ast@cs.vu.nl (09/26/89)
NAME
bawk - text processor
SYNOPSIS
bawk rules [file] ...
DESCRIPTION
Bawk is a text processing program that searches files for
specific patterns and performs "actions" for every occurrance
of these patterns. The patterns can be "regular expressions"
as used in the UNIX "ex" editor. The actions are expressed
using a subset of the "C" language.
The patterns and actions are usually placed in a "rules" file
whose name must be the first argument in the command line.
All other arguments are taken to be the names of text files on
which the rules are to be applied.
The special file name "-" may also be used anywhere on the
command line to take input from the standard input device.
The command:
bawk - prog.c - prog.h
would read the patterns and actions rules from the standard
input, then apply them to the files "prog.c", the standard
input and "prog.h" in that order.
The general format of a rules file is:
<pattern> { <action> }
<pattern> { <action> }
...
There may be any number of these <pattern> { <action> }
sequences in the rules file. Bawk reads a line of input from
the current input file and applies every <pattern> { <action> }
in sequence to the line.
If the <pattern> corresponding to any { <action> } is missing,
the action is applied to every line of input. The default
{ <action> } is to print the matched input line.
PATTERNS
The <pattern>'s may consist of any valid C expression. If the
<pattern> consists of two expressions seperated by a comma, it
is taken to be a range and the <action> is performed on all
lines of input that match the range. <pattern>'s may contain
"regular expressions" delimited by an '@' symbol. Regular
expressions can be thought of as a generalized "wildcard"
string matching mechanism, similar to that used by many
operating systems to specify file names. Regular expressions
may contain any of the following characters:
x An ordinary character (not mentioned below)
matches that character.
'\' The backslash quotes any character.
"\$" matches a dollar-sign.
'^' A circumflex at the beginning of an expression
matches the beginning of a line.
'$' A dollar-sign at the end of an expression
matches the end of a line.
'.' A period matches any single character except
newline.
':x' A colon matches a class of characters described
by the character following it:
':a' ":a" matches any alphabetic;
':d' ":d" matches digits;
':n' ":n" matches alphanumerics;
': ' ": " matches spaces, tabs, and other control
characters, such as newline.
'*' An expression followed by an asterisk matches
zero or more occurrances of that expression:
"fo*" matches "f", "fo", "foo", "fooo", etc.
'+' An expression followed by a plus sign matches
one or more occurrances of that expression:
"fo+" matches "fo", "foo", "fooo", etc.
'-' An expression followed by a minus sign
optionally matches the expression.
'[]' A string enclosed in square brackets matches
any single character in that string, but no
others. If the first character in the string
is a circumflex, the expression matches any
character except newline and the characters in
the string. For example, "[xyz]" matches "xx"
and "zyx", while "[^xyz]" matches "abc" but not
"axb". A range of characters may be specified
by two characters separated by "-". Note that,
[a-z] matches alphabetics, while [z-a] never
matches.
For example, the following rules file would print every line
that contained a valid C identifier:
@[a-zA-Z][a-zA-Z0-9]@
And this rules file would print all lines between and including
the ones that contained the word "START" and "END":
@START@, @END@
ACTIONS
Actions are expressed as a subset of the C language. All
variables are global and default to int's if not formally
declared. Variable declarations may appear anywhere within
an action. Only char's and int's and pointers and arrays of
char and int are allowed. Bawk allows only decimal integer
constants to be used - no hex (0xnn) or octal (0nn). String
and character constants may contain all of the special C
escapes (\n, \r, etc.).
Bawk supports the "if", "else", "while" and "break" flow of
control constructs, which behave exactly as in C.
Also supported are the following unary and binary operators,
listed in order from highest to lowest precedence:
operator type associativity
() [] unary left to right
! ~ ++ -- - * & unary right to left
* / % binary left to right
+ - binary left to right
<< >> binary left to right
< <= > >= binary left to right
== != binary left to right
& binary left to right
^ binary left to right
| binary left to right
&& binary left to right
|| binary left to right
= binary right to left
Comments are introduced by a '#' symbol and are terminated by
the first newline character. The standard "/*" and "*/"
comment delimiters are not supported and will result in a
syntax error.
FIELDS
When bawk reads a line from the current input file, the
record is automatically seperated into "fields". A field is
simply a string of consecutive characters delimited by either
the beginning or end of line, or a "field seperator" character
Initially, the field seperators are the space and tab character.
The special unary operator '$' is used to reference one of the
fields in the current input record (line). The fields are
numbered sequentially starting at 1. The expression "$0"
references the entire input line.
Similarly, the "record seperator" is used to determine the end
of an input "line", initially the newline character.
The field and record seperators may be changed programatically
by one of the actions and will remain in effect until changed
again.
Fields behave exactly like strings; and can be used in the same
context as a character array. These "arrays" can be considered
to have been declared as:
char ($n)[ 128 ];
In other words, they are 128 bytes long. Notice that the
parentheses are necessary because the operators [] and $
associate from right to left; without them, the statement
would have parsed as:
char $(1[ 128 ]);
which is obviously ridiculous.
If the contents of one of these field arrays is altered, the
"$0" field will reflect this change. For example, this
expression:
*$4 = 'A';
will change the first character of the fourth field to an upper-
case letter 'A'. Then, when the following input line:
120 PRINT "Name address Zip"
is processed, it would be printed as:
120 PRINT "Name Address Zip"
Fields may also be modified with the strcpy() function (see
below). For example, the expression:
strcpy( $4, "Addr." );
applied to the same line above would yield:
120 PRINT "Name Addr. Zip"
PREDEFINED VARIABLES
The following variables are pre-defined:
FS Field seperator (see below).
RS Record seperator (see below also).
NF Number of fields in current input
record (line).
NR Number of records processed thus far.
FILENAME Name of current input file.
BEGIN A special <pattern> that matches the
beginning of input text, before the
first record is read.
END A special <pattern> that matches the
end of input text, after the last
record has been read.
Bawk also provides some useful builtin functions for string
manipulation and printing:
printf(arg..) Exactly the printf() function from C.
getline() Reads the next record from the current
input file and returns 0 on end of file.
nextfile() Closes out the current input file and
begins processing the next file in the
list (if any).
strlen(s) Returns the length of its string argument.
strcpy(s,t) Copies the string "t" to the string "s".
strcmp(s,t) Compares the "s" to "t" and returns 0 if
they match.
toupper(c) Returns its character argument converted
to upper-case.
tolower(c) Returns its character argument converted
to lower-case.
match(s,@re@) Compares the string "s" to the regular
expression "re" and returns the number
of matches found (zero if none).
EXAMPLES
The following rules file will scan a C program, counting the
number of mismatched parentheses, brackets, and braces.
/[()\[\]{}]/
{
parens = parens + match( $0, @(@ );
parens = parens - match( $0, @)@ );
bracks = bracks + match( $0, @[@ );
bracks = bracks - match( $0, @]@ );
braces = braces + match( $0, @{@ );
braces = braces - match( $0, @}@ );
}
END { printf("parens=%d, brackets=%d, braces=%d\n",
parens, bracks, braces );
}
This program will capitalize the first word in every sentence of
a document:
BEGIN
{
RS = '.'; # set record seperator to a period
}
{
if ( match( $1, @^[a-z]@ ) )
*$1 = toupper( *$1 );
printf( "%s\n", $0 );
}
LIMITATIONS
Bawk was originally written in BDS C, but every attempt was made
to keep the code as portable as possible. The program should
be compilable with any "standard" C compiler. On CP/M systems
compiled with BDS C, bawk takes up about 24K.
An input record may be no longer than 128 characters. If longer
records are encountered, they terminate prematurely and the
next record starts where the previous one was hacked off.
A single pattern or action statement may be no longer than about
4K characters, excluding comments and whitespace. Since the
program is semi-compiled the tokenized version will probably
wind up being smaller than the source code, so the 4K figure is
only approximate.
AUTHOR
Bob Brodt
486 Linden Ave.
Bogota, NJ 07603
ACKNOWLEDGEMENTS
The concept for bawk (and 3/4 of the name!) was taken from
the program "awk" written by Afred V. Aho, Brian W. Kernighan
and Peter J. Weinberger. My apologies for any irreverences.
The regular expression compiler/parser was borrowed from a
program called "grep" and has been highly modified. Grep is
distributed by the DEC Users Society (DECUS) and is Copyright
(C) 1980 by DECUS. The author acknowledges DECUS with a nod of
thanks for giving their general permission and okey-dokey to
copy or modify the grep program.
UNIX is a trademark of AT&T Bell Labs.