that name, the word is replaced by the full pathname of the
1769 command.
1770 Notes
1772 Filename expansion is performed on the right hand side of a parameter
1773 assignment, including those appearing after commands of the typeset
1774 family. In this case, the right hand side will be treated as a
1775 colon-separated list in the manner of the PATH parameter, so that a `~'
1776 or an `=' following a `:' is eligible for expansion. All such behav‐
1777 iour can be disabled by quoting the `~', the `=', or the whole expres‐
1778 sion (but not simply the colon); the EQUALS option is also respected.
1779 If the option MAGIC_EQUAL_SUBST is set, any unquoted shell argument in
1781 the form `identifier=expression' becomes eligible for file expansion as
1782 described in the previous paragraph. Quoting the first `=' also in‐
1783 hibits this.
1784
1786 If a word contains an unquoted instance of one of the characters `*',
1787 `(', `|', `<', `[', or `?', it is regarded as a pattern for filename
1788 generation, unless the GLOB option is unset. If the EXTENDED_GLOB op‐
1789 tion is set, the `^' and `#' characters also denote a pattern; other‐
1790 wise they are not treated specially by the shell.
1791 The word is replaced with a list of sorted filenames that match the
1793 pattern. If no matching pattern is found, the shell gives an error
1794 message, unless the NULL_GLOB option is set, in which case the word is
1795 deleted; or unless the NOMATCH option is unset, in which case the word
1796 is left unchanged.
1797 In filename generation, the character `/' must be matched explicitly;
1799 also, a `.' must be matched explicitly at the beginning of a pattern or
1800 after a `/', unless the GLOB_DOTS option is set. No filename genera‐
1801 tion pattern matches the files `.' or `..'. In other instances of pat‐
1802 tern matching, the `/' and `.' are not treated specially.
1803 Glob Operators
1805 * Matches any string, including the null string.
1806 ? Matches any character.
1808 [...] Matches any of the enclosed characters. Ranges of characters
1810 can be specified by separating two characters by a `-'. A `-'
1811 or `]' may be matched by including it as the first character in
1812 the list. There are also several named classes of characters,
1813 in the form `[:name:]' with the following meanings. The first
1814 set use the macros provided by the operating system to test for
1815 the given character combinations, including any modifications
1816 due to local language settings, see ctype(3):
1817 [:alnum:]
1819 The character is alphanumeric
1820 [:alpha:]
1822 The character is alphabetic
1823 [:ascii:]
1825 The character is 7-bit, i.e. is a single-byte character
1826 without the top bit set.
1827 [:blank:]
1829 The character is a blank character
1830 [:cntrl:]
1832 The character is a control character
1833 [:digit:]
1835 The character is a decimal digit
1836 [:graph:]
1838 The character is a printable character other than white‐
1839 space
1840 [:lower:]
1842 The character is a lowercase letter
1843 [:print:]
1845 The character is printable
1846 [:punct:]
1848 The character is printable but neither alphanumeric nor
1849 whitespace
1850 [:space:]
1852 The character is whitespace
1853 [:upper:]
1855 The character is an uppercase letter
1856 [:xdigit:]
1858 The character is a hexadecimal digit
1859 Another set of named classes is handled internally by the shell
1861 and is not sensitive to the locale:
1862 [:IDENT:]
1864 The character is allowed to form part of a shell identi‐
1865 fier, such as a parameter name; this test respects the
1866 POSIX_IDENTIFIERS option
1867 [:IFS:]
1869 The character is used as an input field separator, i.e.
1870 is contained in the IFS parameter
1871 [:IFSSPACE:]
1873 The character is an IFS white space character; see the
1874 documentation for IFS in the zshparam(1) manual page.
1875 [:INCOMPLETE:]
1877 Matches a byte that starts an incomplete multibyte char‐
1878 acter. Note that there may be a sequence of more than
1879 one bytes that taken together form the prefix of a multi‐
1880 byte character. To test for a potentially incomplete
1881 byte sequence, use the pattern `[[:INCOMPLETE:]]*'. This
1882 will never match a sequence starting with a valid multi‐
1883 byte character.
1884 [:INVALID:]
1886 Matches a byte that does not start a valid multibyte
1887 character. Note this may be a continuation byte of an
1888 incomplete multibyte character as any part of a multibyte
1889 string consisting of invalid and incomplete multibyte
1890 characters is treated as single bytes.
1891 [:WORD:]
1893 The character is treated as part of a word; this test is
1894 sensitive to the value of the WORDCHARS parameter
1895 Note that the square brackets are additional to those enclosing
1897 the whole set of characters, so to test for a single alphanu‐
1898 meric character you need `[[:alnum:]]'. Named character sets
1899 can be used alongside other types, e.g. `[[:alpha:]0-9]'.
1900 [^...]
1902 [!...] Like [...], except that it matches any character which is not in
1903 the given set.
1904 <[x]-[y]>
1906 Matches any number in the range x to y, inclusive. Either of
1907 the numbers may be omitted to make the range open-ended; hence
1908 `<->' matches any number. To match individual digits, the [...]
1909 form is more efficient.
1910 Be careful when using other wildcards adjacent to patterns of
1912 this form; for example, <0-9>* will actually match any number
1913 whatsoever at the start of the string, since the `<0-9>' will
1914 match the first digit, and the `*' will match any others. This
1915 is a trap for the unwary, but is in fact an inevitable conse‐
1916 quence of the rule that the longest possible match always suc‐
1917 ceeds. Expressions such as `<0-9>[^[:digit:]]*' can be used in‐
1918 stead.
1919 (...) Matches the enclosed pattern. This is used for grouping. If
1921 the KSH_GLOB option is set, then a `@', `*', `+', `?' or `!' im‐
1922 mediately preceding the `(' is treated specially, as detailed
1923 below. The option SH_GLOB prevents bare parentheses from being
1924 used in this way, though the KSH_GLOB option is still available.
1925 Note that grouping cannot extend over multiple directories: it
1927 is an error to have a `/' within a group (this only applies for
1928 patterns used in filename generation). There is one exception:
1929 a group of the form (pat/)# appearing as a complete path segment
1930 can match a sequence of directories. For example, foo/(a*/)#bar
1931 matches foo/bar, foo/any/bar, foo/any/anyother/bar, and so on.
1932 x|y Matches either x or y. This operator has lower precedence than
1934 any other. The `|' character must be within parentheses, to
1935 avoid interpretation as a pipeline. The alternatives are tried
1936 in order from left to right.
1937 ^x (Requires EXTENDED_GLOB to be set.) Matches anything except the
1939 pattern x. This has a higher precedence than `/', so `^foo/bar'
1940 will search directories in `.' except `./foo' for a file named
1941 `bar'.
1942 x~y (Requires EXTENDED_GLOB to be set.) Match anything that matches
1944 the pattern x but does not match y. This has lower precedence
1945 than any operator except `|', so `*/*~foo/bar' will search for
1946 all files in all directories in `.' and then exclude `foo/bar'
1947 if there was such a match. Multiple patterns can be excluded by
1948 `foo~bar~baz'. In the exclusion pattern (y), `/' and `.' are
1949 not treated specially the way they usually are in globbing.
1950 x# (Requires EXTENDED_GLOB to be set.) Matches zero or more occur‐
1952 rences of the pattern x. This operator has high precedence;
1953 `12#' is equivalent to `1(2#)', rather than `[1m(12)#'. It is an
1954 error for an unquoted `#' to follow something which cannot be
1955 repeated; this includes an empty string, a pattern already fol‐
1956 lowed by `##', or parentheses when part of a KSH_GLOB pattern
1957 (for example, `!(foo)#' is invalid and must be replaced by
1958 `*(!(foo))').
1959 x## (Requires EXTENDED_GLOB to be set.) Matches one or more occur‐
1961 rences of the pattern x. This operator has high precedence;
1962 `12##' is equivalent to `1(2##)', rather than `[1m(12)##'. No more
1963 than two active `#' characters may appear together. (Note the
1964 potential clash with glob qualifiers in the form `1(2##)' which
1965 should therefore be avoided.)
1966 ksh-like Glob Operators
1968 If the KSH_GLOB option is set, the effects of parentheses can be modi‐
1969 fied by a preceding `@', `*', `+', `?' or `!'. This character need not
1970 be unquoted to have special effects, but the `(' must be.
1971 @(...) Match the pattern in the parentheses. (Like `(...)'.)
1973 *(...) Match any number of occurrences. (Like `(...)#', except that
1975 recursive directory searching is not supported.)
1976 +(...) Match at least one occurrence. (Like `(...)##', except that re‐
1978 cursive directory searching is not supported.)
1979 ?(...) Match zero or one occurrence. (Like `(|...)'.)
1981 !(...) Match anything but the expression in parentheses. (Like
1983 `(^(...))'.)
1984 Precedence
1986 The precedence of the operators given above is (highest) `^', `/', `~',
1987 `|' (lowest); the remaining operators are simply treated from left to
1988 right as part of a string, with `#' and `##' applying to the shortest
1989 possible preceding unit (i.e. a character, `?', `[...]', `<...>', or a
1990 parenthesised expression). As mentioned above, a `/' used as a direc‐
1991 tory separator may not appear inside parentheses, while a `|' must do
1992 so; in patterns used in other contexts than filename generation (for
1993 example, in case statements and tests within `[[...]]'), a `/' is not
1994 special; and `/' is also not special after a `~' appearing outside
1995 parentheses in a filename pattern.
1996 Globbing Flags
1998 There are various flags which affect any text to their right up to the
1999 end of the enclosing group or to the end of the pattern; they require
2000 the EXTENDED_GLOB option. All take the form (#X) where X may have one
2001 of the following forms:
2002 i Case insensitive: upper or lower case characters in the pattern
2004 match upper or lower case characters.
2005 l Lower case characters in the pattern match upper or lower case
2007 characters; upper case characters in the pattern still only
2008 match upper case characters.
2009 I Case sensitive: locally negates the effect of i or l from that
2011 point on.
2012 b Activate backreferences for parenthesised groups in the pattern;
2014 this does not work in filename generation. When a pattern with
2015 a set of active parentheses is matched, the strings matched by
2016 the groups are stored in the array $match, the indices of the
2017 beginning of the matched parentheses in the array $mbegin, and
2018 the indices of the end in the array $mend, with the first ele‐
2019 ment of each array corresponding to the first parenthesised
2020 group, and so on. These arrays are not otherwise special to the
2021 shell. The indices use the same convention as does parameter
2022 substitution, so that elements of $mend and $mbegin may be used
2023 in subscripts; the KSH_ARRAYS option is respected. Sets of
2024 globbing flags are not considered parenthesised groups; only the
2025 first nine active parentheses can be referenced.
2026 For example,
2028 foo="a_string_with_a_message"
2030 if [[ $foo = (a|an)_(#b)(*) ]]; then
2031 print ${foo[$mbegin[1],$mend[1]]}
2032 fi
2033 prints `string_with_a_message'. Note that the first set of
2035 parentheses is before the (#b) and does not create a backrefer‐
2036 ence.
2037 Backreferences work with all forms of pattern matching other
2039 than filename generation, but note that when performing matches
2040 on an entire array, such as ${array#pattern}, or a global sub‐
2041 stitution, such as ${param//pat/repl}, only the data for the
2042 last match remains available. In the case of global replace‐
2043 ments this may still be useful. See the example for the m flag
2044 below.
2045 The numbering of backreferences strictly follows the order of
2047 the opening parentheses from left to right in the pattern
2048 string, although sets of parentheses may be nested. There are
2049 special rules for parentheses followed by `#' or `##'. Only the
2050 last match of the parenthesis is remembered: for example, in `[[
2051 abab = (#b)([ab])# ]]', only the final `b' is stored in
2052 match[1]. Thus extra parentheses may be necessary to match the
2053 complete segment: for example, use `X((ab|cd)#)Y' to match a
2054 whole string of either `ab' or `cd' between `X' and `Y', using
2055 the value of $match[1] rather than $match[2].
2056 If the match fails none of the parameters is altered, so in some
2058 cases it may be necessary to initialise them beforehand. If
2059 some of the backreferences fail to match -- which happens if
2060 they are in an alternate branch which fails to match, or if they
2061 are followed by # and matched zero times -- then the matched
2062 string is set to the empty string, and the start and end indices
2063 are set to -1.
2064 Pattern matching with backreferences is slightly slower than
2066 without.
2067 B Deactivate backreferences, negating the effect of the b flag
2069 from that point on.
2070 cN,M The flag (#cN,M) can be used anywhere that the # or ## operators
2072 can be used except in the expressions `(*/)#' and `(*/)##' in
2073 filename generation, where `/' has special meaning; it cannot be
2074 combined with other globbing flags and a bad pattern error oc‐
2075 curs if it is misplaced. It is equivalent to the form {N,M} in
2076 regular expressions. The previous character or group is re‐
2077 quired to match between N and M times, inclusive. The form
2078 (#cN) requires exactly N matches; (#c,M) is equivalent to speci‐
2079 fying N as 0; (#cN,) specifies that there is no maximum limit on
2080 the number of matches.
2081 m Set references to the match data for the entire string matched;
2083 this is similar to backreferencing and does not work in filename
2084 generation. The flag must be in effect at the end of the pat‐
2085 tern, i.e. not local to a group. The parameters $MATCH, $MBEGIN
2086 and $MEND will be set to the string matched and to the indices
2087 of the beginning and end of the string, respectively. This is
2088 most useful in parameter substitutions, as otherwise the string
2089 matched is obvious.
2090 For example,
2092 arr=(veldt jynx grimps waqf zho buck)
2094 print ${arr//(#m)[aeiou]/${(U)MATCH}}
2095 forces all the matches (i.e. all vowels) into uppercase, print‐
2097 ing `vEldt jynx grImps wAqf zhO bUck'.
2098 Unlike backreferences, there is no speed penalty for using match
2100 references, other than the extra substitutions required for the
2101 replacement strings in cases such as the example shown.
2102 M Deactivate the m flag, hence no references to match data will be
2104 created.
2105 anum Approximate matching: num errors are allowed in the string
2107 matched by the pattern. The rules for this are described in the
2108 next subsection.
2109 s, e Unlike the other flags, these have only a local effect, and each
2111 must appear on its own: `(#s)' and `(#e)' are the only valid
2112 forms. The `(#s)' flag succeeds only at the start of the test
2113 string, and the `(#e)' flag succeeds only at the end of the test
2114 string; they correspond to `^' and `$' in standard regular ex‐
2115 pressions. They are useful for matching path segments in pat‐
2116 terns other than those in filename generation (where path seg‐
2117 ments are in any case treated separately). For example,
2118 `*((#s)|/)test((#e)|/)*' matches a path segment `test' in any of
2119 the following strings: test, test/at/start, at/end/test,
2120 in/test/middle.
2121 Another use is in parameter substitution; for example `${ar‐
2123 ray/(#s)A*Z(#e)}' will remove only elements of an array which
2124 match the complete pattern `A*Z'. There are other ways of per‐
2125 forming many operations of this type, however the combination of
2126 the substitution operations `/' and `//' with the `(#s)' and
2127 `(#e)' flags provides a single simple and memorable method.
2128 Note that assertions of the form `(^(#s))' also work, i.e. match
2130 anywhere except at the start of the string, although this actu‐
2131 ally means `anything except a zero-length portion at the start
2132 of the string'; you need to use `(""~(#s))' to match a
2133 zero-length portion of the string not at the start.
2134 q A `q' and everything up to the closing parenthesis of the glob‐
2136 bing flags are ignored by the pattern matching code. This is
2137 intended to support the use of glob qualifiers, see below. The
2138 result is that the pattern `(#b)(*).c(#q.)' can be used both for
2139 globbing and for matching against a string. In the former case,
2140 the `(#q.)' will be treated as a glob qualifier and the `(#b)'
2141 will not be useful, while in the latter case the `(#b)' is use‐
2142 ful for backreferences and the `(#q.)' will be ignored. Note
2143 that colon modifiers in the glob qualifiers are also not applied
2144 in ordinary pattern matching.
2145 u Respect the current locale in determining the presence of multi‐
2147 byte characters in a pattern, provided the shell was compiled
2148 with MULTIBYTE_SUPPORT. This overrides the MULTIBYTE option;
2149 the default behaviour is taken from the option. Compare U.
2150 (Mnemonic: typically multibyte characters are from Unicode in
2151 the UTF-8 encoding, although any extension of ASCII supported by
2152 the system library may be used.)
2153 U All characters are considered to be a single byte long. The op‐
2155 posite of u. This overrides the MULTIBYTE option.
2156 For example, the test string fooxx can be matched by the pattern
2158 (#i)FOOXX, but not by (#l)FOOXX, (#i)FOO(#I)XX or ((#i)FOOX)X. The
2159 string (#ia2)readme specifies case-insensitive matching of readme with
2160 up to two errors.
2161 When using the ksh syntax for grouping both KSH_GLOB and EXTENDED_GLOB
2163 must be set and the left parenthesis should be preceded by @. Note
2164 also that the flags do not affect letters inside [...] groups, in other
2165 words (#i)[a-z] still matches only lowercase letters. Finally, note
2166 that when examining whole paths case-insensitively every directory must
2167 be searched for all files which match, so that a pattern of the form
2168 (#i)/foo/bar/... is potentially slow.
2169 Approximate Matching
2171 When matching approximately, the shell keeps a count of the errors
2172 found, which cannot exceed the number specified in the (#anum) flags.
2173 Four types of error are recognised:
2174 1. Different characters, as in fooxbar and fooybar.
2176 2. Transposition of characters, as in banana and abnana.
2178 3. A character missing in the target string, as with the pattern
2180 road and target string rod.
2181 4. An extra character appearing in the target string, as with stove
2183 and strove.
2184 Thus, the pattern (#a3)abcd matches dcba, with the errors occurring by
2186 using the first rule twice and the second once, grouping the string as
2187 [d][cb][a] and [a][bc][d].
2188 Non-literal parts of the pattern must match exactly, including charac‐
2190 ters in character ranges: hence (#a1)??? matches strings of length
2191 four, by applying rule 4 to an empty part of the pattern, but not
2192 strings of length two, since all the ? must match. Other characters
2193 which must match exactly are initial dots in filenames (unless the
2194 GLOB_DOTS option is set), and all slashes in filenames, so that a/bc is
2195 two errors from ab/c (the slash cannot be transposed with another char‐
2196 acter). Similarly, errors are counted separately for non-contiguous
2197 strings in the pattern, so that (ab|cd)ef is two errors from aebf.
2198 When using exclusion via the ~ operator, approximate matching is
2200 treated entirely separately for the excluded part and must be activated
2201 separately. Thus, (#a1)README~READ_ME matches READ.ME but not READ_ME,
2202 as the trailing READ_ME is matched without approximation. However,
2203 (#a1)README~(#a1)READ_ME does not match any pattern of the form READ?ME
2204 as all such forms are now excluded.
2205 Apart from exclusions, there is only one overall error count; however,
2207 the maximum errors allowed may be altered locally, and this can be de‐
2208 limited by grouping. For example, (#a1)cat((#a0)dog)fox allows one er‐
2209 ror in total, which may not occur in the dog section, and the pattern
2210 (#a1)cat(#a0)dog(#a1)fox is equivalent. Note that the point at which
2211 an error is first found is the crucial one for establishing whether to
2212 use approximation; for example, (#a1)abc(#a0)xyz will not match
2213 abcdxyz, because the error occurs at the `x', where approximation is
2214 turned off.
2215 Entire path segments may be matched approximately, so that
2217 `(#a1)/foo/d/is/available/at/the/bar' allows one error in any path seg‐
2218 ment. This is much less efficient than without the (#a1), however,
2219 since every directory in the path must be scanned for a possible ap‐
2220 proximate match. It is best to place the (#a1) after any path segments
2221 which are known to be correct.
2222 Recursive Globbing
2224 A pathname component of the form `(foo/)#' matches a path consisting of
2225 zero or more directories matching the pattern foo.
2226 As a shorthand, `**/' is equivalent to `(*/)#'; note that this there‐
2228 fore matches files in the current directory as well as subdirectories.
2229 Thus:
2230 ls -ld -- (*/)#bar
2232 or
2234 ls -ld -- **/bar
2236 does a recursive directory search for files named `bar' (potentially
2238 including the file `bar' in the current directory). This form does not
2239 follow symbolic links; the alternative form `***/' does, but is other‐
2240 wise identical. Neither of these can be combined with other forms of
2241 globbing within the same path segment; in that case, the `*' operators
2242 revert to their usual effect.
2243 Even shorter forms are available when the option GLOB_STAR_SHORT is
2245 set. In that case if no / immediately follows a ** or *** they are
2246 treated as if both a / plus a further * are present. Hence:
2247 setopt GLOBSTARSHORT
2249 ls -ld -- **.c
2250 is equivalent to
2252 ls -ld -- **/*.c
2254 Glob Qualifiers
2256 Patterns used for filename generation may end in a list of qualifiers
2257 enclosed in parentheses. The qualifiers specify which filenames that
2258 otherwise match the given pattern will be inserted in the argument
2259 list.
2260 If the option BARE_GLOB_QUAL is set, then a trailing set of parentheses
2262 containing no `|' or `(' characters (or `~' if it is special) is taken
2263 as a set of glob qualifiers. A glob subexpression that would normally
2264 be taken as glob qualifiers, for example `(^x)', can be forced to be
2265 treated as part of the glob pattern by doubling the parentheses, in
2266 this case producing `((^x))'.
2267 If the option EXTENDED_GLOB is set, a different syntax for glob quali‐
2269 fiers is available, namely `(#qx)' where x is any of the same glob
2270 qualifiers used in the other format. The qualifiers must still appear
2271 at the end of the pattern. However, with this syntax multiple glob
2272 qualifiers may be chained together. They are treated as a logical AND
2273 of the individual sets of flags. Also, as the syntax is unambiguous,
2274 the expression will be treated as glob qualifiers just as long any
2275 parentheses contained within it are balanced; appearance of `|', `(' or
2276 `~' does not negate the effect. Note that qualifiers will be recog‐
2277 nised in this form even if a bare glob qualifier exists at the end of
2278 the pattern, for example `*(#q*)(.)' will recognise executable regular
2279 files if both options are set; however, mixed syntax should probably be
2280 avoided for the sake of clarity. Note that within conditions using the
2281 `[[' form the presence of a parenthesised expression (#q...) at the end
2282 of a string indicates that globbing should be performed; the expression
2283 may include glob qualifiers, but it is also valid if it is simply (#q).
2284 This does not apply to the right hand side of pattern match operators
2285 as the syntax already has special significance.
2286 A qualifier may be any one of the following:
2288 / directories
2290 F `full' (i.e. non-empty) directories. Note that the opposite
2292 sense (^F) expands to empty directories and all non-directories.
2293 Use (/^F) for empty directories.
2294 . plain files
2296 @ symbolic links
2298 = sockets
2300 p named pipes (FIFOs)
2302 * executable plain files (0100 or 0010 or 0001)
2304 % device files (character or block special)
2306 %b block special files
2308 %c character special files
2310 r owner-readable files (0400)
2312 w owner-writable files (0200)
2314 x owner-executable files (0100)
2316 A group-readable files (0040)
2318 I group-writable files (0020)
2320 E group-executable files (0010)
2322 R world-readable files (0004)
2324 W world-writable files (0002)
2326 X world-executable files (0001)
2328 s setuid files (04000)
2330 S setgid files (02000)
2332 t files with the sticky bit (01000)
2334 fspec files with access rights matching spec. This spec may be a octal
2336 number optionally preceded by a `=', a `+', or a `-'. If none of
2337 these characters is given, the behavior is the same as for `='.
2338 The octal number describes the mode bits to be expected, if com‐
2339 bined with a `=', the value given must match the file-modes ex‐
2340 actly, with a `+', at least the bits in the given number must be
2341 set in the file-modes, and with a `-', the bits in the number
2342 must not be set. Giving a `?' instead of a octal digit anywhere
2343 in the number ensures that the corresponding bits in the
2344 file-modes are not checked, this is only useful in combination
2345 with `='.
2346 If the qualifier `f' is followed by any other character anything
2348 up to the next matching character (`[', `{', and `<' match `]',
2349 `}', and `>' respectively, any other character matches itself)
2350 is taken as a list of comma-separated sub-specs. Each sub-spec
2351 may be either an octal number as described above or a list of
2352 any of the characters `u', `g', `o', and `a', followed by a `=',
2353 a `+', or a `-', followed by a list of any of the characters
2354 `r', `w', `x', `s', and `t', or an octal digit. The first list
2355 of characters specify which access rights are to be checked. If
2356 a `u' is given, those for the owner of the file are used, if a
2357 `g' is given, those of the group are checked, a `o' means to
2358 test those of other users, and the `a' says to test all three
2359 groups. The `=', `+', and `-' again says how the modes are to be
2360 checked and have the same meaning as described for the first
2361 form above. The second list of characters finally says which ac‐
2362 cess rights are to be expected: `r' for read access, `w' for
2363 write access, `x' for the right to execute the file (or to
2364 search a directory), `s' for the setuid and setgid bits, and `t'
2365 for the sticky bit.
2366 Thus, `*(f70?)' gives the files for which the owner has read,
2368 write, and execute permission, and for which other group members
2369 have no rights, independent of the permissions for other users.
2370 The pattern `*(f-100)' gives all files for which the owner does
2371 not have execute permission, and `*(f:gu+w,o-rx:)' gives the
2372 files for which the owner and the other members of the group
2373 have at least write permission, and for which other users don't
2374 have read or execute permission.
2375 estring
2377 +cmd The string will be executed as shell code. The filename will be
2378 included in the list if and only if the code returns a zero sta‐
2379 tus (usually the status of the last command).
2380 In the first form, the first character after the `e' will be
2382 used as a separator and anything up to the next matching separa‐
2383 tor will be taken as the string; `[', `{', and `<' match `]',
2384 `}', and `>', respectively, while any other character matches
2385 itself. Note that expansions must be quoted in the string to
2386 prevent them from being expanded before globbing is done.
2387 string is then executed as shell code. The string globqual is
2388 appended to the array zsh_eval_context the duration of execu‐
2389 tion.
2390 During the execution of string the filename currently being
2392 tested is available in the parameter REPLY; the parameter may be
2393 altered to a string to be inserted into the list instead of the
2394 original filename. In addition, the parameter reply may be set
2395 to an array or a string, which overrides the value of REPLY. If
2396 set to an array, the latter is inserted into the command line
2397 word by word.
2398 For example, suppose a directory contains a single file
2400 `lonely'. Then the expression `*(e:'reply=(${REPLY}{1,2})':)'
2401 will cause the words `lonely1' and `lonely2' to be inserted into
2402 the command line. Note the quoting of string.
2403 The form +cmd has the same effect, but no delimiters appear
2405 around cmd. Instead, cmd is taken as the longest sequence of
2406 characters following the + that are alphanumeric or underscore.
2407 Typically cmd will be the name of a shell function that contains
2408 the appropriate test. For example,
2409 nt() { [[ $REPLY -nt $NTREF ]] }
2411 NTREF=reffile
2412 ls -ld -- *(+nt)
2413 lists all files in the directory that have been modified more
2415 recently than reffile.
2416 ddev files on the device dev
2418 l[-|+]ct
2420 files having a link count less than ct (-), greater than ct (+),
2421 or equal to ct
2422 U files owned by the effective user ID
2424 G files owned by the effective group ID
2426 uid files owned by user ID id if that is a number. Otherwise, id
2428 specifies a user name: the character after the `u' will be taken
2429 as a separator and the string between it and the next matching
2430 separator will be taken as a user name. The starting separators
2431 `[', `{', and `<' match the final separators `]', `}', and `>',
2432 respectively; any other character matches itself. The selected
2433 files are those owned by this user. For example, `u:foo:' or
2434 `u[foo]' selects files owned by user `foo'.
2435 gid like uid but with group IDs or names
2437 a[Mwhms][-|+]n
2439 files accessed exactly n days ago. Files accessed within the
2440 last n days are selected using a negative value for n (-n).
2441 Files accessed more than n days ago are selected by a positive n
2442 value (+n). Optional unit specifiers `M', `w', `h', `m' or `s'
2443 (e.g. `ah5') cause the check to be performed with months (of 30
2444 days), weeks, hours, minutes or seconds instead of days, respec‐
2445 tively. An explicit `d' for days is also allowed.
2446 Any fractional part of the difference between the access time
2448 and the current part in the appropriate units is ignored in the
2449 comparison. For instance, `echo *(ah-5)' would echo files ac‐
2450 cessed within the last five hours, while `echo *(ah+5)' would
2451 echo files accessed at least six hours ago, as times strictly
2452 between five and six hours are treated as five hours.
2453 m[Mwhms][-|+]n
2455 like the file access qualifier, except that it uses the file
2456 modification time.
2457 c[Mwhms][-|+]n
2459 like the file access qualifier, except that it uses the file in‐
2460 ode change time.
2461 L[+|-]n
2463 files less than n bytes (-), more than n bytes (+), or exactly n
2464 bytes in length.
2465 If this flag is directly followed by a size specifier `k' (`K'),
2467 `m' (`M'), or `p' (`P') (e.g. `Lk-50') the check is performed
2468 with kilobytes, megabytes, or blocks (of 512 bytes) instead.
2469 (On some systems additional specifiers are available for giga‐
2470 bytes, `g' or `G', and terabytes, `t' or `T'.) If a size speci‐
2471 fier is used a file is regarded as "exactly" the size if the
2472 file size rounded up to the next unit is equal to the test size.
2473 Hence `*(Lm1)' matches files from 1 byte up to 1 Megabyte inclu‐
2474 sive. Note also that the set of files "less than" the test size
2475 only includes files that would not match the equality test;
2476 hence `*(Lm-1)' only matches files of zero size.
2477 ^ negates all qualifiers following it
2479 - toggles between making the qualifiers work on symbolic links
2481 (the default) and the files they point to, if any; any symbolic
2482 link for whose target the `stat' system call fails (whatever the
2483 cause of the failure) is treated as a file in its own right
2484 M sets the MARK_DIRS option for the current pattern
2486 T appends a trailing qualifier mark to the filenames, analogous to
2488 the LIST_TYPES option, for the current pattern (overrides M)
2489 N sets the NULL_GLOB option for the current pattern
2491 D sets the GLOB_DOTS option for the current pattern
2493 n sets the NUMERIC_GLOB_SORT option for the current pattern
2495 Yn enables short-circuit mode: the pattern will expand to at most n
2497 filenames. If more than n matches exist, only the first n
2498 matches in directory traversal order will be considered.
2499 Implies oN when no oc qualifier is used.
2501 oc specifies how the names of the files should be sorted. The fol‐
2503 lowing values of c sort in the following ways:
2504 n By name.
2506 L By the size (length) of the files.
2507 l By number of links.
2508 a By time of last access, youngest first.
2509 m By time of last modification, youngest first.
2510 c By time of last inode change, youngest first.
2511 d By directories: files in subdirectories appear before
2512 those in the current directory at each level of the
2513 search -- this is best combined with other criteria, for
2514 example `odon' to sort on names for files within the same
2515 directory.
2516 N No sorting is performed.
2517 estring
2518 +cmd Sort by shell code (see below).
2519 Note that the modifiers ^ and - are used, so `*(^-oL)' gives a
2521 list of all files sorted by file size in descending order, fol‐
2522 lowing any symbolic links. Unless oN is used, multiple order
2523 specifiers may occur to resolve ties.
2524 The default sorting is n (by name) unless the Y glob qualifier
2526 is used, in which case it is N (unsorted).
2527 oe and o+ are special cases; they are each followed by shell
2529 code, delimited as for the e glob qualifier and the + glob qual‐
2530 ifier respectively (see above). The code is executed for each
2531 matched file with the parameter REPLY set to the name of the
2532 file on entry and globsort appended to zsh_eval_context. The
2533 code should modify the parameter REPLY in some fashion. On re‐
2534 turn, the value of the parameter is used instead of the file
2535 name as the string on which to sort. Unlike other sort opera‐
2536 tors, oe and o+ may be repeated, but note that the maximum num‐
2537 ber of sort operators of any kind that may appear in any glob
2538 expression is 12.
2539 Oc like `o', but sorts in descending order; i.e. `*(^oc)' is the
2541 same as `*(Oc)' and `*(^Oc)' is the same as `*(oc)'; `Od' puts
2542 files in the current directory before those in subdirectories at
2543 each level of the search.
2544 [beg[,end]]
2546 specifies which of the matched filenames should be included in
2547 the returned list. The syntax is the same as for array sub‐
2548 scripts. beg and the optional end may be mathematical expres‐
2549 sions. As in parameter subscripting they may be negative to make
2550 them count from the last match backward. E.g.: `*(-OL[1,3])'
2551 gives a list of the names of the three largest files.
2552 Pstring
2554 The string will be prepended to each glob match as a separate
2555 word. string is delimited in the same way as arguments to the e
2556 glob qualifier described above. The qualifier can be repeated;
2557 the words are prepended separately so that the resulting command
2558 line contains the words in the same order they were given in the
2559 list of glob qualifiers.
2560 A typical use for this is to prepend an option before all occur‐
2562 rences of a file name; for example, the pattern `*(P:-f:)' pro‐
2563 duces the command line arguments `-f file1 -f file2 ...'
2564 If the modifier ^ is active, then string will be appended in‐
2566 stead of prepended. Prepending and appending is done indepen‐
2567 dently so both can be used on the same glob expression; for ex‐
2568 ample by writing `*(P:foo:^P:bar:^P:baz:)' which produces the
2569 command line arguments `foo baz file1 bar ...'
2570 More than one of these lists can be combined, separated by commas. The
2572 whole list matches if at least one of the sublists matches (they are
2573 `or'ed, the qualifiers in the sublists are `and'ed). Some qualifiers,
2574 however, affect all matches generated, independent of the sublist in
2575 which they are given. These are the qualifiers `M', `T', `N', `D',
2576 `n', `o', `O' and the subscripts given in brackets (`[...]').
2577 If a `:' appears in a qualifier list, the remainder of the expression
2579 in parenthesis is interpreted as a modifier (see the section `Modi‐
2580 fiers' in the section `History Expansion'). Each modifier must be in‐
2581 troduced by a separate `:'. Note also that the result after modifica‐
2582 tion does not have to be an existing file. The name of any existing
2583 file can be followed by a modifier of the form `(:...)' even if no ac‐
2584 tual filename generation is performed, although note that the presence
2585 of the parentheses causes the entire expression to be subjected to any
2586 global pattern matching options such as NULL_GLOB. Thus:
2587 ls -ld -- *(-/)
2589 lists all directories and symbolic links that point to directories, and
2591 ls -ld -- *(-@)
2593 lists all broken symbolic links, and
2595 ls -ld -- *(%W)
2597 lists all world-writable device files in the current directory, and
2599 ls -ld -- *(W,X)
2601 lists all files in the current directory that are world-writable or
2603 world-executable, and
2604 print -rC1 /tmp/foo*(u0^@:t)
2606 outputs the basename of all root-owned files beginning with the string
2608 `foo' in /tmp, ignoring symlinks, and
2609 ls -ld -- *.*~(lex|parse).[ch](^D^l1)
2611 lists all files having a link count of one whose names contain a dot
2613 (but not those starting with a dot, since GLOB_DOTS is explicitly
2614 switched off) except for lex.c, lex.h, parse.c and parse.h.
2615 print -rC1 b*.pro(#q:s/pro/shmo/)(#q.:s/builtin/shmiltin/)
2617 demonstrates how colon modifiers and other qualifiers may be chained
2619 together. The ordinary qualifier `.' is applied first, then the colon
2620 modifiers in order from left to right. So if EXTENDED_GLOB is set and
2621 the base pattern matches the regular file builtin.pro, the shell will
2622 print `shmiltin.shmo'.
2623zsh 5.9 May 14, 2022 ZSHEXPN(1)