Awk -- A Pattern Scanning and Processing Language (Second Edition) Alfred V. Aho Brian W. Kernighan Peter J. Weinberger Bell Laboratories Murray Hill, New Jersey 07974 ABSTRACT Awk is a programming language whose basic operation is to search a set of files for pat- terns, and to perform specified actions upon lines or fields of lines which contain instances of those patterns. Awk makes certain data selection and transformation operations easy to express; for example, the awk program length > 72 prints all input lines whose length exceeds 72 characters; the program NF % 2 == 0 prints all lines with an even number of fields; and the program { $1 = log($1); print } replaces the first field of each line by its loga- rithm. Awk patterns may include arbitrary boolean combinations of regular expressions and of rela- tional operators on strings, numbers, fields, variables, and array elements. Actions may include the same pattern-matching constructions as in patterns, as well as arithmetic and string expressions and assignments, if-else, while, for statements, and multiple output streams. This report contains a user's guide, a dis- cussion of the design and implementation of awk, and some timing statistics. September 1, 1978 - 2 - Awk -- A Pattern Scanning and Processing Language (Second Edition) Alfred V. Aho Brian W. Kernighan Peter J. Weinberger Bell Laboratories Murray Hill, New Jersey 07974 1. Introduction Awk is a programming language designed to make many common information retrieval and text manipulation tasks easy to state and to perform. The basic operation of awk is to scan a set of input lines in order, searching for lines which match any of a set of patterns which the user has specified. For each pattern, an action can be specified; this action will be performed on each line that matches the pattern. program grep unix program manual will recognize the approach, al- though Readers familiar with the UNIX* ---------------------------------------------------- - 2 - in awk the patterns may be more general than in grep, and the actions allowed are more involved than merely printing the matching line. For example, the awk pro- gram {print $3, $2} prints the third and second columns of a table in that order. The program $2 ~ /A|B|C/ prints all input lines with an A, B, or C in the second field. The program $1 != prev { print; prev = $1 } prints all lines in which the first field is different from the previous first field. 1.1. Usage The command awk program [files] executes the awk commands in the string program on the set of named files, or on the standard input if there are no files. The statements can also be placed in a file pfile, and executed by the command awk -f pfile [files] ---------------------------------------------------- *UNIX is a Trademark of Bell Laboratories. - 3 - 1.2. Program Structure An awk program is a sequence of statements of the form: pattern { action } pattern { action } ... Each line of input is matched against each of the patterns in turn. For each pattern that matches, the associated ac- tion is executed. When all the patterns have been tested, the next line is fetched and the matching starts over. Either the pattern or the action may be left out, but not both. If there is no action for a pattern, the matching line is simply copied to the output. (Thus a line which matches several patterns can be printed several times.) If there is no pattern for an action, then the action is per- formed for every input line. A line which matches no pat- tern is ignored. Since patterns and actions are both optional, actions must be enclosed in braces to distinguish them from pat- terns. 1.3. Records and Fields Awk input is divided into ``records'' terminated by a record separator. The default record separator is a new- line, so by default awk processes its input a line at a time. The number of the current record is available in a - 4 - variable named NR. Each input record is considered to be divided into ``fields.'' Fields are normally separated by white space -- blanks or tabs -- but the input field separator may be changed, as described below. Fields are referred to as $1, $2, and so forth, where $1 is the first field, and $0 is the whole input record itself. Fields may be assigned to. The number of fields in the current record is available in a variable named NF. The variables FS and RS refer to the input field and record separators; they may be changed at any time to any single character. The optional command-line argument -Fc may also be used to set FS to the character c. If the record separator is empty, an empty input line is taken as the record separator, and blanks, tabs and new- lines are treated as field separators. The variable FILENAME contains the name of the current input file. 1.4. Printing An action may have no pattern, in which case the action is executed for all lines. The simplest action is to print some or all of a record; this is accomplished by the awk command print. The awk program { print } - 5 - prints each record, thus copying the input to the output in- tact. More useful is to print a field or fields from each record. For instance, print $2, $1 prints the first two fields in reverse order. Items sepa- rated by a comma in the print statement will be separated by the current output field separator when output. Items not separated by commas will be concatenated, so print $1 $2 runs the first and second fields together. The predefined variables NF and NR can be used; for ex- ample { print NR, NF, $0 } prints each record preceded by the record number and the number of fields. Output may be diverted to multiple files; the program { print $1 >"foo1"; print $2 >"foo2" } writes the first field, $1, on the file foo1, and the second field on file foo2. The >> notation can also be used: print $1 >>"foo" appends the output to the file foo. (In each case, the out- - 6 - put files are created if necessary.) The file name can be a variable or a field as well as a constant; for example, print $1 >$2 uses the contents of field 2 as a file name. Naturally there is a limit on the number of output files; currently it is 10. Similarly, output can be piped into another process (on UNIX only); for instance, print | "mail bwk" mails the output to bwk. The variables OFS and ORS may be used to change the current output field separator and output record separator. The output record separator is appended to the output of the print statement. Awk also provides the printf statement for output for- matting: printf format expr, expr, ... formats the expressions in the list according to the speci- fication in format and prints them. For example, printf "%8.2f %10ld\n", $1, $2 prints $1 as a floating point number 8 digits wide, with two - 7 - after the decimal point, and $2 as a 10-digit long decimal number, followed by a newline. No output separators are produced automatically; you must add them yourself, as in this example. The version of printf is identical to that used with C. C programm language prentice hall 1978 2. Patterns A pattern in front of an action acts as a selector that determines whether the action is to be executed. A variety of expressions may be used as patterns: regular expressions, arithmetic relational expressions, string-valued expres- sions, and arbitrary boolean combinations of these. 2.1. BEGIN and END The special pattern BEGIN matches the beginning of the input, before the first record is read. The pattern END matches the end of the input, after the last record has been processed. BEGIN and END thus provide a way to gain control before and after processing, for initialization and wrapup. As an example, the field separator can be set to a colon by BEGIN { FS = ":" } ... rest of program ... Or the input lines may be counted by END { print NR } - 8 - If BEGIN is present, it must be the first pattern; END must be the last if used. 2.2. Regular Expressions The simplest regular expression is a literal string of characters enclosed in slashes, like /smith/ This is actually a complete awk program which will print all lines which contain any occurrence of the name ``smith''. If a line contains ``smith'' as part of a larger word, it will also be printed, as in blacksmithing Awk regular expressions include the regular expression forms found in the UNIX text editor ed unix program manual and grep (without back-referencing). In addition, awk al- lows parentheses for grouping, | for alternatives, + for ``one or more'', and ? for ``zero or one'', all as in lex. Character classes may be abbreviated: [a-zA-Z0-9] is the set of all letters and digits. As an example, the awk program /[Aa]ho|[Ww]einberger|[Kk]ernighan/ will print all lines which contain any of the names ``Aho,'' ``Weinberger'' or ``Kernighan,'' whether capitalized or not. Regular expressions (with the extensions listed above) - 9 - must be enclosed in slashes, just as in ed and sed. Within a regular expression, blanks and the regular expression metacharacters are significant. To turn of the magic mean- ing of one of the regular expression characters, precede it with a backslash. An example is the pattern /\/.*\// which matches any string of characters enclosed in slashes. One can also specify that any field or variable matches a regular expression (or does not match it) with the opera- tors ~ and !~. The program $1 ~ /[jJ]ohn/ prints all lines where the first field matches ``john'' or ``John.'' Notice that this will also match ``Johnson'', ``St. Johnsbury'', and so on. To restrict it to exactly [jJ]ohn, use $1 ~ /^[jJ]ohn$/ The caret ^ refers to the beginning of a line or field; the dollar sign $ refers to the end. 2.3. Relational Expressions An awk pattern can be a relational expression involving the usual relational operators <, <=, ==, !=, >=, and >. An example is - 10 - $2 > $1 + 100 which selects lines where the second field is at least 100 greater than the first field. Similarly, NF % 2 == 0 prints lines with an even number of fields. In relational tests, if neither operand is numeric, a string comparison is made; otherwise it is numeric. Thus, $1 >= "s" selects lines that begin with an s, t, u, etc. In the ab- sence of any other information, fields are treated as strings, so the program $1 > $2 will perform a string comparison. 2.4. Combinations of Patterns A pattern can be any boolean combination of patterns, using the operators || (or), && (and), and ! (not). For ex- ample, $1 >= "s" && $1 < "t" && $1 != "smith" selects lines where the first field begins with ``s'', but is not ``smith''. && and || guarantee that their operands will be evaluated from left to right; evaluation stops as - 11 - soon as the truth or falsehood is determined. 2.5. Pattern Ranges The ``pattern'' that selects an action may also consist of two patterns separated by a comma, as in pat1, pat2 { ... } In this case, the action is performed for each line between an occurrence of pat1 and the next occurrence of pat2 (in- clusive). For example, /start/, /stop/ prints all lines between start and stop, while NR == 100, NR == 200 { ... } does the action for lines 100 through 200 of the input. 3. Actions An awk action is a sequence of action statements termi- nated by newlines or semicolons. These action statements can be used to do a variety of bookkeeping and string manip- ulating tasks. 3.1. Built-in Functions Awk provides a ``length'' function to compute the length of a string of characters. This program prints each record, preceded by its length: - 12 - {print length, $0} length by itself is a ``pseudo-variable'' which yields the length of the current record; length(argument) is a function which yields the length of its argument, as in the equiva- lent {print length($0), $0} The argument may be any expression. Awk also provides the arithmetic functions sqrt, log, exp, and int, for square root, base e logarithm, exponen- tial, and integer part of their respective arguments. The name of one of these built-in functions, without argument or parentheses, stands for the value of the func- tion on the whole record. The program length < 10 || length > 20 prints lines whose length is less than 10 or greater than 20. The function substr(s, m, n) produces the substring of s that begins at position m (origin 1) and is at most n characters long. If n is omitted, the substring goes to the end of s. The function index(s1, s2) returns the position where the string s2 occurs in s1, or zero if it does not. The function sprintf(f, e1, e2, ...) produces the value of the expressions e1, e2, etc., in the printf format speci- - 13 - fied by f. Thus, for example, x = sprintf("%8.2f %10ld", $1, $2) sets x to the string produced by formatting the values of $1 and $2. 3.2. Variables, Expressions, and Assignments Awk variables take on numeric (floating point) or string values according to context. For example, in x = 1 x is clearly a number, while in x = "smith" it is clearly a string. Strings are converted to numbers and vice versa whenever context demands it. For instance, x = "3" + "4" assigns 7 to x. Strings which cannot be interpreted as num- bers in a numerical context will generally have numeric val- ue zero, but it is unwise to count on this behavior. By default, variables (other than built-ins) are ini- tialized to the null string, which has numerical value zero; this eliminates the need for most BEGIN sections. For exam- ple, the sums of the first two fields can be computed by { s1 += $1; s2 += $2 } - 14 - END { print s1, s2 } Arithmetic is done internally in floating point. The arithmetic operators are +, -, *, /, and % (mod). The C in- crement ++ and decrement -- operators are also available, and so are the assignment operators +=, -=, *=, /=, and %=. These operators may all be used in expressions. 3.3. Field Variables Fields in awk share essentially all of the properties of variables -- they may be used in arithmetic or string op- erations, and may be assigned to. Thus one can replace the first field with a sequence number like this: { $1 = NR; print } or accumulate two fields into a third, like this: { $1 = $2 + $3; print $0 } or assign a string to a field: { if ($3 > 1000) $3 = "too big" print } which replaces the third field by ``too big'' when it is, and in any case prints the record. Field references may be numerical expressions, as in - 15 - { print $i, $(i+1), $(i+n) } Whether a field is deemed numeric or string depends on con- text; in ambiguous cases like if ($1 == $2) ... fields are treated as strings. Each input line is split into fields automatically as necessary. It is also possible to split any variable or string into fields: n = split(s, array, sep) splits the the string s into array[1], ..., array[n]. The number of elements found is returned. If the sep argument is provided, it is used as the field separator; otherwise FS is used as the separator. 3.4. String Concatenation Strings may be concatenated. For example length($1 $2 $3) returns the length of the first three fields. Or in a print statement, print $1 " is " $2 prints the two fields separated by `` is ''. Variables and numeric expressions may also appear in concatenations. - 16 - 3.5. Arrays Array elements are not declared; they spring into exis- tence by being mentioned. Subscripts may have any non-null value, including non-numeric strings. As an example of a conventional numeric subscript, the statement x[NR] = $0 assigns the current input record to the NR-th element of the array x. In fact, it is possible in principle (though per- haps slow) to process the entire input in a random order with the awk program { x[NR] = $0 } END { ... program ... } The first action merely records each input line in the array x. Array elements may be named by non-numeric values, which gives awk a capability rather like the associative memory of Snobol tables. Suppose the input contains fields with values like apple, orange, etc. Then the program /apple/ { x["apple"]++ } /orange/ { x["orange"]++ } END { print x["apple"], x["orange"] } increments counts for the named array elements, and prints them at the end of the input. - 17 - 3.6. Flow-of-Control Statements Awk provides the basic flow-of-control statements if- else, while, for, and statement grouping with braces, as in C. We showed the if statement in section 3.3 without de- scribing it. The condition in parentheses is evaluated; if it is true, the statement following the if is done. The else part is optional. The while statement is exactly like that of C. For ex- ample, to print all input fields one per line, i = 1 while (i <= NF) { print $i ++i } The for statement is also exactly that of C: for (i = 1; i <= NF; i++) print $i does the same job as the while statement above. There is an alternate form of the for statement which is suited for accessing the elements of an associative ar- ray: for (i in array) statement - 18 - does statement with i set in turn to each element of array. The elements are accessed in an apparently random order. Chaos will ensue if i is altered, or if any new elements are accessed during the loop. The expression in the condition part of an if, while or for can include relational operators like <, <=, >, >=, == (``is equal to''), and != (``not equal to''); regular ex- pression matches with the match operators ~ and !~; the log- ical operators ||, &&, and !; and of course parentheses for grouping. The break statement causes an immediate exit from an enclosing while or for; the continue statement causes the next iteration to begin. The statement next causes awk to skip immediately to the next record and begin scanning the patterns from the top. The statement exit causes the program to behave as if the end of the input had occurred. Comments may be placed in awk programs: they begin with the character # and end with the end of the line, as in print x, y # this is a comment 4. Design The UNIX system already provides several programs that operate by passing input through a selection mechanism. - 19 - Grep, the first and simplest, merely prints all lines which match a single specified pattern. Egrep provides more gen- eral patterns, i.e., regular expressions in full generality; fgrep searches for a set of keywords with a particularly fast algorithm. Sed unix programm manual provides most of the editing facilities of the editor ed, applied to a stream of input. None of these programs provides numeric capabili- ties, logical relations, or variables. Lex lesk lexical analyzer cstr provides general regular expression recognition capabilities, and, by serving as a C program generator, is essentially open-ended in its capabil- ities. The use of lex, however, requires a knowledge of C programming, and a lex program must be compiled and loaded before use, which discourages its use for one-shot applica- tions. Awk is an attempt to fill in another part of the matrix of possibilities. It provides general regular expression capabilities and an implicit input/output loop. But it also provides convenient numeric processing, variables, more gen- eral selection, and control flow in the actions. It does not require compilation or a knowledge of C. Finally, awk provides a convenient way to access fields within lines; it is unique in this respect. Awk also tries to integrate strings and numbers com- pletely, by treating all quantities as both string and nu- meric, deciding which representation is appropriate as late - 20 - as possible. In most cases the user can simply ignore the differences. Most of the effort in developing awk went into deciding what awk should or should not do (for instance, it doesn't do string substitution) and what the syntax should be (no explicit operator for concatenation) rather than on writing or debugging the code. We have tried to make the syntax powerful but easy to use and well adapted to scanning files. For example, the absence of declarations and implicit ini- tializations, while probably a bad idea for a general-pur- pose programming language, is desirable in a language that is meant to be used for tiny programs that may even be com- posed on the command line. In practice, awk usage seems to fall into two broad categories. One is what might be called ``report genera- tion'' -- processing an input to extract counts, sums, sub- totals, etc. This also includes the writing of trivial data validation programs, such as verifying that a field contains only numeric information or that certain delimiters are properly balanced. The combination of textual and numeric processing is invaluable here. A second area of use is as a data transformer, convert- ing data from the form produced by one program into that ex- pected by another. The simplest examples merely select fields, perhaps with rearrangements. - 21 - 5. Implementation The actual implementation of awk uses the language de- velopment tools available on the UNIX operating system. The grammar is specified with yacc; yacc johnson cstr the lexi- cal analysis is done by lex; the regular expression recog- nizers are deterministic finite automata constructed direct- ly from the expressions. An awk program is translated into a parse tree which is then directly executed by a simple in- terpreter. Awk was designed for ease of use rather than processing speed; the delayed evaluation of variable types and the ne- cessity to break input into fields makes high speed diffi- cult to achieve in any case. Nonetheless, the program has not proven to be unworkably slow. Table I below shows the execution (user + system) time on a PDP-11/70 of the UNIX programs wc, grep, egrep, fgrep, sed, lex, and awk on the following simple tasks: 1. count the number of lines. 2. print all lines containing ``doug''. 3. print all lines containing ``doug'', ``ken'' or ``dmr''. 4. print the third field of each line. - 22 - 5. print the third and second fields of each line, in that order. 6. append all lines containing ``doug'', ``ken'', and ``dmr'' to files ``jdoug'', ``jken'', and ``jdmr'', re- spectively. 7. print each line prefixed by ``line-number : ''. 8. sum the fourth column of a table. The program wc merely counts words, lines and characters in its input; we have already mentioned the others. In all cases the input was a file containing 10,000 lines as creat- ed by the command ls -l; each line has the form -rw-rw-rw- 1 ava 123 Oct 15 17:05 xxx The total length of this input is 452,960 characters. Times for lex do not include compile or load. As might be expected, awk is not as fast as the spe- cialized tools wc, sed, or the programs in the grep family, but is faster than the more general tool lex. In all cases, the tasks were about as easy to express as awk programs as programs in these other languages; tasks involving fields were considerably easier to express as awk programs. Some of the test programs are shown in awk, sed and lex. $LIST$ Task Program 1 2 3 4 5 6 7 8 -------------------------------------------------------------------- - 23 - | | | | | | | | | wc | 8.6 | | | | | | | | | | | | | | | | | grep | 11.7 | 13.1 | | | | | | | | | | | | | | | | egrep | 6.2 | 11.5 | 11.6 | | | | | | | | | | | | | | | fgrep | 7.7 | 13.8 | 16.1 | | | | | | | | | | | | | | | sed | 10.2 | 11.6 | 15.8 | 29.0 | 30.5 | 16.1 | | | | | | | | | | | | lex | 65.1 | 150.1 | 144.2 | 67.7 | 70.3 | 104.0 | 81.7 | 92.8 | | | | | | | | | | awk | 15.0 | 25.6 | 29.9 | 33.3 | 38.9 | 46.4 | 71.4 | 31.1 | | | | | | | | | | --------+------+-------+-------+------+------+-------+------+------+ Table I. Execution Times of Programs. (Times are in sec.) The programs for some 6. /ken/ {print >"jken"} of these jobs are shown /doug/ {print >"jdoug"} below. The lex programs are /dmr/ {print >"jdmr"} generally too long to show. 7. {print NR ": " $0} AWK: 8. {sum = sum + $4} 1. END {print NR} END {print sum} 2. /doug/ SED: 3. /ken|doug|dmr/ 1. $= 4. {print $3} 2. /doug/p 5. {print $3, $2} 3. /doug/p /doug/d - 24 - /ken/p ^.*doug.*$ printf("%s\n", yytext); /ken/d . ; /dmr/p \n ; /dmr/d 4. /[^ ]* [ ]*[^ ]* [ ]*\([^ ]*\) .*/s//\1/p 5. /[^ ]* [ ]*\([^ ]*\) [ ]*\([^ ]*\) .*/s//\2 \1/p 6. /ken/w jken /doug/w jdoug /dmr/w jdmr LEX: 1. %{ int i; %} %% \n i++; . ; %% yywrap() { printf("%d\n", i); } 2. %%