1GAWK(1) Utility Commands GAWK(1)
2
6 gawk - pattern scanning and processing language
7
9 gawk [ POSIX or GNU style options ] -f program-file [ -- ] file ...
10 gawk [ POSIX or GNU style options ] [ -- ] program-text file ...
11 pgawk [ POSIX or GNU style options ] -f program-file [ -- ] file ...
13 pgawk [ POSIX or GNU style options ] [ -- ] program-text file ...
14 dgawk [ POSIX or GNU style options ] -f program-file [ -- ] file ...
16
18 Gawk is the GNU Project's implementation of the AWK programming lan‐
19 guage. It conforms to the definition of the language in the POSIX
20 1003.1 Standard. This version in turn is based on the description in
21 The AWK Programming Language, by Aho, Kernighan, and Weinberger. Gawk
22 provides the additional features found in the current version of UNIX
23 awk and a number of GNU-specific extensions.
24 The command line consists of options to gawk itself, the AWK program
26 text (if not supplied via the -f or --file options), and values to be
27 made available in the ARGC and ARGV pre-defined AWK variables.
28 Pgawk is the profiling version of gawk. It is identical in every way
30 to gawk, except that programs run more slowly, and it automatically
31 produces an execution profile in the file awkprof.out when done. See
32 the --profile option, below.
33 Dgawk is an awk debugger. Instead of running the program directly, it
35 loads the AWK source code and then prompts for debugging commands.
36 Unlike gawk and pgawk, dgawk only processes AWK program source provided
37 with the -f option. The debugger is documented in GAWK: Effective AWK
38 Programming.
39
41 Gawk options may be either traditional POSIX-style one letter options,
42 or GNU-style long options. POSIX options start with a single “-”,
43 while long options start with “--”. Long options are provided for both
44 GNU-specific features and for POSIX-mandated features.
45 Gawk- specific options are typically used in long-option form. Argu‐
47 ments to long options are either joined with the option by an = sign,
48 with no intervening spaces, or they may be provided in the next command
49 line argument. Long options may be abbreviated, as long as the abbre‐
50 viation remains unique.
51 Additionally, each long option has a corresponding short option, so
53 that the option's functionality may be used from within #! executable
54 scripts.
55
57 Gawk accepts the following options. Standard options are listed first,
58 followed by options for gawk extensions, listed alphabetically by short
59 option.
60 -f program-file
62 --file program-file
63 Read the AWK program source from the file program-file, instead
64 of from the first command line argument. Multiple -f (or
65 --file) options may be used.
66 -F fs
68 --field-separator fs
69 Use fs for the input field separator (the value of the FS prede‐
70 fined variable).
71 -v var=val
73 --assign var=val
74 Assign the value val to the variable var, before execution of
75 the program begins. Such variable values are available to the
76 BEGIN block of an AWK program.
77 -b
79 --characters-as-bytes
80 Treat all input data as single-byte characters. In other words,
81 don't pay any attention to the locale information when attempt‐
82 ing to process strings as multibyte characters. The --posix
83 option overrides this one.
84 -c
86 --traditional
87 Run in compatibility mode. In compatibility mode, gawk behaves
88 identically to UNIX awk; none of the GNU-specific extensions are
89 recognized. See GNU EXTENSIONS, below, for more information.
90 -C
92 --copyright
93 Print the short version of the GNU copyright information message
94 on the standard output and exit successfully.
95 -d[file]
97 --dump-variables[=file]
98 Print a sorted list of global variables, their types and final
99 values to file. If no file is provided, gawk uses a file named
100 awkvars.out in the current directory.
101 Having a list of all the global variables is a good way to look
102 for typographical errors in your programs. You would also use
103 this option if you have a large program with a lot of functions,
104 and you want to be sure that your functions don't inadvertently
105 use global variables that you meant to be local. (This is a
106 particularly easy mistake to make with simple variable names
107 like i, j, and so on.)
108 -e program-text
110 --source program-text
111 Use program-text as AWK program source code. This option allows
112 the easy intermixing of library functions (used via the -f and
113 --file options) with source code entered on the command line.
114 It is intended primarily for medium to large AWK programs used
115 in shell scripts.
116 -E file
118 --exec file
119 Similar to -f, however, this is option is the last one pro‐
120 cessed. This should be used with #! scripts, particularly for
121 CGI applications, to avoid passing in options or source code (!)
122 on the command line from a URL. This option disables command-
123 line variable assignments.
124 -g
126 --gen-pot
127 Scan and parse the AWK program, and generate a GNU .pot (Porta‐
128 ble Object Template) format file on standard output with entries
129 for all localizable strings in the program. The program itself
130 is not executed. See the GNU gettext distribution for more
131 information on .pot files.
132 -h
134 --help Print a relatively short summary of the available options on the
135 standard output. (Per the GNU Coding Standards, these options
136 cause an immediate, successful exit.)
137 -L [value]
139 --lint[=value]
140 Provide warnings about constructs that are dubious or non-porta‐
141 ble to other AWK implementations. With an optional argument of
142 fatal, lint warnings become fatal errors. This may be drastic,
143 but its use will certainly encourage the development of cleaner
144 AWK programs. With an optional argument of invalid, only warn‐
145 ings about things that are actually invalid are issued. (This is
146 not fully implemented yet.)
147 -n
149 --non-decimal-data
150 Recognize octal and hexadecimal values in input data. Use this
151 option with great caution!
152 -N
154 --use-lc-numeric
155 This forces gawk to use the locale's decimal point character
156 when parsing input data. Although the POSIX standard requires
157 this behavior, and gawk does so when --posix is in effect, the
158 default is to follow traditional behavior and use a period as
159 the decimal point, even in locales where the period is not the
160 decimal point character. This option overrides the default
161 behavior, without the full draconian strictness of the --posix
162 option.
163 -O
165 --optimize
166 Enable optimizations upon the internal representation of the
167 program. Currently, this includes just simple constant-folding.
168 The gawk maintainer hopes to add additional optimizations over
169 time.
170 -p[prof_file]
172 --profile[=prof_file]
173 Send profiling data to prof_file. The default is awkprof.out.
174 When run with gawk, the profile is just a “pretty printed” ver‐
175 sion of the program. When run with pgawk, the profile contains
176 execution counts of each statement in the program in the left
177 margin and function call counts for each user-defined function.
178 -P
180 --posix
181 This turns on compatibility mode, with the following additional
182 restrictions:
183 · \x escape sequences are not recognized.
185 · Only space and tab act as field separators when FS is set to a
187 single space, newline does not.
188 · You cannot continue lines after ? and :.
190 · The synonym func for the keyword function is not recognized.
192 · The operators ** and **= cannot be used in place of ^ and ^=.
194 -r
196 --re-interval
197 Enable the use of interval expressions in regular expression
198 matching (see Regular Expressions, below). Interval expressions
199 were not traditionally available in the AWK language. The POSIX
200 standard added them, to make awk and egrep consistent with each
201 other. They are enabled by default, but this option remains for
202 use with --traditional.
203 -R
205 --command file
206 Dgawk only. Read stored debugger commands from file.
207 -S
209 --sandbox
210 Runs gawk in sandbox mode, disabling the system() function,
211 input redirection with getline, output redirection with print
212 and printf, and loading dynamic extensions. Command execution
213 (through pipelines) is also disabled. This effectively blocks a
214 script from accessing local resources (except for the files
215 specified on the command line).
216 -t
218 --lint-old
219 Provide warnings about constructs that are not portable to the
220 original version of Unix awk.
221 -V
223 --version
224 Print version information for this particular copy of gawk on
225 the standard output. This is useful mainly for knowing if the
226 current copy of gawk on your system is up to date with respect
227 to whatever the Free Software Foundation is distributing. This
228 is also useful when reporting bugs. (Per the GNU Coding Stan‐
229 dards, these options cause an immediate, successful exit.)
230 -- Signal the end of options. This is useful to allow further argu‐
232 ments to the AWK program itself to start with a “-”. This pro‐
233 vides consistency with the argument parsing convention used by
234 most other POSIX programs.
235 In compatibility mode, any other options are flagged as invalid, but
237 are otherwise ignored. In normal operation, as long as program text
238 has been supplied, unknown options are passed on to the AWK program in
239 the ARGV array for processing. This is particularly useful for running
240 AWK programs via the “#!” executable interpreter mechanism.
241
243 An AWK program consists of a sequence of pattern-action statements and
244 optional function definitions.
245 @include "filename" pattern { action statements }
247 function name(parameter list) { statements }
248 Gawk first reads the program source from the program-file(s) if speci‐
250 fied, from arguments to --source, or from the first non-option argument
251 on the command line. The -f and --source options may be used multiple
252 times on the command line. Gawk reads the program text as if all the
253 program-files and command line source texts had been concatenated
254 together. This is useful for building libraries of AWK functions,
255 without having to include them in each new AWK program that uses them.
256 It also provides the ability to mix library functions with command line
257 programs.
258 In addition, lines beginning with @include may be used to include other
260 source files into your program, making library use even easier.
261 The environment variable AWKPATH specifies a search path to use when
263 finding source files named with the -f option. If this variable does
264 not exist, the default path is ".:/usr/local/share/awk". (The actual
265 directory may vary, depending upon how gawk was built and installed.)
266 If a file name given to the -f option contains a “/” character, no path
267 search is performed.
268 Gawk executes AWK programs in the following order. First, all variable
270 assignments specified via the -v option are performed. Next, gawk com‐
271 piles the program into an internal form. Then, gawk executes the code
272 in the BEGIN block(s) (if any), and then proceeds to read each file
273 named in the ARGV array (up to ARGV[ARGC]). If there are no files
274 named on the command line, gawk reads the standard input.
275 If a filename on the command line has the form var=val it is treated as
277 a variable assignment. The variable var will be assigned the value
278 val. (This happens after any BEGIN block(s) have been run.) Command
279 line variable assignment is most useful for dynamically assigning val‐
280 ues to the variables AWK uses to control how input is broken into
281 fields and records. It is also useful for controlling state if multi‐
282 ple passes are needed over a single data file.
283 If the value of a particular element of ARGV is empty (""), gawk skips
285 over it.
286 For each input file, if a BEGINFILE rule exists, gawk executes the
288 associated code before processing the contents of the file. Similarly,
289 gawk executes the code associated with ENDFILE after processing the
290 file.
291 For each record in the input, gawk tests to see if it matches any pat‐
293 tern in the AWK program. For each pattern that the record matches, the
294 associated action is executed. The patterns are tested in the order
295 they occur in the program.
296 Finally, after all the input is exhausted, gawk executes the code in
298 the END block(s) (if any).
299 Command Line Directories
301 According to POSIX, files named on the awk command line must be text
302 files. The behavior is ``undefined'' if they are not. Most versions
303 of awk treat a directory on the command line as a fatal error.
304 Starting with version 4.0 of gawk, a directory on the command line pro‐
306 duces a warning, but is otherwise skipped. If either of the --posix or
307 --traditional options is given, then gawk reverts to treating directo‐
308 ries on the command line as a fatal error.
309
311 AWK variables are dynamic; they come into existence when they are first
312 used. Their values are either floating-point numbers or strings, or
313 both, depending upon how they are used. AWK also has one dimensional
314 arrays; arrays with multiple dimensions may be simulated. Several pre-
315 defined variables are set as a program runs; these are described as
316 needed and summarized below.
317 Records
319 Normally, records are separated by newline characters. You can control
320 how records are separated by assigning values to the built-in variable
321 RS. If RS is any single character, that character separates records.
322 Otherwise, RS is a regular expression. Text in the input that matches
323 this regular expression separates the record. However, in compatibil‐
324 ity mode, only the first character of its string value is used for sep‐
325 arating records. If RS is set to the null string, then records are
326 separated by blank lines. When RS is set to the null string, the new‐
327 line character always acts as a field separator, in addition to what‐
328 ever value FS may have.
329 Fields
331 As each input record is read, gawk splits the record into fields, using
332 the value of the FS variable as the field separator. If FS is a single
333 character, fields are separated by that character. If FS is the null
334 string, then each individual character becomes a separate field. Oth‐
335 erwise, FS is expected to be a full regular expression. In the special
336 case that FS is a single space, fields are separated by runs of spaces
337 and/or tabs and/or newlines. (But see the section POSIX COMPATIBILITY,
338 below). NOTE: The value of IGNORECASE (see below) also affects how
339 fields are split when FS is a regular expression, and how records are
340 separated when RS is a regular expression.
341 If the FIELDWIDTHS variable is set to a space separated list of num‐
343 bers, each field is expected to have fixed width, and gawk splits up
344 the record using the specified widths. The value of FS is ignored.
345 Assigning a new value to FS or FPAT overrides the use of FIELDWIDTHS.
346 Similarly, if the FPAT variable is set to a string representing a regu‐
348 lar expression, each field is made up of text that matches that regular
349 expression. In this case, the regular expression describes the fields
350 themselves, instead of the text that separates the fields. Assigning a
351 new value to FS or FIELDWIDTHS overrides the use of FPAT.
352 Each field in the input record may be referenced by its position, $1,
354 $2, and so on. $0 is the whole record. Fields need not be referenced
355 by constants:
356 n = 5
358 print $n
359 prints the fifth field in the input record.
361 The variable NF is set to the total number of fields in the input
363 record.
364 References to non-existent fields (i.e. fields after $NF) produce the
366 null-string. However, assigning to a non-existent field (e.g., $(NF+2)
367 = 5) increases the value of NF, creates any intervening fields with the
368 null string as their value, and causes the value of $0 to be recom‐
369 puted, with the fields being separated by the value of OFS. References
370 to negative numbered fields cause a fatal error. Decrementing NF
371 causes the values of fields past the new value to be lost, and the
372 value of $0 to be recomputed, with the fields being separated by the
373 value of OFS.
374 Assigning a value to an existing field causes the whole record to be
376 rebuilt when $0 is referenced. Similarly, assigning a value to $0
377 causes the record to be resplit, creating new values for the fields.
378 Built-in Variables
380 Gawk's built-in variables are:
381 ARGC The number of command line arguments (does not include
383 options to gawk, or the program source).
384 ARGIND The index in ARGV of the current file being processed.
386 ARGV Array of command line arguments. The array is indexed from
388 0 to ARGC - 1. Dynamically changing the contents of ARGV
389 can control the files used for data.
390 BINMODE On non-POSIX systems, specifies use of “binary” mode for
392 all file I/O. Numeric values of 1, 2, or 3, specify that
393 input files, output files, or all files, respectively,
394 should use binary I/O. String values of "r", or "w" spec‐
395 ify that input files, or output files, respectively, should
396 use binary I/O. String values of "rw" or "wr" specify that
397 all files should use binary I/O. Any other string value is
398 treated as "rw", but generates a warning message.
399 CONVFMT The conversion format for numbers, "%.6g", by default.
401 ENVIRON An array containing the values of the current environment.
403 The array is indexed by the environment variables, each
404 element being the value of that variable (e.g., ENVI‐
405 RON["HOME"] might be /home/arnold). Changing this array
406 does not affect the environment seen by programs which gawk
407 spawns via redirection or the system() function.
408 ERRNO If a system error occurs either doing a redirection for
410 getline, during a read for getline, or during a close(),
411 then ERRNO will contain a string describing the error. The
412 value is subject to translation in non-English locales.
413 FIELDWIDTHS A whitespace separated list of field widths. When set,
415 gawk parses the input into fields of fixed width, instead
416 of using the value of the FS variable as the field separa‐
417 tor. See Fields, above.
418 FILENAME The name of the current input file. If no files are speci‐
420 fied on the command line, the value of FILENAME is “-”.
421 However, FILENAME is undefined inside the BEGIN block
422 (unless set by getline).
423 FNR The input record number in the current input file.
425 FPAT A regular expression describing the contents of the fields
427 in a record. When set, gawk parses the input into fields,
428 where the fields match the regular expression, instead of
429 using the value of the FS variable as the field separator.
430 See Fields, above.
431 FS The input field separator, a space by default. See Fields,
433 above.
434 IGNORECASE Controls the case-sensitivity of all regular expression and
436 string operations. If IGNORECASE has a non-zero value,
437 then string comparisons and pattern matching in rules,
438 field splitting with FS and FPAT, record separating with
439 RS, regular expression matching with ~ and !~, and the gen‐
440 sub(), gsub(), index(), match(), patsplit(), split(), and
441 sub() built-in functions all ignore case when doing regular
442 expression operations. NOTE: Array subscripting is not
443 affected. However, the asort() and asorti() functions are
444 affected.
445 Thus, if IGNORECASE is not equal to zero, /aB/ matches all
446 of the strings "ab", "aB", "Ab", and "AB". As with all AWK
447 variables, the initial value of IGNORECASE is zero, so all
448 regular expression and string operations are normally case-
449 sensitive.
450 LINT Provides dynamic control of the --lint option from within
452 an AWK program. When true, gawk prints lint warnings. When
453 false, it does not. When assigned the string value
454 "fatal", lint warnings become fatal errors, exactly like
455 --lint=fatal. Any other true value just prints warnings.
456 NF The number of fields in the current input record.
458 NR The total number of input records seen so far.
460 OFMT The output format for numbers, "%.6g", by default.
462 OFS The output field separator, a space by default.
464 ORS The output record separator, by default a newline.
466 PROCINFO The elements of this array provide access to information
468 about the running AWK program. On some systems, there may
469 be elements in the array, "group1" through "groupn" for
470 some n, which is the number of supplementary groups that
471 the process has. Use the in operator to test for these
472 elements. The following elements are guaranteed to be
473 available:
474 PROCINFO["egid"] the value of the getegid(2) system
476 call.
477 PROCINFO["strftime"]
479 The default time format string for
480 strftime().
481 PROCINFO["euid"] the value of the geteuid(2) system
483 call.
484 PROCINFO["FS"] "FS" if field splitting with FS is in
486 effect, "FPAT" if field splitting with
487 FPAT is in effect, or "FIELDWIDTHS" if
488 field splitting with FIELDWIDTHS is in
489 effect.
490 PROCINFO["gid"] the value of the getgid(2) system call.
492 PROCINFO["pgrpid"] the process group ID of the current
494 process.
495 PROCINFO["pid"] the process ID of the current process.
497 PROCINFO["ppid"] the parent process ID of the current
499 process.
500 PROCINFO["uid"] the value of the getuid(2) system call.
502 PROCINFO["sorted_in"]
504 If this element exists in PROCINFO,
505 then its value controls the order in
506 which array elements are traversed in
507 for loops. Supported values are
508 "@ind_str_asc", "@ind_num_asc",
509 "@val_type_asc", "@val_str_asc",
510 "@val_num_asc", "@ind_str_desc",
511 "@ind_num_desc", "@val_type_desc",
512 "@val_str_desc", "@val_num_desc", and
513 "@unsorted". The value can also be the
514 name of any comparison function defined
515 as follows:
516 function cmp_func(i1, v1, i2, v2)
518 where i1 and i2 are the indices, and v1 and v2 are the cor‐
520 responding values of the two elements being compared. It
521 should return a number less than, equal to, or greater than
522 0, depending on how the elements of the array are to be
523 ordered.
524 PROCINFO["version"]
526 the version of gawk.
527 RS The input record separator, by default a newline.
529 RT The record terminator. Gawk sets RT to the input text that
531 matched the character or regular expression specified by
532 RS.
533 RSTART The index of the first character matched by match(); 0 if
535 no match. (This implies that character indices start at
536 one.)
537 RLENGTH The length of the string matched by match(); -1 if no
539 match.
540 SUBSEP The character used to separate multiple subscripts in array
542 elements, by default "\034".
543 TEXTDOMAIN The text domain of the AWK program; used to find the local‐
545 ized translations for the program's strings.
546 Arrays
548 Arrays are subscripted with an expression between square brackets ([
549 and ]). If the expression is an expression list (expr, expr ...) then
550 the array subscript is a string consisting of the concatenation of the
551 (string) value of each expression, separated by the value of the SUBSEP
552 variable. This facility is used to simulate multiply dimensioned
553 arrays. For example:
554 i = "A"; j = "B"; k = "C"
556 x[i, j, k] = "hello, world\n"
557 assigns the string "hello, world\n" to the element of the array x which
559 is indexed by the string "A\034B\034C". All arrays in AWK are associa‐
560 tive, i.e. indexed by string values.
561 The special operator in may be used to test if an array has an index
563 consisting of a particular value:
564 if (val in array)
566 print array[val]
567 If the array has multiple subscripts, use (i, j) in array.
569 The in construct may also be used in a for loop to iterate over all the
571 elements of an array.
572 An element may be deleted from an array using the delete statement.
574 The delete statement may also be used to delete the entire contents of
575 an array, just by specifying the array name without a subscript.
576 gawk supports true multidimensional arrays. It does not require that
578 such arrays be ``rectangular'' as in C or C++. For example:
579 a[1] = 5
580 a[2][1] = 6
581 a[2][2] = 7
582 Variable Typing And Conversion
584 Variables and fields may be (floating point) numbers, or strings, or
585 both. How the value of a variable is interpreted depends upon its con‐
586 text. If used in a numeric expression, it will be treated as a number;
587 if used as a string it will be treated as a string.
588 To force a variable to be treated as a number, add 0 to it; to force it
590 to be treated as a string, concatenate it with the null string.
591 When a string must be converted to a number, the conversion is accom‐
593 plished using strtod(3). A number is converted to a string by using
594 the value of CONVFMT as a format string for sprintf(3), with the
595 numeric value of the variable as the argument. However, even though
596 all numbers in AWK are floating-point, integral values are always con‐
597 verted as integers. Thus, given
598 CONVFMT = "%2.2f"
600 a = 12
601 b = a ""
602 the variable b has a string value of "12" and not "12.00".
604 NOTE: When operating in POSIX mode (such as with the --posix command
606 line option), beware that locale settings may interfere with the way
607 decimal numbers are treated: the decimal separator of the numbers you
608 are feeding to gawk must conform to what your locale would expect, be
609 it a comma (,) or a period (.).
610 Gawk performs comparisons as follows: If two variables are numeric,
612 they are compared numerically. If one value is numeric and the other
613 has a string value that is a “numeric string,” then comparisons are
614 also done numerically. Otherwise, the numeric value is converted to a
615 string and a string comparison is performed. Two strings are compared,
616 of course, as strings.
617 Note that string constants, such as "57", are not numeric strings, they
619 are string constants. The idea of “numeric string” only applies to
620 fields, getline input, FILENAME, ARGV elements, ENVIRON elements and
621 the elements of an array created by split() or patsplit() that are
622 numeric strings. The basic idea is that user input, and only user
623 input, that looks numeric, should be treated that way.
624 Uninitialized variables have the numeric value 0 and the string value
626 "" (the null, or empty, string).
627 Octal and Hexadecimal Constants
629 You may use C-style octal and hexadecimal constants in your AWK program
630 source code. For example, the octal value 011 is equal to decimal 9,
631 and the hexadecimal value 0x11 is equal to decimal 17.
632 String Constants
634 String constants in AWK are sequences of characters enclosed between
635 double quotes (like "value"). Within strings, certain escape sequences
636 are recognized, as in C. These are:
637 \\ A literal backslash.
639 \a The “alert” character; usually the ASCII BEL character.
641 \b backspace.
643 \f form-feed.
645 \n newline.
647 \r carriage return.
649 \t horizontal tab.
651 \v vertical tab.
653 \xhex digits
655 The character represented by the string of hexadecimal digits fol‐
656 lowing the \x. As in ANSI C, all following hexadecimal digits are
657 considered part of the escape sequence. (This feature should tell
658 us something about language design by committee.) E.g., "\x1B" is
659 the ASCII ESC (escape) character.
660 \ddd The character represented by the 1-, 2-, or 3-digit sequence of
662 octal digits. E.g., "\033" is the ASCII ESC (escape) character.
663 \c The literal character c.
665 The escape sequences may also be used inside constant regular expres‐
667 sions (e.g., /[ \t\f\n\r\v]/ matches whitespace characters).
668 In compatibility mode, the characters represented by octal and hexadec‐
670 imal escape sequences are treated literally when used in regular
671 expression constants. Thus, /a\52b/ is equivalent to /a\*b/.
672
674 AWK is a line-oriented language. The pattern comes first, and then the
675 action. Action statements are enclosed in { and }. Either the pattern
676 may be missing, or the action may be missing, but, of course, not both.
677 If the pattern is missing, the action is executed for every single
678 record of input. A missing action is equivalent to
679 { print }
681 which prints the entire record.
683 Comments begin with the # character, and continue until the end of the
685 line. Blank lines may be used to separate statements. Normally, a
686 statement ends with a newline, however, this is not the case for lines
687 ending in a comma, {, ?, :, &&, or ||. Lines ending in do or else also
688 have their statements automatically continued on the following line.
689 In other cases, a line can be continued by ending it with a “\”, in
690 which case the newline is ignored.
691 Multiple statements may be put on one line by separating them with a
693 “;”. This applies to both the statements within the action part of a
694 pattern-action pair (the usual case), and to the pattern-action state‐
695 ments themselves.
696 Patterns
698 AWK patterns may be one of the following:
699 BEGIN
701 END
702 BEGINFILE
703 ENDFILE
704 /regular expression/
705 relational expression
706 pattern && pattern
707 pattern || pattern
708 pattern ? pattern : pattern
709 (pattern)
710 ! pattern
711 pattern1, pattern2
712 BEGIN and END are two special kinds of patterns which are not tested
714 against the input. The action parts of all BEGIN patterns are merged
715 as if all the statements had been written in a single BEGIN block.
716 They are executed before any of the input is read. Similarly, all the
717 END blocks are merged, and executed when all the input is exhausted (or
718 when an exit statement is executed). BEGIN and END patterns cannot be
719 combined with other patterns in pattern expressions. BEGIN and END
720 patterns cannot have missing action parts.
721 BEGINFILE and ENDFILE are additional special patterns whose bodies are
723 executed before reading the first record of each command line input
724 file and after reading the last record of each file. Inside the BEGIN‐
725 FILE rule, the value of ERRNO will be the empty string if the file
726 could be opened successfully. Otherwise, there is some problem with
727 the file and the code should use nextfile to skip it. If that is not
728 done, gawk produces its usual fatal error for files that cannot be
729 opened.
730 For /regular expression/ patterns, the associated statement is executed
732 for each input record that matches the regular expression. Regular
733 expressions are the same as those in egrep(1), and are summarized
734 below.
735 A relational expression may use any of the operators defined below in
737 the section on actions. These generally test whether certain fields
738 match certain regular expressions.
739 The &&, ||, and ! operators are logical AND, logical OR, and logical
741 NOT, respectively, as in C. They do short-circuit evaluation, also as
742 in C, and are used for combining more primitive pattern expressions.
743 As in most languages, parentheses may be used to change the order of
744 evaluation.
745 The ?: operator is like the same operator in C. If the first pattern
747 is true then the pattern used for testing is the second pattern, other‐
748 wise it is the third. Only one of the second and third patterns is
749 evaluated.
750 The pattern1, pattern2 form of an expression is called a range pattern.
752 It matches all input records starting with a record that matches pat‐
753 tern1, and continuing until a record that matches pattern2, inclusive.
754 It does not combine with any other sort of pattern expression.
755 Regular Expressions
757 Regular expressions are the extended kind found in egrep. They are
758 composed of characters as follows:
759 c matches the non-metacharacter c.
761 \c matches the literal character c.
763 . matches any character including newline.
765 ^ matches the beginning of a string.
767 $ matches the end of a string.
769 [abc...] character list, matches any of the characters abc....
771 [^abc...] negated character list, matches any character except abc....
773 r1|r2 alternation: matches either r1 or r2.
775 r1r2 concatenation: matches r1, and then r2.
777 r+ matches one or more r's.
779 r* matches zero or more r's.
781 r? matches zero or one r's.
783 (r) grouping: matches r.
785 r{n}
787 r{n,}
788 r{n,m} One or two numbers inside braces denote an interval expres‐
789 sion. If there is one number in the braces, the preceding
790 regular expression r is repeated n times. If there are two
791 numbers separated by a comma, r is repeated n to m times.
792 If there is one number followed by a comma, then r is
793 repeated at least n times.
794 \y matches the empty string at either the beginning or the end
796 of a word.
797 \B matches the empty string within a word.
799 \< matches the empty string at the beginning of a word.
801 \> matches the empty string at the end of a word.
803 \s matches any whitespace character.
805 \S matches any nonwhitespace character.
807 \w matches any word-constituent character (letter, digit, or
809 underscore).
810 \W matches any character that is not word-constituent.
812 \` matches the empty string at the beginning of a buffer
814 (string).
815 \' matches the empty string at the end of a buffer.
817 The escape sequences that are valid in string constants (see below) are
819 also valid in regular expressions.
820 Character classes are a feature introduced in the POSIX standard. A
822 character class is a special notation for describing lists of charac‐
823 ters that have a specific attribute, but where the actual characters
824 themselves can vary from country to country and/or from character set
825 to character set. For example, the notion of what is an alphabetic
826 character differs in the USA and in France.
827 A character class is only valid in a regular expression inside the
829 brackets of a character list. Character classes consist of [:, a key‐
830 word denoting the class, and :]. The character classes defined by the
831 POSIX standard are:
832 [:alnum:] Alphanumeric characters.
834 [:alpha:] Alphabetic characters.
836 [:blank:] Space or tab characters.
838 [:cntrl:] Control characters.
840 [:digit:] Numeric characters.
842 [:graph:] Characters that are both printable and visible. (A space is
844 printable, but not visible, while an a is both.)
845 [:lower:] Lowercase alphabetic characters.
847 [:print:] Printable characters (characters that are not control char‐
849 acters.)
850 [:punct:] Punctuation characters (characters that are not letter, dig‐
852 its, control characters, or space characters).
853 [:space:] Space characters (such as space, tab, and formfeed, to name
855 a few).
856 [:upper:] Uppercase alphabetic characters.
858 [:xdigit:] Characters that are hexadecimal digits.
860 For example, before the POSIX standard, to match alphanumeric charac‐
862 ters, you would have had to write /[A-Za-z0-9]/. If your character set
863 had other alphabetic characters in it, this would not match them, and
864 if your character set collated differently from ASCII, this might not
865 even match the ASCII alphanumeric characters. With the POSIX character
866 classes, you can write /[[:alnum:]]/, and this matches the alphabetic
867 and numeric characters in your character set, no matter what it is.
868 Two additional special sequences can appear in character lists. These
870 apply to non-ASCII character sets, which can have single symbols
871 (called collating elements) that are represented with more than one
872 character, as well as several characters that are equivalent for col‐
873 lating, or sorting, purposes. (E.g., in French, a plain “e” and a
874 grave-accented “e`” are equivalent.)
875 Collating Symbols
877 A collating symbol is a multi-character collating element
878 enclosed in [. and .]. For example, if ch is a collating ele‐
879 ment, then [[.ch.]] is a regular expression that matches this
880 collating element, while [ch] is a regular expression that
881 matches either c or h.
882 Equivalence Classes
884 An equivalence class is a locale-specific name for a list of
885 characters that are equivalent. The name is enclosed in [= and
886 =]. For example, the name e might be used to represent all of
887 “e,” “e´,” and “e`.” In this case, [[=e=]] is a regular expres‐
888 sion that matches any of e, e´, or e`.
889 These features are very valuable in non-English speaking locales. The
891 library functions that gawk uses for regular expression matching cur‐
892 rently only recognize POSIX character classes; they do not recognize
893 collating symbols or equivalence classes.
894 The \y, \B, \<, \>, \s, \S, \w, \W, \`, and \' operators are specific
896 to gawk; they are extensions based on facilities in the GNU regular
897 expression libraries.
898 The various command line options control how gawk interprets characters
900 in regular expressions.
901 No options
903 In the default case, gawk provide all the facilities of POSIX
904 regular expressions and the GNU regular expression operators
905 described above.
906 --posix
908 Only POSIX regular expressions are supported, the GNU operators
909 are not special. (E.g., \w matches a literal w).
910 --traditional
912 Traditional Unix awk regular expressions are matched. The GNU
913 operators are not special, and interval expressions are not
914 available. Characters described by octal and hexadecimal escape
915 sequences are treated literally, even if they represent regular
916 expression metacharacters.
917 --re-interval
919 Allow interval expressions in regular expressions, even if
920 --traditional has been provided.
921 Actions
923 Action statements are enclosed in braces, { and }. Action statements
924 consist of the usual assignment, conditional, and looping statements
925 found in most languages. The operators, control statements, and
926 input/output statements available are patterned after those in C.
927 Operators
929 The operators in AWK, in order of decreasing precedence, are
930 (...) Grouping
932 $ Field reference.
934 ++ -- Increment and decrement, both prefix and postfix.
936 ^ Exponentiation (** may also be used, and **= for the
938 assignment operator).
939 + - ! Unary plus, unary minus, and logical negation.
941 * / % Multiplication, division, and modulus.
943 + - Addition and subtraction.
945 space String concatenation.
947 | |& Piped I/O for getline, print, and printf.
949 < > <= >= != ==
951 The regular relational operators.
952 ~ !~ Regular expression match, negated match. NOTE: Do not use
954 a constant regular expression (/foo/) on the left-hand side
955 of a ~ or !~. Only use one on the right-hand side. The
956 expression /foo/ ~ exp has the same meaning as (($0 ~
957 /foo/) ~ exp). This is usually not what was intended.
958 in Array membership.
960 && Logical AND.
962 || Logical OR.
964 ?: The C conditional expression. This has the form expr1 ?
966 expr2 : expr3. If expr1 is true, the value of the expres‐
967 sion is expr2, otherwise it is expr3. Only one of expr2
968 and expr3 is evaluated.
969 = += -= *= /= %= ^=
971 Assignment. Both absolute assignment (var = value) and
972 operator-assignment (the other forms) are supported.
973 Control Statements
975 The control statements are as follows:
976 if (condition) statement [ else statement ]
978 while (condition) statement
979 do statement while (condition)
980 for (expr1; expr2; expr3) statement
981 for (var in array) statement
982 break
983 continue
984 delete array[index]
985 delete array
986 exit [ expression ]
987 { statements }
988 switch (expression) {
989 case value|regex : statement
990 ...
991 [ default: statement ]
992 }
993 I/O Statements
995 The input/output statements are as follows:
996 close(file [, how]) Close file, pipe or co-process. The optional how
998 should only be used when closing one end of a
999 two-way pipe to a co-process. It must be a
1000 string value, either "to" or "from".
1001 getline Set $0 from next input record; set NF, NR, FNR.
1003 getline <file Set $0 from next record of file; set NF.
1005 getline var Set var from next input record; set NR, FNR.
1007 getline var <file Set var from next record of file.
1009 command | getline [var]
1011 Run command piping the output either into $0 or
1012 var, as above.
1013 command |& getline [var]
1015 Run command as a co-process piping the output
1016 either into $0 or var, as above. Co-processes
1017 are a gawk extension. (command can also be a
1018 socket. See the subsection Special File Names,
1019 below.)
1020 next Stop processing the current input record. The
1022 next input record is read and processing starts
1023 over with the first pattern in the AWK program.
1024 If the end of the input data is reached, the END
1025 block(s), if any, are executed.
1026 nextfile Stop processing the current input file. The next
1028 input record read comes from the next input file.
1029 FILENAME and ARGIND are updated, FNR is reset to
1030 1, and processing starts over with the first pat‐
1031 tern in the AWK program. If the end of the input
1032 data is reached, the END block(s), if any, are
1033 executed.
1034 print Print the current record. The output record is
1036 terminated with the value of the ORS variable.
1037 print expr-list Print expressions. Each expression is separated
1039 by the value of the OFS variable. The output
1040 record is terminated with the value of the ORS
1041 variable.
1042 print expr-list >file Print expressions on file. Each expression is
1044 separated by the value of the OFS variable. The
1045 output record is terminated with the value of the
1046 ORS variable.
1047 printf fmt, expr-list Format and print. See The printf Statement,
1049 below.
1050 printf fmt, expr-list >file
1052 Format and print on file.
1053 system(cmd-line) Execute the command cmd-line, and return the exit
1055 status. (This may not be available on non-POSIX
1056 systems.)
1057 fflush([file]) Flush any buffers associated with the open output
1059 file or pipe file. If file is missing or if it
1060 is the null string, then flush all open output
1061 files and pipes.
1062 Additional output redirections are allowed for print and printf.
1064 print ... >> file
1066 Appends output to the file.
1067 print ... | command
1069 Writes on a pipe.
1070 print ... |& command
1072 Sends data to a co-process or socket. (See also the subsection
1073 Special File Names, below.)
1074 The getline command returns 1 on success, 0 on end of file, and -1 on
1076 an error. Upon an error, ERRNO contains a string describing the prob‐
1077 lem.
1078 NOTE: Failure in opening a two-way socket will result in a non-fatal
1080 error being returned to the calling function. If using a pipe, co-
1081 process, or socket to getline, or from print or printf within a loop,
1082 you must use close() to create new instances of the command or socket.
1083 AWK does not automatically close pipes, sockets, or co-processes when
1084 they return EOF.
1085 The printf Statement
1087 The AWK versions of the printf statement and sprintf() function (see
1088 below) accept the following conversion specification formats:
1089 %c A single character. If the argument used for %c is numeric, it
1091 is treated as a character and printed. Otherwise, the argument
1092 is assumed to be a string, and the only first character of that
1093 string is printed.
1094 %d, %i A decimal number (the integer part).
1096 %e, %E A floating point number of the form [-]d.dddddde[+-]dd. The %E
1098 format uses E instead of e.
1099 %f, %F A floating point number of the form [-]ddd.dddddd. If the sys‐
1101 tem library supports it, %F is available as well. This is like
1102 %f, but uses capital letters for special “not a number” and
1103 “infinity” values. If %F is not available, gawk uses %f.
1104 %g, %G Use %e or %f conversion, whichever is shorter, with nonsignifi‐
1106 cant zeros suppressed. The %G format uses %E instead of %e.
1107 %o An unsigned octal number (also an integer).
1109 %u An unsigned decimal number (again, an integer).
1111 %s A character string.
1113 %x, %X An unsigned hexadecimal number (an integer). The %X format
1115 uses ABCDEF instead of abcdef.
1116 %% A single % character; no argument is converted.
1118 Optional, additional parameters may lie between the % and the control
1120 letter:
1121 count$ Use the count'th argument at this point in the formatting. This
1123 is called a positional specifier and is intended primarily for
1124 use in translated versions of format strings, not in the origi‐
1125 nal text of an AWK program. It is a gawk extension.
1126 - The expression should be left-justified within its field.
1128 space For numeric conversions, prefix positive values with a space,
1130 and negative values with a minus sign.
1131 + The plus sign, used before the width modifier (see below), says
1133 to always supply a sign for numeric conversions, even if the
1134 data to be formatted is positive. The + overrides the space
1135 modifier.
1136 # Use an “alternate form” for certain control letters. For %o,
1138 supply a leading zero. For %x, and %X, supply a leading 0x or
1139 0X for a nonzero result. For %e, %E, %f and %F, the result
1140 always contains a decimal point. For %g, and %G, trailing zeros
1141 are not removed from the result.
1142 0 A leading 0 (zero) acts as a flag, that indicates output should
1144 be padded with zeroes instead of spaces. This applies only to
1145 the numeric output formats. This flag only has an effect when
1146 the field width is wider than the value to be printed.
1147 width The field should be padded to this width. The field is normally
1149 padded with spaces. If the 0 flag has been used, it is padded
1150 with zeroes.
1151 .prec A number that specifies the precision to use when printing. For
1153 the %e, %E, %f and %F, formats, this specifies the number of
1154 digits you want printed to the right of the decimal point. For
1155 the %g, and %G formats, it specifies the maximum number of sig‐
1156 nificant digits. For the %d, %i, %o, %u, %x, and %X formats, it
1157 specifies the minimum number of digits to print. For %s, it
1158 specifies the maximum number of characters from the string that
1159 should be printed.
1160 The dynamic width and prec capabilities of the ANSI C printf() routines
1162 are supported. A * in place of either the width or prec specifications
1163 causes their values to be taken from the argument list to printf or
1164 sprintf(). To use a positional specifier with a dynamic width or pre‐
1165 cision, supply the count$ after the * in the format string. For exam‐
1166 ple, "%3$*2$.*1$s".
1167 Special File Names
1169 When doing I/O redirection from either print or printf into a file, or
1170 via getline from a file, gawk recognizes certain special filenames
1171 internally. These filenames allow access to open file descriptors
1172 inherited from gawk's parent process (usually the shell). These file
1173 names may also be used on the command line to name data files. The
1174 filenames are:
1175 /dev/stdin The standard input.
1177 /dev/stdout The standard output.
1179 /dev/stderr The standard error output.
1181 /dev/fd/n The file associated with the open file descriptor n.
1183 These are particularly useful for error messages. For example:
1185 print "You blew it!" > "/dev/stderr"
1187 whereas you would otherwise have to use
1189 print "You blew it!" | "cat 1>&2"
1191 The following special filenames may be used with the |& co-process
1193 operator for creating TCP/IP network connections:
1194 /inet/tcp/lport/rhost/rport
1196 /inet4/tcp/lport/rhost/rport
1197 /inet6/tcp/lport/rhost/rport
1198 Files for a TCP/IP connection on local port lport to remote host
1199 rhost on remote port rport. Use a port of 0 to have the system
1200 pick a port. Use /inet4 to force an IPv4 connection, and /inet6
1201 to force an IPv6 connection. Plain /inet uses the system
1202 default (most likely IPv4).
1203 /inet/udp/lport/rhost/rport
1205 /inet4/udp/lport/rhost/rport
1206 /inet6/udp/lport/rhost/rport
1207 Similar, but use UDP/IP instead of TCP/IP.
1208 Numeric Functions
1210 AWK has the following built-in arithmetic functions:
1211 atan2(y, x) Return the arctangent of y/x in radians.
1213 cos(expr) Return the cosine of expr, which is in radians.
1215 exp(expr) The exponential function.
1217 int(expr) Truncate to integer.
1219 log(expr) The natural logarithm function.
1221 rand() Return a random number N, between 0 and 1, such that 0 ≤
1223 N < 1.
1224 sin(expr) Return the sine of expr, which is in radians.
1226 sqrt(expr) The square root function.
1228 srand([expr]) Use expr as the new seed for the random number generator.
1230 If no expr is provided, use the time of day. The return
1231 value is the previous seed for the random number genera‐
1232 tor.
1233 String Functions
1235 Gawk has the following built-in string functions:
1236 asort(s [, d [, how] ]) Return the number of elements in the source
1238 array s. Sort the contents of s using gawk's
1239 normal rules for comparing values, and replace
1240 the indices of the sorted values s with sequen‐
1241 tial integers starting with 1. If the optional
1242 destination array d is specified, then first
1243 duplicate s into d, and then sort d, leaving
1244 the indices of the source array s unchanged.
1245 The optional string how controls the direction
1246 and the comparison mode. Valid values for how
1247 are any of the strings valid for
1248 PROCINFO["sorted_in"]. It can also be the name
1249 of a user-defined comparison function as
1250 described in PROCINFO["sorted_in"].
1251 asorti(s [, d [, how] ])
1253 Return the number of elements in the source
1254 array s. The behavior is the same as that of
1255 asort(), except that the array indices are used
1256 for sorting, not the array values. When done,
1257 the array is indexed numerically, and the val‐
1258 ues are those of the original indices. The
1259 original values are lost; thus provide a second
1260 array if you wish to preserve the original.
1261 The purpose of the optional string how is the
1262 same as described in asort() above.
1263 gensub(r, s, h [, t]) Search the target string t for matches of the
1265 regular expression r. If h is a string begin‐
1266 ning with g or G, then replace all matches of r
1267 with s. Otherwise, h is a number indicating
1268 which match of r to replace. If t is not sup‐
1269 plied, use $0 instead. Within the replacement
1270 text s, the sequence \n, where n is a digit
1271 from 1 to 9, may be used to indicate just the
1272 text that matched the n'th parenthesized subex‐
1273 pression. The sequence \0 represents the
1274 entire matched text, as does the character &.
1275 Unlike sub() and gsub(), the modified string is
1276 returned as the result of the function, and the
1277 original target string is not changed.
1278 gsub(r, s [, t]) For each substring matching the regular expres‐
1280 sion r in the string t, substitute the string
1281 s, and return the number of substitutions. If
1282 t is not supplied, use $0. An & in the
1283 replacement text is replaced with the text that
1284 was actually matched. Use \& to get a literal
1285 &. (This must be typed as "\\&"; see GAWK:
1286 Effective AWK Programming for a fuller discus‐
1287 sion of the rules for &'s and backslashes in
1288 the replacement text of sub(), gsub(), and gen‐
1289 sub().)
1290 index(s, t) Return the index of the string t in the string
1292 s, or 0 if t is not present. (This implies
1293 that character indices start at one.)
1294 length([s]) Return the length of the string s, or the
1296 length of $0 if s is not supplied. As a non-
1297 standard extension, with an array argument,
1298 length() returns the number of elements in the
1299 array.
1300 match(s, r [, a]) Return the position in s where the regular
1302 expression r occurs, or 0 if r is not present,
1303 and set the values of RSTART and RLENGTH. Note
1304 that the argument order is the same as for the
1305 ~ operator: str ~ re. If array a is provided,
1306 a is cleared and then elements 1 through n are
1307 filled with the portions of s that match the
1308 corresponding parenthesized subexpression in r.
1309 The 0'th element of a contains the portion of s
1310 matched by the entire regular expression r.
1311 Subscripts a[n, "start"], and a[n, "length"]
1312 provide the starting index in the string and
1313 length respectively, of each matching sub‐
1314 string.
1315 patsplit(s, a [, r [, seps] ])
1317 Split the string s into the array a and the
1318 separators array seps on the regular expression
1319 r, and return the number of fields. Element
1320 values are the portions of s that matched r.
1321 The value of seps[i] is the separator that
1322 appeared in front of a[i+1]. If r is omitted,
1323 FPAT is used instead. The arrays a and seps
1324 are cleared first. Splitting behaves identi‐
1325 cally to field splitting with FPAT, described
1326 above.
1327 split(s, a [, r [, seps] ])
1329 Split the string s into the array a and the
1330 separators array seps on the regular expression
1331 r, and return the number of fields. If r is
1332 omitted, FS is used instead. The arrays a and
1333 seps are cleared first. seps[i] is the field
1334 separator matched by r between a[i] and a[i+1].
1335 If r is a single space, then leading whitespace
1336 in s goes into the extra array element seps[0]
1337 and trailing whitespace goes into the extra
1338 array element seps[n], where n is the return
1339 value of split(s, a, r, seps). Splitting
1340 behaves identically to field splitting,
1341 described above.
1342 sprintf(fmt, expr-list) Prints expr-list according to fmt, and returns
1344 the resulting string.
1345 strtonum(str) Examine str, and return its numeric value. If
1347 str begins with a leading 0, strtonum() assumes
1348 that str is an octal number. If str begins
1349 with a leading 0x or 0X, strtonum() assumes
1350 that str is a hexadecimal number. Otherwise,
1351 decimal is assumed.
1352 sub(r, s [, t]) Just like gsub(), but replace only the first
1354 matching substring.
1355 substr(s, i [, n]) Return the at most n-character substring of s
1357 starting at i. If n is omitted, use the rest
1358 of s.
1359 tolower(str) Return a copy of the string str, with all the
1361 uppercase characters in str translated to their
1362 corresponding lowercase counterparts. Non-
1363 alphabetic characters are left unchanged.
1364 toupper(str) Return a copy of the string str, with all the
1366 lowercase characters in str translated to their
1367 corresponding uppercase counterparts. Non-
1368 alphabetic characters are left unchanged.
1369 Gawk is multibyte aware. This means that index(), length(), substr()
1371 and match() all work in terms of characters, not bytes.
1372 Time Functions
1374 Since one of the primary uses of AWK programs is processing log files
1375 that contain time stamp information, gawk provides the following func‐
1376 tions for obtaining time stamps and formatting them.
1377 mktime(datespec)
1379 Turn datespec into a time stamp of the same form as returned
1380 by systime(), and return the result. The datespec is a
1381 string of the form YYYY MM DD HH MM SS[ DST]. The contents
1382 of the string are six or seven numbers representing respec‐
1383 tively the full year including century, the month from 1 to
1384 12, the day of the month from 1 to 31, the hour of the day
1385 from 0 to 23, the minute from 0 to 59, the second from 0 to
1386 60, and an optional daylight saving flag. The values of
1387 these numbers need not be within the ranges specified; for
1388 example, an hour of -1 means 1 hour before midnight. The
1389 origin-zero Gregorian calendar is assumed, with year 0 pre‐
1390 ceding year 1 and year -1 preceding year 0. The time is
1391 assumed to be in the local timezone. If the daylight saving
1392 flag is positive, the time is assumed to be daylight saving
1393 time; if zero, the time is assumed to be standard time; and
1394 if negative (the default), mktime() attempts to determine
1395 whether daylight saving time is in effect for the specified
1396 time. If datespec does not contain enough elements or if the
1397 resulting time is out of range, mktime() returns -1.
1398 strftime([format [, timestamp[, utc-flag]]])
1400 Format timestamp according to the specification in format.
1401 If utc-flag is present and is non-zero or non-null, the
1402 result is in UTC, otherwise the result is in local time. The
1403 timestamp should be of the same form as returned by sys‐
1404 time(). If timestamp is missing, the current time of day is
1405 used. If format is missing, a default format equivalent to
1406 the output of date(1) is used. The default format is avail‐
1407 able in PROCINFO["strftime"]. See the specification for the
1408 strftime() function in ANSI C for the format conversions that
1409 are guaranteed to be available.
1410 systime() Return the current time of day as the number of seconds since
1412 the Epoch (1970-01-01 00:00:00 UTC on POSIX systems).
1413 Bit Manipulations Functions
1415 Gawk supplies the following bit manipulation functions. They work by
1416 converting double-precision floating point values to uintmax_t inte‐
1417 gers, doing the operation, and then converting the result back to
1418 floating point. The functions are:
1419 and(v1, v2) Return the bitwise AND of the values provided by v1
1421 and v2.
1422 compl(val) Return the bitwise complement of val.
1424 lshift(val, count) Return the value of val, shifted left by count
1426 bits.
1427 or(v1, v2) Return the bitwise OR of the values provided by v1
1429 and v2.
1430 rshift(val, count) Return the value of val, shifted right by count
1432 bits.
1433 xor(v1, v2) Return the bitwise XOR of the values provided by v1
1435 and v2.
1436 Type Function
1438 The following function is for use with multidimensional arrays.
1439 isarray(x)
1441 Return true if x is an array, false otherwise.
1442 Internationalization Functions
1444 The following functions may be used from within your AWK program for
1445 translating strings at run-time. For full details, see GAWK: Effective
1446 AWK Programming.
1447 bindtextdomain(directory [, domain])
1449 Specify the directory where gawk looks for the .mo files, in
1450 case they will not or cannot be placed in the ``standard'' loca‐
1451 tions (e.g., during testing). It returns the directory where
1452 domain is ``bound.''
1453 The default domain is the value of TEXTDOMAIN. If directory is
1454 the null string (""), then bindtextdomain() returns the current
1455 binding for the given domain.
1456 dcgettext(string [, domain [, category]])
1458 Return the translation of string in text domain domain for
1459 locale category category. The default value for domain is the
1460 current value of TEXTDOMAIN. The default value for category is
1461 "LC_MESSAGES".
1462 If you supply a value for category, it must be a string equal to
1463 one of the known locale categories described in GAWK: Effective
1464 AWK Programming. You must also supply a text domain. Use
1465 TEXTDOMAIN if you want to use the current domain.
1466 dcngettext(string1 , string2 , number [, domain [, category]])
1468 Return the plural form used for number of the translation of
1469 string1 and string2 in text domain domain for locale category
1470 category. The default value for domain is the current value of
1471 TEXTDOMAIN. The default value for category is "LC_MESSAGES".
1472 If you supply a value for category, it must be a string equal to
1473 one of the known locale categories described in GAWK: Effective
1474 AWK Programming. You must also supply a text domain. Use
1475 TEXTDOMAIN if you want to use the current domain.
1476
1478 Functions in AWK are defined as follows:
1479 function name(parameter list) { statements }
1481 Functions are executed when they are called from within expressions in
1483 either patterns or actions. Actual parameters supplied in the function
1484 call are used to instantiate the formal parameters declared in the
1485 function. Arrays are passed by reference, other variables are passed
1486 by value.
1487 Since functions were not originally part of the AWK language, the pro‐
1489 vision for local variables is rather clumsy: They are declared as extra
1490 parameters in the parameter list. The convention is to separate local
1491 variables from real parameters by extra spaces in the parameter list.
1492 For example:
1493 function f(p, q, a, b) # a and b are local
1495 {
1496 ...
1497 }
1498 /abc/ { ... ; f(1, 2) ; ... }
1500 The left parenthesis in a function call is required to immediately fol‐
1502 low the function name, without any intervening whitespace. This avoids
1503 a syntactic ambiguity with the concatenation operator. This restric‐
1504 tion does not apply to the built-in functions listed above.
1505 Functions may call each other and may be recursive. Function parame‐
1507 ters used as local variables are initialized to the null string and the
1508 number zero upon function invocation.
1509 Use return expr to return a value from a function. The return value is
1511 undefined if no value is provided, or if the function returns by “fall‐
1512 ing off” the end.
1513 As a gawk extension, functions may be called indirectly. To do this,
1515 assign the name of the function to be called, as a string, to a vari‐
1516 able. Then use the variable as if it were the name of a function, pre‐
1517 fixed with an @ sign, like so:
1518 function myfunc()
1519 {
1520 print "myfunc called"
1521 ...
1522 }
1523 { ...
1525 the_func = "myfunc"
1526 @the_func() # call through the_func to myfunc
1527 ...
1528 }
1529 If --lint has been provided, gawk warns about calls to undefined func‐
1531 tions at parse time, instead of at run time. Calling an undefined
1532 function at run time is a fatal error.
1533 The word func may be used in place of function.
1535
1537 You can dynamically add new built-in functions to the running gawk
1538 interpreter. The full details are beyond the scope of this manual
1539 page; see GAWK: Effective AWK Programming for the details.
1540 extension(object, function)
1542 Dynamically link the shared object file named by object, and
1543 invoke function in that object, to perform initialization.
1544 These should both be provided as strings. Return the value
1545 returned by function.
1546 Using this feature at the C level is not pretty, but it is unlikely to
1548 go away. Additional mechanisms may be added at some point.
1549
1551 pgawk accepts two signals. SIGUSR1 causes it to dump a profile and
1552 function call stack to the profile file, which is either awkprof.out,
1553 or whatever file was named with the --profile option. It then contin‐
1554 ues to run. SIGHUP causes pgawk to dump the profile and function call
1555 stack and then exit.
1556
1558 String constants are sequences of characters enclosed in double quotes.
1559 In non-English speaking environments, it is possible to mark strings in
1560 the AWK program as requiring translation to the local natural language.
1561 Such strings are marked in the AWK program with a leading underscore
1562 (“_”). For example,
1563 gawk 'BEGIN { print "hello, world" }'
1565 always prints hello, world. But,
1567 gawk 'BEGIN { print _"hello, world" }'
1569 might print bonjour, monde in France.
1571 There are several steps involved in producing and running a localizable
1573 AWK program.
1574 1. Add a BEGIN action to assign a value to the TEXTDOMAIN variable to
1576 set the text domain to a name associated with your program:
1577 BEGIN { TEXTDOMAIN = "myprog" }
1579 This allows gawk to find the .mo file associated with your program.
1581 Without this step, gawk uses the messages text domain, which likely
1582 does not contain translations for your program.
1583 2. Mark all strings that should be translated with leading under‐
1585 scores.
1586 3. If necessary, use the dcgettext() and/or bindtextdomain() functions
1588 in your program, as appropriate.
1589 4. Run gawk --gen-pot -f myprog.awk > myprog.pot to generate a .po
1591 file for your program.
1592 5. Provide appropriate translations, and build and install the corre‐
1594 sponding .mo files.
1595 The internationalization features are described in full detail in GAWK:
1597 Effective AWK Programming.
1598
1600 A primary goal for gawk is compatibility with the POSIX standard, as
1601 well as with the latest version of UNIX awk. To this end, gawk incor‐
1602 porates the following user visible features which are not described in
1603 the AWK book, but are part of the Bell Laboratories version of awk, and
1604 are in the POSIX standard.
1605 The book indicates that command line variable assignment happens when
1607 awk would otherwise open the argument as a file, which is after the
1608 BEGIN block is executed. However, in earlier implementations, when
1609 such an assignment appeared before any file names, the assignment would
1610 happen before the BEGIN block was run. Applications came to depend on
1611 this “feature.” When awk was changed to match its documentation, the
1612 -v option for assigning variables before program execution was added to
1613 accommodate applications that depended upon the old behavior. (This
1614 feature was agreed upon by both the Bell Laboratories and the GNU
1615 developers.)
1616 When processing arguments, gawk uses the special option “--” to signal
1618 the end of arguments. In compatibility mode, it warns about but other‐
1619 wise ignores undefined options. In normal operation, such arguments
1620 are passed on to the AWK program for it to process.
1621 The AWK book does not define the return value of srand(). The POSIX
1623 standard has it return the seed it was using, to allow keeping track of
1624 random number sequences. Therefore srand() in gawk also returns its
1625 current seed.
1626 Other new features are: The use of multiple -f options (from MKS awk);
1628 the ENVIRON array; the \a, and \v escape sequences (done originally in
1629 gawk and fed back into the Bell Laboratories version); the tolower()
1630 and toupper() built-in functions (from the Bell Laboratories version);
1631 and the ANSI C conversion specifications in printf (done first in the
1632 Bell Laboratories version).
1633
1635 There is one feature of historical AWK implementations that gawk sup‐
1636 ports: It is possible to call the length() built-in function not only
1637 with no argument, but even without parentheses! Thus,
1638 a = length # Holy Algol 60, Batman!
1640 is the same as either of
1642 a = length()
1644 a = length($0)
1645 Using this feature is poor practice, and gawk issues a warning about
1647 its use if --lint is specified on the command line.
1648
1650 Gawk has a number of extensions to POSIX awk. They are described in
1651 this section. All the extensions described here can be disabled by
1652 invoking gawk with the --traditional or --posix options.
1653 The following features of gawk are not available in POSIX awk.
1655 · No path search is performed for files named via the -f option.
1657 Therefore the AWKPATH environment variable is not special.
1658 · There is no facility for doing file inclusion (gawk's @include mecha‐
1660 nism).
1661 · The \x escape sequence. (Disabled with --posix.)
1663 · The ability to continue lines after ? and :. (Disabled with
1665 --posix.)
1666 · Octal and hexadecimal constants in AWK programs.
1668 · The ARGIND, BINMODE, ERRNO, LINT, RT and TEXTDOMAIN variables are not
1670 special.
1671 · The IGNORECASE variable and its side-effects are not available.
1673 · The FIELDWIDTHS variable and fixed-width field splitting.
1675 · The FPAT variable and field splitting based on field values.
1677 · The PROCINFO array is not available.
1679 · The use of RS as a regular expression.
1681 · The special file names available for I/O redirection are not recog‐
1683 nized.
1684 · The |& operator for creating co-processes.
1686 · The BEGINFILE and ENDFILE special patterns are not available.
1688 · The ability to split out individual characters using the null string
1690 as the value of FS, and as the third argument to split().
1691 · An optional fourth argument to split() to receive the separator
1693 texts.
1694 · The optional second argument to the close() function.
1696 · The optional third argument to the match() function.
1698 · The ability to use positional specifiers with printf and sprintf().
1700 · The ability to pass an array to length().
1702 · The use of delete array to delete the entire contents of an array.
1704 · The use of nextfile to abandon processing of the current input file.
1706 · The and(), asort(), asorti(), bindtextdomain(), compl(), dcgettext(),
1708 dcngettext(), gensub(), lshift(), mktime(), or(), patsplit(),
1709 rshift(), strftime(), strtonum(), systime() and xor() functions.
1710 · Localizable strings.
1712 · Adding new built-in functions dynamically with the extension() func‐
1714 tion.
1715 The AWK book does not define the return value of the close() function.
1717 Gawk's close() returns the value from fclose(3), or pclose(3), when
1718 closing an output file or pipe, respectively. It returns the process's
1719 exit status when closing an input pipe. The return value is -1 if the
1720 named file, pipe or co-process was not opened with a redirection.
1721 When gawk is invoked with the --traditional option, if the fs argument
1723 to the -F option is “t”, then FS is set to the tab character. Note
1724 that typing gawk -F\t ... simply causes the shell to quote the “t,”
1725 and does not pass “\t” to the -F option. Since this is a rather ugly
1726 special case, it is not the default behavior. This behavior also does
1727 not occur if --posix has been specified. To really get a tab character
1728 as the field separator, it is best to use single quotes: gawk -F'\t'
1729 ....
1730
1732 The AWKPATH environment variable can be used to provide a list of
1733 directories that gawk searches when looking for files named via the -f
1734 and --file options.
1735 For socket communication, two special environment variables can be used
1737 to control the number of retries (GAWK_SOCK_RETRIES), and the interval
1738 between retries (GAWK_MSEC_SLEEP). The interval is in milliseconds. On
1739 systems that do not support usleep(3), the value is rounded up to an
1740 integral number of seconds.
1741 If POSIXLY_CORRECT exists in the environment, then gawk behaves exactly
1743 as if --posix had been specified on the command line. If --lint has
1744 been specified, gawk issues a warning message to this effect.
1745
1747 If the exit statement is used with a value, then gawk exits with the
1748 numeric value given to it.
1749 Otherwise, if there were no problems during execution, gawk exits with
1751 the value of the C constant EXIT_SUCCESS. This is usually zero.
1752 If an error occurs, gawk exits with the value of the C constant
1754 EXIT_FAILURE. This is usually one.
1755 If gawk exits because of a fatal error, the exit status is 2. On non-
1757 POSIX systems, this value may be mapped to EXIT_FAILURE.
1758
1760 This man page documents gawk, version 4.0.
1761
1763 The original version of UNIX awk was designed and implemented by Alfred
1764 Aho, Peter Weinberger, and Brian Kernighan of Bell Laboratories. Brian
1765 Kernighan continues to maintain and enhance it.
1766 Paul Rubin and Jay Fenlason, of the Free Software Foundation, wrote
1768 gawk, to be compatible with the original version of awk distributed in
1769 Seventh Edition UNIX. John Woods contributed a number of bug fixes.
1770 David Trueman, with contributions from Arnold Robbins, made gawk com‐
1771 patible with the new version of UNIX awk. Arnold Robbins is the cur‐
1772 rent maintainer.
1773 The initial DOS port was done by Conrad Kwok and Scott Garfinkle.
1775 Scott Deifik maintains the port to MS-DOS using DJGPP. Eli Zaretskii
1776 maintains the port to MS-Windows using MinGW. Pat Rankin did the port
1777 to VMS, and Michal Jaegermann did the port to the Atari ST. The port
1778 to OS/2 was done by Kai Uwe Rommel, with contributions and help from
1779 Darrel Hankerson. Andreas Buening now maintains the OS/2 port. The
1780 late Fred Fish supplied support for the Amiga, and Martin Brown pro‐
1781 vided the BeOS port. Stephen Davies provided the original Tandem port,
1782 and Matthew Woehlke provided changes for Tandem's POSIX-compliant sys‐
1783 tems. Dave Pitts provided the port to z/OS.
1784 See the README file in the gawk distribution for up-to-date information
1786 about maintainers and which ports are currently supported.
1787
1789 If you find a bug in gawk, please send electronic mail to bug-
1790 gawk@gnu.org. Please include your operating system and its revision,
1791 the version of gawk (from gawk --version), which C compiler you used to
1792 compile it, and a test program and data that are as small as possible
1793 for reproducing the problem.
1794 Before sending a bug report, please do the following things. First,
1796 verify that you have the latest version of gawk. Many bugs (usually
1797 subtle ones) are fixed at each release, and if yours is out of date,
1798 the problem may already have been solved. Second, please see if set‐
1799 ting the environment variable LC_ALL to LC_ALL=C causes things to
1800 behave as you expect. If so, it's a locale issue, and may or may not
1801 really be a bug. Finally, please read this man page and the reference
1802 manual carefully to be sure that what you think is a bug really is,
1803 instead of just a quirk in the language.
1804 Whatever you do, do NOT post a bug report in comp.lang.awk. While the
1806 gawk developers occasionally read this newsgroup, posting bug reports
1807 there is an unreliable way to report bugs. Instead, please use the
1808 electronic mail addresses given above.
1809 If you're using a GNU/Linux or BSD-based system, you may wish to submit
1811 a bug report to the vendor of your distribution. That's fine, but
1812 please send a copy to the official email address as well, since there's
1813 no guarantee that the bug report will be forwarded to the gawk main‐
1814 tainer.
1815
1817 The -F option is not necessary given the command line variable assign‐
1818 ment feature; it remains only for backwards compatibility.
1819 Syntactically invalid single character programs tend to overflow the
1821 parse stack, generating a rather unhelpful message. Such programs are
1822 surprisingly difficult to diagnose in the completely general case, and
1823 the effort to do so really is not worth it.
1824
1826 egrep(1), getpid(2), getppid(2), getpgrp(2), getuid(2), geteuid(2),
1827 getgid(2), getegid(2), getgroups(2), usleep(3)
1828 The AWK Programming Language, Alfred V. Aho, Brian W. Kernighan, Peter
1830 J. Weinberger, Addison-Wesley, 1988. ISBN 0-201-07981-X.
1831 GAWK: Effective AWK Programming, Edition 4.0, shipped with the gawk
1833 source. The current version of this document is available online at
1834 http://www.gnu.org/software/gawk/manual.
1835
1837 Print and sort the login names of all users:
1838 BEGIN { FS = ":" }
1840 { print $1 | "sort" }
1841 Count lines in a file:
1843 { nlines++ }
1845 END { print nlines }
1846 Precede each line by its number in the file:
1848 { print FNR, $0 }
1850 Concatenate and line number (a variation on a theme):
1852 { print NR, $0 }
1854 Run an external command for particular lines of data:
1856 tail -f access_log |
1858 awk '/myhome.html/ { system("nmap " $1 ">> logdir/myhome.html") }'
1859
1861 Brian Kernighan of Bell Laboratories provided valuable assistance dur‐
1862 ing testing and debugging. We thank him.
1863
1865 Copyright © 1989, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
1866 2001, 2002, 2003, 2004, 2005, 2007, 2009, 2010, 2011 Free Software
1867 Foundation, Inc.
1868 Permission is granted to make and distribute verbatim copies of this
1870 manual page provided the copyright notice and this permission notice
1871 are preserved on all copies.
1872 Permission is granted to copy and distribute modified versions of this
1874 manual page under the conditions for verbatim copying, provided that
1875 the entire resulting derived work is distributed under the terms of a
1876 permission notice identical to this one.
1877 Permission is granted to copy and distribute translations of this man‐
1879 ual page into another language, under the above conditions for modified
1880 versions, except that this permission notice may be stated in a trans‐
1881 lation approved by the Foundation.
1882Free Software Foundation Dec 07 2012 GAWK(1)