![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to awk.1 CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [Research Unix] / researchv10dc / man / adm / man1|
Return to awk.1 CVS log | Up to [Research Unix] / researchv10dc / man / adm / man1 |
1.1 root 1: .TH AWK 1
2: .CT 1 files prog_other
3: .SH NAME
4: awk \- pattern-directed scanning and processing language
5: .SH SYNOPSIS
6: .B awk
7: [
8: .BI -F fs
9: ]
10: [
11: .BI -v
12: .I var=value
13: ]
14: [
15: .I prog
16: ]
17: [
18: .I file ...
19: ]
20: .SH DESCRIPTION
21: .I Awk
22: scans each input
23: .I file
24: for lines that match any of a set of patterns specified literally in
25: .IR prog
26: or in one or more files
27: specified as
28: .B -f
29: .IR file .
30: With each pattern
31: there can be an associated action that will be performed
32: when a line of a
33: .I file
34: matches the pattern.
35: Each line is matched against the
36: pattern portion of every pattern-action statement;
37: the associated action is performed for each matched pattern.
38: The file name
39: .L -
40: means the standard input.
41: Any
42: .IR file
43: of the form
44: .I var=value
45: is treated as an assignment, not a filename,
46: and is executed at the time it would have been opened if it were a filename.
47: The option
48: .B -v
49: followed by
50: .I var=value
51: is an assignment to be done before
52: .I prog
53: is executed;
54: any number of
55: .B -v
56: options may be present.
57: .PP
58: An input line is made up of fields separated by white space,
59: or by regular expression
60: .BR FS .
61: The fields are denoted
62: .BR $1 ,
63: .BR $2 ,
64: \&..., while
65: .B $0
66: refers to the entire line.
67: .PP
68: A pattern-action statement has the form
69: .IP
70: .IB pattern " { " action " }
71: .PP
72: A missing
73: .BI { " action " }
74: means print the line;
75: a missing pattern always matches.
76: Pattern-action statements are separated by newlines or semicolons.
77: .PP
78: An action is a sequence of statements.
79: A statement can be one of the following:
80: .PP
81: .EX
82: .ta \w'\f(CWdelete array[expression]'u
83: if(\fI expression \fP)\fI statement \fP\fR[ \fPelse\fI statement \fP\fR]\fP
84: while(\fI expression \fP)\fI statement\fP
85: for(\fI expression \fP;\fI expression \fP;\fI expression \fP)\fI statement\fP
86: for(\fI var \fPin\fI array \fP)\fI statement\fP
87: do\fI statement \fPwhile(\fI expression \fP)
88: break
89: continue
90: {\fR [\fP\fI statement ... \fP\fR] \fP}
91: \fIexpression\fP #\fR commonly\fP\fI var = expression\fP
92: print\fR [ \fP\fIexpression-list \fP\fR] \fP\fR[ \fP>\fI expression \fP\fR]\fP
93: printf\fI format \fP\fR[ \fP,\fI expression-list \fP\fR] \fP\fR[ \fP>\fI expression \fP\fR]\fP
94: return\fR [ \fP\fIexpression \fP\fR]\fP
95: next #\fR skip remaining patterns on this input line\fP
96: delete\fI array\fP[\fI expression \fP] #\fR delete an array element\fP
97: exit\fR [ \fP\fIexpression \fP\fR]\fP #\fR exit immediately; status is \fP\fIexpression\fP
98: .EE
99: .DT
100: .PP
101: Statements are terminated by
102: semicolons, newlines or right braces.
103: An empty
104: .I expression-list
105: stands for
106: .BR $0 .
107: String constants are quoted \&\f(CW"\ "\fR,
108: with the usual C escapes recognized within.
109: Expressions take on string or numeric values as appropriate,
110: and are built using the operators
111: .B + - * / % ^
112: (exponentiation), and concatenation (indicated by white space).
113: The operators
114: .B
115: ! ++ -- += -= *= /= %= ^= > >= < <= == != && || ?:
116: are also available in expressions.
117: Variables may be scalars, array elements
118: (denoted
119: .IB x [ i ] )
120: or fields.
121: Variables are initialized to the null string.
122: Array subscripts may be any string,
123: not necessarily numeric;
124: this allows for a form of associative memory.
125: Multiple subscripts such as
126: .B [i,j,k]
127: are permitted; the constituents are concatenated,
128: separated by the value of
129: .BR SUBSEP .
130: .PP
131: The
132: .B print
133: statement prints its arguments on the standard output
134: (or on a file if
135: .BI > file
136: or
137: .BI >> file
138: is present or on a pipe if
139: .BI | cmd
140: is present), separated by the current output field separator,
141: and terminated by the output record separator.
142: .I file
143: and
144: .I cmd
145: may be literal names or parenthesized expressions;
146: identical string values in different statements denote
147: the same open file.
148: The
149: .B printf
150: statement formats its expression list according to the format
151: (see
152: .IR printf (3)) .
153: The built-in function
154: .BI close( expr )
155: closes the file or pipe
156: .IR expr .
157: .PP
158: The mathematical functions
159: .BR exp ,
160: .BR log ,
161: .BR sqrt ,
162: .BR sin ,
163: .BR cos ,
164: and
165: .BR atan2
166: are built in.
167: Other built-in functions:
168: .TF length
169: .TP
170: .B length
171: the length of its argument
172: taken as a string,
173: or of
174: .B $0
175: if no argument.
176: .TP
177: .B rand
178: random number on (0,1)
179: .TP
180: .B srand
181: sets seed for
182: .B rand
183: and returns the previous seed.
184: .TP
185: .B int
186: truncates to an integer value
187: .TP
188: .BI substr( s , " m" , " n\fB)
189: the
190: .IR n -character
191: substring of
192: .I s
193: that begins at position
194: .IR m
195: counted from 1.
196: .TP
197: .BI index( s , " t" )
198: the position in
199: .I s
200: where the string
201: .I t
202: occurs, or 0 if it does not.
203: .TP
204: .BI match( s , " r" )
205: the position in
206: .I s
207: where the regular expression
208: .I r
209: occurs, or 0 if it does not.
210: The variables
211: .B RSTART
212: and
213: .B RLENGTH
214: are set to the position and length of the matched string.
215: .TP
216: .BI split( s , " a" , " fs\fB)
217: splits the string
218: .I s
219: into array elements
220: .IB a [1] ,
221: .IB a [2] ,
222: \&...,
223: .IB a [ n ] ,
224: and returns
225: .IR n .
226: The separation is done with the regular expression
227: .I fs
228: or with the field separator
229: .B FS
230: if
231: .I fs
232: is not given.
233: .TP
234: .BI sub( r , " t" , " s\fB)
235: substitutes
236: .I t
237: for the first occurrence of the regular expression
238: .I r
239: in the string
240: .IR s .
241: If
242: .I s
243: is not given,
244: .B $0
245: is used.
246: .TP
247: .B gsub
248: same as
249: .B sub
250: except that all occurrences of the regular expression
251: are replaced;
252: .B sub
253: and
254: .B gsub
255: return the number of replacements.
256: .TP
257: .BI sprintf( fmt , " expr" , " ...\fB )
258: the string resulting from formatting
259: .I expr ...
260: according to the
261: .IR printf (3)
262: format
263: .I fmt
264: .TP
265: .BI system( cmd )
266: executes
267: .I cmd
268: and returns its exit status
269: .PD
270: .PP
271: The ``function''
272: .B getline
273: sets
274: .B $0 to
275: the next input record from the current input file;
276: .B getline
277: .BI < file
278: sets
279: .B $0
280: to the next record from
281: .IR file .
282: .B getline
283: .I x
284: sets variable
285: .I x
286: instead.
287: Finally,
288: .IB cmd " | getline
289: pipes the output of
290: .I cmd
291: into
292: .BR getline ;
293: each call of
294: .B getline
295: returns the next line of output from
296: .IR cmd .
297: In all cases,
298: .B getline
299: returns 1 for a successful input,
300: 0 for end of file, and \-1 for an error.
301: .PP
302: Patterns are arbitrary Boolean combinations
303: (with
304: .BR "! || &&" )
305: of regular expressions and
306: relational expressions.
307: Regular expressions are as in
308: .IR egrep ;
309: see
310: .IR gre (1).
311: Isolated regular expressions
312: in a pattern apply to the entire line.
313: Regular expressions may also occur in
314: relational expressions, using the operators
315: .BR ~
316: and
317: .BR !~ .
318: .BI / re /
319: is a constant regular expression;
320: any string (constant or variable) may be used
321: as a regular expression, except in the position of an isolated regular expression
322: in a pattern.
323: .PP
324: A pattern may consist of two patterns separated by a comma;
325: in this case, the action is performed for all lines
326: from an occurrence of the first pattern
327: though an occurrence of the second.
328: .PP
329: A relational expression is one of the following:
330: .IP
331: .I expression matchop regular-expression
332: .br
333: .I expression relop expression
334: .br
335: .IB expression " in " array-name
336: .br
337: .BI ( expr , expr,... ") in " array-name
338: .PP
339: where a relop is any of the six relational operators in C,
340: and a matchop is either
341: .B ~
342: (matches)
343: or
344: .B !~
345: (does not match).
346: A conditional is an arithmetic expression,
347: a relational expression,
348: or a Boolean combination
349: of these.
350: .PP
351: The special patterns
352: .B BEGIN
353: and
354: .B END
355: may be used to capture control before the first input line is read
356: and after the last.
357: .B BEGIN
358: and
359: .B END
360: do not combine with other patterns.
361: .PP
362: Variable names with special meanings:
363: .TF FILENAME
364: .TP
365: .B FS
366: regular expression used to separate fields; also settable
367: by option
368: .BI -F fs.
369: .TP
370: .BR NF
371: number of fields in the current record
372: .TP
373: .B NR
374: ordinal number of the current record
375: .TP
376: .B FNR
377: ordinal number of the current record in the current file
378: .TP
379: .B FILENAME
380: the name of the current input file
381: .TP
382: .B RS
383: input record separator (default newline)
384: .TP
385: .B OFS
386: output field separator (default blank)
387: .TP
388: .B ORS
389: output record separator (default newline)
390: .TP
391: .B OFMT
392: output format for numbers (default
393: .BR "%.6g" )
394: .TP
395: .B SUBSEP
396: separates multiple subscripts (default 034)
397: .TP
398: .B ARGC
399: argument count, assignable
400: .TP
401: .B ARGV
402: argument array, assignable;
403: non-null members are taken as filenames
404: .TP
405: .B ENVIRON
406: array of environment variables; subscripts are names.
407: .PD
408: .PP
409: Functions may be defined (at the position of a pattern-action statement) thus:
410: .IP
411: .L
412: function foo(a, b, c) { ...; return x }
413: .PP
414: Parameters are passed by value if scalar and by reference if array name;
415: functions may be called recursively.
416: Parameters are local to the function; all other variables are global.
417: Thus local variables may be created by providing excess parameters in
418: the function definition.
419: .SH EXAMPLES
420: .TP
421: .L
422: length > 72
423: Print lines longer than 72 characters.
424: .TP
425: .L
426: { print $2, $1 }
427: Print first two fields in opposite order.
428: .PP
429: .EX
430: BEGIN { FS = ",[ \et]*|[ \et]+" }
431: { print $2, $1 }
432: .EE
433: .ns
434: .IP
435: Same, with input fields separated by comma and/or blanks and tabs.
436: .PP
437: .EX
438: { s += $1 }
439: END { print "sum is", s, " average is", s/NR }
440: .EE
441: .ns
442: .IP
443: Add up first column, print sum and average.
444: .TP
445: .L
446: /start/, /stop/
447: Print all lines between start/stop pairs.
448: .PP
449: .EX
450: BEGIN { # Simulate echo(1)
451: for (i = 1; i < ARGC; i++) printf "%s ", ARGV[i]
452: printf "\en"
453: exit }
454: .EE
455: .SH SEE ALSO
456: .IR gre (1),
457: .IR lex (1),
458: .IR sed (1)
459: .br
460: A. V. Aho, B. W. Kernighan, P. J. Weinberger,
461: .I
462: The AWK Programming Language,
463: Addison-Wesley, 1988.
464: .SH BUGS
465: There are no explicit conversions between numbers and strings.
466: To force an expression to be treated as a number add 0 to it;
467: to force it to be treated as a string concatenate
468: \&\f(CW""\fP to it.
469: .br
470: The scope rules for variables in functions are a botch;
471: the syntax is worse.
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.