1PSED(1) User Contributed Perl Documentation PSED(1)
2
6 psed - a stream editor
7
9 psed [-an] script [file ...]
10 psed [-an] [-e script] [-f script-file] [file ...]
11 s2p [-an] [-e script] [-f script-file]
13
15 A stream editor reads the input stream consisting of the specified
16 files (or standard input, if none are given), processes is line by line
17 by applying a script consisting of edit commands, and writes resulting
18 lines to standard output. The filename '"-"' may be used to read
19 standard input.
20 The edit script is composed from arguments of -e options and script-
22 files, in the given order. A single script argument may be specified as
23 the first parameter.
24 If this program is invoked with the name s2p, it will act as a sed-to-
26 Perl translator. See "SED SCRIPT TRANSLATION".
27 sed returns an exit code of 0 on success or >0 if an error occurred.
29
31 -a A file specified as argument to the w edit command is by default
32 opened before input processing starts. Using -a, opening of such
33 files is delayed until the first line is actually written to the
34 file.
35 -e script
37 The editing commands defined by script are appended to the script.
38 Multiple commands must be separated by newlines or semicolons.
39 -f script-file
41 Editing commands from the specified script-file are read and
42 appended to the script.
43 -n By default, a line is written to standard output after the editing
45 script has been applied to it. The -n option suppresses automatic
46 printing.
47
49 sed command syntax is defined as
50 [address[,address]][!]function[argument]
52 with whitespace being permitted before or after addresses, and between
54 the function character and the argument. The addresses and the address
55 inverter ("!") are used to restrict the application of a command to the
56 selected line(s) of input.
57 Each command must be on a line of its own, except where noted in the
59 synopses below.
60 The edit cycle performed on each input line consist of reading the line
62 (without its trailing newline character) into the pattern space,
63 applying the applicable commands of the edit script, writing the final
64 contents of the pattern space and a newline to the standard output. A
65 hold space is provided for saving the contents of the pattern space for
66 later use.
67 Addresses
69 A sed address is either a line number or a pattern, which may be
70 combined arbitrarily to construct ranges. Lines are numbered across all
71 input files.
72 Any address may be followed by an exclamation mark ('"!"'), selecting
74 all lines not matching that address.
75 number
77 The line with the given number is selected.
78 $ A dollar sign ("$") is the line number of the last line of the
80 input stream.
81 /regular expression/
83 A pattern address is a basic regular expression (see "BASIC REGULAR
84 EXPRESSIONS"), between the delimiting character "/". Any other
85 character except "\" or newline may be used to delimit a pattern
86 address when the initial delimiter is prefixed with a backslash
87 ('"\"').
88 If no address is given, the command selects every line.
90 If one address is given, it selects the line (or lines) matching the
92 address.
93 Two addresses select a range that begins whenever the first address
95 matches, and ends (including that line) when the second address
96 matches. If the first (second) address is a matching pattern, the
97 second address is not applied to the very same line to determine the
98 end of the range. Likewise, if the second address is a matching
99 pattern, the first address is not applied to the very same line to
100 determine the begin of another range. If both addresses are line
101 numbers, and the second line number is less than the first line number,
102 then only the first line is selected.
103 Functions
105 The maximum permitted number of addresses is indicated with each
106 function synopsis below.
107 The argument text consists of one or more lines following the command.
109 Embedded newlines in text must be preceded with a backslash. Other
110 backslashes in text are deleted and the following character is taken
111 literally.
112 [1addr]a\ text
114 Write text (which must start on the line following the command) to
115 standard output immediately before reading the next line of input,
116 either by executing the N function or by beginning a new cycle.
117 [2addr]b [label]
119 Branch to the : function with the specified label. If no label is
120 given, branch to the end of the script.
121 [2addr]c\ text
123 The line, or range of lines, selected by the address is deleted.
124 The text (which must start on the line following the command) is
125 written to standard output. With an address range, this occurs at
126 the end of the range.
127 [2addr]d
129 Deletes the pattern space and starts the next cycle.
130 [2addr]D
132 Deletes the pattern space through the first embedded newline or to
133 the end. If the pattern space becomes empty, a new cycle is
134 started, otherwise execution of the script is restarted.
135 [2addr]g
137 Replace the contents of the pattern space with the hold space.
138 [2addr]G
140 Append a newline and the contents of the hold space to the pattern
141 space.
142 [2addr]h
144 Replace the contents of the hold space with the pattern space.
145 [2addr]H
147 Append a newline and the contents of the pattern space to the hold
148 space.
149 [1addr]i\ text
151 Write the text (which must start on the line following the command)
152 to standard output.
153 [2addr]l
155 Print the contents of the pattern space: non-printable characters
156 are shown in C-style escaped form; long lines are split and have a
157 trailing ^'"\"' at the point of the split; the true end of a line
158 is marked with a '"$"'. Escapes are: '\a', '\t', '\n', '\f', '\r',
159 '\e' for BEL, HT, LF, FF, CR, ESC, respectively, and '\' followed
160 by a three-digit octal number for all other non-printable
161 characters.
162 [2addr]n
164 If automatic printing is enabled, write the pattern space to the
165 standard output. Replace the pattern space with the next line of
166 input. If there is no more input, processing is terminated.
167 [2addr]N
169 Append a newline and the next line of input to the pattern space.
170 If there is no more input, processing is terminated.
171 [2addr]p
173 Print the pattern space to the standard output. (Use the -n option
174 to suppress automatic printing at the end of a cycle if you want to
175 avoid double printing of lines.)
176 [2addr]P
178 Prints the pattern space through the first embedded newline or to
179 the end.
180 [1addr]q
182 Branch to the end of the script and quit without starting a new
183 cycle.
184 [1addr]r file
186 Copy the contents of the file to standard output immediately before
187 the next attempt to read a line of input. Any error encountered
188 while reading file is silently ignored.
189 [2addr]s/regular expression/replacement/flags
191 Substitute the replacement string for the first substring in the
192 pattern space that matches the regular expression. Any character
193 other than backslash or newline can be used instead of a slash to
194 delimit the regular expression and the replacement. To use the
195 delimiter as a literal character within the regular expression and
196 the replacement, precede the character by a backslash ('"\"').
197 Literal newlines may be embedded in the replacement string by
199 preceding a newline with a backslash.
200 Within the replacement, an ampersand ('"&"') is replaced by the
202 string matching the regular expression. The strings '"\1"' through
203 '"\9"' are replaced by the corresponding subpattern (see "BASIC
204 REGULAR EXPRESSIONS"). To get a literal '"&"' or '"\"' in the
205 replacement text, precede it by a backslash.
206 The following flags modify the behaviour of the s command:
208 g The replacement is performed for all matching, non-
210 overlapping substrings of the pattern space.
211 1..9 Replace only the n-th matching substring of the pattern
213 space.
214 p If the substitution was made, print the new value of the
216 pattern space.
217 w file If the substitution was made, write the new value of the
219 pattern space to the specified file.
220 [2addr]t [label]
222 Branch to the : function with the specified label if any s
223 substitutions have been made since the most recent reading of an
224 input line or execution of a t function. If no label is given,
225 branch to the end of the script.
226 [2addr]w file
228 The contents of the pattern space are written to the file.
229 [2addr]x
231 Swap the contents of the pattern space and the hold space.
232 [2addr]y/string1/string2/
234 In the pattern space, replace all characters occurring in string1
235 by the character at the corresponding position in string2. It is
236 possible to use any character (other than a backslash or newline)
237 instead of a slash to delimit the strings. Within string1 and
238 string2, a backslash followed by any character other than a newline
239 is that literal character, and a backslash followed by an 'n' is
240 replaced by a newline character.
241 [1addr]=
243 Prints the current line number on the standard output.
244 [0addr]: [label]
246 The command specifies the position of the label. It has no other
247 effect.
248 [2addr]{ [command]
250 [0addr]}
251 These two commands begin and end a command list. The first command
252 may be given on the same line as the opening { command. The
253 commands within the list are jointly selected by the address(es)
254 given on the { command (but may still have individual addresses).
255 [0addr]# [comment]
257 The entire line is ignored (treated as a comment). If, however, the
258 first two characters in the script are '"#n"', automatic printing
259 of output is suppressed, as if the -n option were given on the
260 command line.
261
263 A Basic Regular Expression (BRE), as defined in POSIX 1003.2, consists
264 of atoms, for matching parts of a string, and bounds, specifying
265 repetitions of a preceding atom.
266 Atoms
268 The possible atoms of a BRE are: ., matching any single character; ^
269 and $, matching the null string at the beginning or end of a string,
270 respectively; a bracket expressions, enclosed in [ and ] (see below);
271 and any single character with no other significance (matching that
272 character). A \ before one of: ., ^, $, [, *, \, matching the character
273 after the backslash. A sequence of atoms enclosed in \( and \) becomes
274 an atom and establishes the target for a backreference, consisting of
275 the substring that actually matches the enclosed atoms. Finally, \
276 followed by one of the digits 0 through 9 is a backreference.
277 A ^ that is not first, or a $ that is not last does not have a special
279 significance and need not be preceded by a backslash to become literal.
280 The same is true for a ], that does not terminate a bracket expression.
281 An unescaped backslash cannot be last in a BRE.
283 Bounds
285 The BRE bounds are: *, specifying 0 or more matches of the preceding
286 atom; \{count\}, specifying that many repetitions; \{minimum,\}, giving
287 a lower limit; and \{minimum,maximum\} finally defines a lower and
288 upper bound.
289 A bound appearing as the first item in a BRE is taken literally.
291 Bracket Expressions
293 A bracket expression is a list of characters, character ranges and
294 character classes enclosed in [ and ] and matches any single character
295 from the represented set of characters.
296 A character range is written as two characters separated by - and
298 represents all characters (according to the character collating
299 sequence) that are not less than the first and not greater than the
300 second. (Ranges are very collating-sequence-dependent, and portable
301 programs should avoid relying on them.)
302 A character class is one of the class names
304 alnum digit punct
306 alpha graph space
307 blank lower upper
308 cntrl print xdigit
309 enclosed in [: and :] and represents the set of characters as defined
311 in ctype(3).
312 If the first character after [ is ^, the sense of matching is inverted.
314 To include a literal '"^"', place it anywhere else but first. To
316 include a literal '"]"' place it first or immediately after an initial
317 ^. To include a literal '"-"' make it the first (or second after ^) or
318 last character, or the second endpoint of a range.
319 The special bracket expression constructs "[[:<:]]" and "[[:>:]]" match
321 the null string at the beginning and end of a word respectively. (Note
322 that neither is identical to Perl's '\b' atom.)
323 Additional Atoms
325 Since some sed implementations provide additional regular expression
326 atoms (not defined in POSIX 1003.2), psed is capable of translating the
327 following backslash escapes:
328 \< This is the same as "[[:>:]]".
330 \> This is the same as "[[:<:]]".
331 \w This is an abbreviation for "[[:alnum:]_]".
332 \W This is an abbreviation for "[^[:alnum:]_]".
333 \y Match the empty string at a word boundary.
334 \B Match the empty string between any two either word or non-word
335 characters.
336 To enable this feature, the environment variable PSEDEXTBRE must be set
338 to a string containing the requested characters, e.g.:
339 "PSEDEXTBRE='<>wW'".
340
342 The environment variable "PSEDEXTBRE" may be set to extend BREs. See
343 "Additional Atoms".
344
346 ambiguous translation for character '%s' in 'y' command
347 The indicated character appears twice, with different translations.
348 '[' cannot be last in pattern
350 A '[' in a BRE indicates the beginning of a bracket expression.
351 '\' cannot be last in pattern
353 A '\' in a BRE is used to make the subsequent character literal.
354 '\' cannot be last in substitution
356 A '\' in a substitution string is used to make the subsequent
357 character literal.
358 conflicting flags '%s'
360 In an s command, either the 'g' flag and an n-th occurrence flag,
361 or multiple n-th occurrence flags are specified. Note that only the
362 digits ^'1' through '9' are permitted.
363 duplicate label %s (first defined at %s)
365 excess address(es)
366 The command has more than the permitted number of addresses.
367 extra characters after command (%s)
369 illegal option '%s'
370 improper delimiter in s command
371 The BRE and substitution may not be delimited with '\' or newline.
372 invalid address after ','
374 invalid backreference (%s)
375 The specified backreference number exceeds the number of
376 backreferences in the BRE.
377 invalid repeat clause '\{%s\}'
379 The repeat clause does not contain a valid integer value, or pair
380 of values.
381 malformed regex, 1st address
383 malformed regex, 2nd address
384 malformed regular expression
385 malformed substitution expression
386 malformed 'y' command argument
387 The first or second string of a y command is syntactically
388 incorrect.
389 maximum less than minimum in '\{%s\}'
391 no script command given
392 There must be at least one -e or one -f option specifying a script
393 or script file.
394 '\' not valid as delimiter in 'y' command
396 option -e requires an argument
397 option -f requires an argument
398 's' command requires argument
399 start of unterminated '{'
400 string lengths in 'y' command differ
401 The translation table strings in a y command must have equal
402 lengths.
403 undefined label '%s'
405 unexpected '}'
406 A } command without a preceding { command was encountered.
407 unexpected end of script
409 The end of the script was reached although a text line after a a, c
410 or i command indicated another line.
411 unknown command '%s'
413 unterminated '['
414 A BRE contains an unterminated bracket expression.
415 unterminated '\('
417 A BRE contains an unterminated backreference.
418 '\{' without closing '\}'
420 A BRE contains an unterminated bounds specification.
421 '\)' without preceding '\('
423 'y' command requires argument
424
426 The basic material for the preceding section was generated by running
427 the sed script
428 #no autoprint
430 s/^.*Warn( *"\([^"]*\)".*$/\1/
431 t process
432 b
433 :process
434 s/$!/%s/g
435 s/$[_[:alnum:]]\{1,\}/%s/g
436 s/\\\\/\\/g
437 s/^/=item /
438 p
439 on the program's own text, and piping the output into "sort -u".
441
443 If this program is invoked with the name s2p it will act as a sed-to-
444 Perl translator. After option processing (all other arguments are
445 ignored), a Perl program is printed on standard output, which will
446 process the input stream (as read from all arguments) in the way
447 defined by the sed script and the option setting used for the
448 translation.
449
451 perl(1), re_format(7)
452
454 The l command will show escape characters (ESC) as '"\e"', but a
455 vertical tab (VT) in octal.
456 Trailing spaces are truncated from labels in :, t and b commands.
458 The meaning of an empty regular expression ('"//"'), as defined by sed,
460 is "the last pattern used, at run time". This deviates from the Perl
461 interpretation, which will re-use the "last successfully executed
462 regular expression". Since keeping track of pattern usage would create
463 terribly cluttered code, and differences would only appear in obscure
464 context (where other sed implementations appear to deviate, too), the
465 Perl semantics was adopted. Note that common usage of this feature,
466 such as in "/abc/s//xyz/", will work as expected.
467 Collating elements (of bracket expressions in BREs) are not
469 implemented.
470
472 This sed implementation conforms to the IEEE Std1003.2-1992 ("POSIX.2")
473 definition of sed, and is compatible with the OpenBSD implementation,
474 except where otherwise noted (see "BUGS").
475
477 This Perl implementation of sed was written by Wolfgang Laun,
478 Wolfgang.Laun@alcatel.at.
479
481 This program is free and open software. You may use, modify,
482 distribute, and sell this program (and any modified variants) in any
483 way you wish, provided you do not restrict others from doing the same.
484perl v5.34.0 2021-07-22 PSED(1)