rsalz@uunet.uu.net (Rich Salz) (06/06/90)
Submitted-by: "Arnold D. Robbins" <arnold@unix.cc.emory.edu>
Posting-number: Volume 22, Issue 92
Archive-name: gawk2.11/part06
#! /bin/sh
# This is a shell archive. Remove anything before this line, then feed it
# into a shell via "sh file" or similar. To overwrite existing files,
# type "sh file -c".
# The tool that generated this appeared in the comp.sources.unix newsgroup;
# send mail to comp-sources-unix@uunet.uu.net if you want that tool.
# Contents: ./gawk.texinfo.03 ./missing.d/strchr.c ./version.sh
# Wrapped by rsalz@litchi.bbn.com on Wed Jun 6 12:24:50 1990
PATH=/bin:/usr/bin:/usr/ucb ; export PATH
echo If this archive is complete, you will see the following message:
echo ' "shar: End of archive 6 (of 16)."'
if test -f './gawk.texinfo.03' -a "${1}" != "-c" ; then
echo shar: Will not clobber existing file \"'./gawk.texinfo.03'\"
else
echo shar: Extracting \"'./gawk.texinfo.03'\" \(49625 characters\)
sed "s/^X//" >'./gawk.texinfo.03' <<'END_OF_FILE'
X printf format, "----", "------" @}
X @{ printf format, $1, $2 @}' BBS-list
X@end example
X
XSee if you can use the @code{printf} statement to line up the headings and
Xtable data for our @file{inventory-shipped} example covered earlier in the
Xsection on the @code{print} statement (@pxref{Print}).
X
X@node Redirection, Special Files, Printf, Printing
X@section Redirecting Output of @code{print} and @code{printf}
X
X@cindex output redirection
X@cindex redirection of output
XSo far we have been dealing only with output that prints to the standard
Xoutput, usually your terminal. Both @code{print} and @code{printf} can be
Xtold to send their output to other places. This is called
X@dfn{redirection}.@refill
X
XA redirection appears after the @code{print} or @code{printf} statement.
XRedirections in @code{awk} are written just like redirections in shell
Xcommands, except that they are written inside the @code{awk} program.
X
X@menu
X* File/Pipe Redirection:: Redirecting Output to Files and Pipes.
X* Close Output:: How to close output files and pipes.
X@end menu
X
X@node File/Pipe Redirection, Close Output, Redirection, Redirection
X@subsection Redirecting Output to Files and Pipes
X
XHere are the three forms of output redirection. They are all shown for
Xthe @code{print} statement, but they work identically for @code{printf}
Xalso.
X
X@table @code
X@item print @var{items} > @var{output-file}
XThis type of redirection prints the items onto the output file
X@var{output-file}. The file name @var{output-file} can be any
Xexpression. Its value is changed to a string and then used as a
Xfile name (@pxref{Expressions}).@refill
X
XWhen this type of redirection is used, the @var{output-file} is erased
Xbefore the first output is written to it. Subsequent writes do not
Xerase @var{output-file}, but append to it. If @var{output-file} does
Xnot exist, then it is created.@refill
X
XFor example, here is how one @code{awk} program can write a list of
XBBS names to a file @file{name-list} and a list of phone numbers to a
Xfile @file{phone-list}. Each output file contains one name or number
Xper line.
X
X@example
Xawk '@{ print $2 > "phone-list"
X print $1 > "name-list" @}' BBS-list
X@end example
X
X@item print @var{items} >> @var{output-file}
XThis type of redirection prints the items onto the output file
X@var{output-file}. The difference between this and the
Xsingle-@samp{>} redirection is that the old contents (if any) of
X@var{output-file} are not erased. Instead, the @code{awk} output is
Xappended to the file.
X
X@cindex pipes for output
X@cindex output, piping
X@item print @var{items} | @var{command}
XIt is also possible to send output through a @dfn{pipe} instead of into a
Xfile. This type of redirection opens a pipe to @var{command} and writes
Xthe values of @var{items} through this pipe, to another process created
Xto execute @var{command}.@refill
X
XThe redirection argument @var{command} is actually an @code{awk}
Xexpression. Its value is converted to a string, whose contents give the
Xshell command to be run.
X
XFor example, this produces two files, one unsorted list of BBS names
Xand one list sorted in reverse alphabetical order:
X
X@example
Xawk '@{ print $1 > "names.unsorted"
X print $1 | "sort -r > names.sorted" @}' BBS-list
X@end example
X
XHere the unsorted list is written with an ordinary redirection while
Xthe sorted list is written by piping through the @code{sort} utility.
X
XHere is an example that uses redirection to mail a message to a mailing
Xlist @samp{bug-system}. This might be useful when trouble is encountered
Xin an @code{awk} script run periodically for system maintenance.
X
X@example
Xprint "Awk script failed:", $0 | "mail bug-system"
Xprint "at record number", FNR, "of", FILENAME | "mail bug-system"
Xclose("mail bug-system")
X@end example
X
XWe call the @code{close} function here because it's a good idea to close
Xthe pipe as soon as all the intended output has been sent to it.
X@xref{Close Output}, for more information on this.
X@end table
X
XRedirecting output using @samp{>}, @samp{>>}, or @samp{|} asks the system
Xto open a file or pipe only if the particular @var{file} or @var{command}
Xyou've specified has not already been written to by your program.@refill
X
X@node Close Output, , File/Pipe Redirection, Redirection
X@subsection Closing Output Files and Pipes
X@cindex closing output files and pipes
X@findex close
X
XWhen a file or pipe is opened, the file name or command associated with
Xit is remembered by @code{awk} and subsequent writes to the same file or
Xcommand are appended to the previous writes. The file or pipe stays
Xopen until @code{awk} exits. This is usually convenient.
X
XSometimes there is a reason to close an output file or pipe earlier
Xthan that. To do this, use the @code{close} function, as follows:
X
X@example
Xclose(@var{filename})
X@end example
X
X@noindent
Xor
X
X@example
Xclose(@var{command})
X@end example
X
XThe argument @var{filename} or @var{command} can be any expression.
XIts value must exactly equal the string used to open the file or pipe
Xto begin with---for example, if you open a pipe with this:
X
X@example
Xprint $1 | "sort -r > names.sorted"
X@end example
X
X@noindent
Xthen you must close it with this:
X
X@example
Xclose("sort -r > names.sorted")
X@end example
X
XHere are some reasons why you might need to close an output file:
X
X@itemize @bullet
X@item
XTo write a file and read it back later on in the same @code{awk}
Xprogram. Close the file when you are finished writing it; then
Xyou can start reading it with @code{getline} (@pxref{Getline}).
X
X@item
XTo write numerous files, successively, in the same @code{awk}
Xprogram. If you don't close the files, eventually you will exceed the
Xsystem limit on the number of open files in one process. So close
Xeach one when you are finished writing it.
X
X@item
XTo make a command finish. When you redirect output through a pipe,
Xthe command reading the pipe normally continues to try to read input
Xas long as the pipe is open. Often this means the command cannot
Xreally do its work until the pipe is closed. For example, if you
Xredirect output to the @code{mail} program, the message is not
Xactually sent until the pipe is closed.
X
X@item
XTo run the same program a second time, with the same arguments.
XThis is not the same thing as giving more input to the first run!
X
XFor example, suppose you pipe output to the @code{mail} program. If you
Xoutput several lines redirected to this pipe without closing it, they make
Xa single message of several lines. By contrast, if you close the pipe
Xafter each line of output, then each line makes a separate message.
X@end itemize
X
X@node Special Files, , Redirection, Printing
X@section Standard I/O Streams
X@cindex standard input
X@cindex standard output
X@cindex standard error output
X@cindex file descriptors
X
XRunning programs conventionally have three input and output streams
Xalready available to them for reading and writing. These are known as
Xthe @dfn{standard input}, @dfn{standard output}, and @dfn{standard error
Xoutput}. These streams are, by default, terminal input and output, but
Xthey are often redirected with the shell, via the @samp{<}, @samp{<<},
X@samp{>}, @samp{>>}, @samp{>&} and @samp{|} operators. Standard error
Xis used only for writing error messages; the reason we have two separate
Xstreams, standard output and standard error, is so that they can be
Xredirected separately.
X
X@c @cindex differences between @code{gawk} and @code{awk}
XIn other implementations of @code{awk}, the only way to write an error
Xmessage to standard error in an @code{awk} program is as follows:
X
X@example
Xprint "Serious error detected!\n" | "cat 1>&2"
X@end example
X
X@noindent
XThis works by opening a pipeline to a shell command which can access the
Xstandard error stream which it inherits from the @code{awk} process.
XThis is far from elegant, and is also inefficient, since it requires a
Xseparate process. So people writing @code{awk} programs have often
Xneglected to do this. Instead, they have sent the error messages to the
Xterminal, like this:
X
X@example
XNF != 4 @{
X printf("line %d skipped: doesn't have 4 fields\n", FNR) > "/dev/tty"
X@}
X@end example
X
X@noindent
XThis has the same effect most of the time, but not always: although the
Xstandard error stream is usually the terminal, it can be redirected, and
Xwhen that happens, writing to the terminal is not correct. In fact, if
X@code{awk} is run from a background job, it may not have a terminal at all.
XThen opening @file{/dev/tty} will fail.
X
X@code{gawk} provides special file names for accessing the three standard
Xstreams. When you redirect input or output in @code{gawk}, if the file name
Xmatches one of these special names, then @code{gawk} directly uses the
Xstream it stands for.
X
X@cindex @file{/dev/stdin}
X@cindex @file{/dev/stdout}
X@cindex @file{/dev/stderr}
X@cindex @file{/dev/fd/}
X@table @file
X@item /dev/stdin
XThe standard input (file descriptor 0).
X
X@item /dev/stdout
XThe standard output (file descriptor 1).
X
X@item /dev/stderr
XThe standard error output (file descriptor 2).
X
X@item /dev/fd/@var{n}
XThe file associated with file descriptor @var{n}. Such a file must have
Xbeen opened by the program initiating the @code{awk} execution (typically
Xthe shell). Unless you take special pains, only descriptors 0, 1 and 2
Xare available.
X@end table
X
XThe file names @file{/dev/stdin}, @file{/dev/stdout}, and @file{/dev/stderr}
Xare aliases for @file{/dev/fd/0}, @file{/dev/fd/1}, and @file{/dev/fd/2},
Xrespectively, but they are more self-explanatory.
X
XThe proper way to write an error message in a @code{gawk} program
Xis to use @file{/dev/stderr}, like this:
X
X@example
XNF != 4 @{
X printf("line %d skipped: doesn't have 4 fields\n", FNR) > "/dev/stderr"
X@}
X@end example
X
XRecognition of these special file names is disabled if @code{gawk} is in
Xcompatibility mode (@pxref{Command Line}).
X
X@node One-liners, Patterns, Printing, Top
X@chapter Useful ``One-liners''
X
X@cindex one-liners
XUseful @code{awk} programs are often short, just a line or two. Here is a
Xcollection of useful, short programs to get you started. Some of these
Xprograms contain constructs that haven't been covered yet. The description
Xof the program will give you a good idea of what is going on, but please
Xread the rest of the manual to become an @code{awk} expert!
X
X@table @code
X@item awk '@{ num_fields = num_fields + NF @}
X@itemx @ @ @ @ @ END @{ print num_fields @}'
XThis program prints the total number of fields in all input lines.
X
X@item awk 'length($0) > 80'
XThis program prints every line longer than 80 characters. The sole
Xrule has a relational expression as its pattern, and has no action (so the
Xdefault action, printing the record, is used).
X
X@item awk 'NF > 0'
XThis program prints every line that has at least one field. This is an
Xeasy way to delete blank lines from a file (or rather, to create a new
Xfile similar to the old file but from which the blank lines have been
Xdeleted).
X
X@item awk '@{ if (NF > 0) print @}'
XThis program also prints every line that has at least one field. Here we
Xallow the rule to match every line, then decide in the action whether
Xto print.
X
X@item awk@ 'BEGIN@ @{@ for (i = 1; i <= 7; i++)
X@itemx @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ print int(101 * rand()) @}'
XThis program prints 7 random numbers from 0 to 100, inclusive.
X
X@item ls -l @var{files} | awk '@{ x += $4 @} ; END @{ print "total bytes: " x @}'
XThis program prints the total number of bytes used by @var{files}.
X
X@item expand@ @var{file}@ |@ awk@ '@{ if (x < length()) x = length() @}
X@itemx @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ END @{ print "maximum line length is " x @}'
XThis program prints the maximum line length of @var{file}. The input
Xis piped through the @code{expand} program to change tabs into spaces,
Xso the widths compared are actually the right-margin columns.
X@end table
X
X@node Patterns, Actions, One-liners, Top
X@chapter Patterns
X@cindex pattern, definition of
X
XPatterns in @code{awk} control the execution of rules: a rule is
Xexecuted when its pattern matches the current input record. This
Xchapter tells all about how to write patterns.
X
X@menu
X* Kinds of Patterns:: A list of all kinds of patterns.
X The following subsections describe them in detail.
X
X* Empty:: The empty pattern, which matches every record.
X
X* Regexp:: Regular expressions such as @samp{/foo/}.
X
X* Comparison Patterns:: Comparison expressions such as @code{$1 > 10}.
X
X* Boolean Patterns:: Combining comparison expressions.
X
X* Expression Patterns:: Any expression can be used as a pattern.
X
X* Ranges:: Using pairs of patterns to specify record ranges.
X
X* BEGIN/END:: Specifying initialization and cleanup rules.
X@end menu
X
X@node Kinds of Patterns, Empty, Patterns, Patterns
X@section Kinds of Patterns
X@cindex patterns, types of
X
XHere is a summary of the types of patterns supported in @code{awk}.
X
X@table @code
X@item /@var{regular expression}/
XA regular expression as a pattern. It matches when the text of the
Xinput record fits the regular expression. (@xref{Regexp, , Regular
XExpressions as Patterns}.)
X
X@item @var{expression}
XA single expression. It matches when its value, converted to a number,
Xis nonzero (if a number) or nonnull (if a string). (@xref{Expression
XPatterns}.)
X
X@item @var{pat1}, @var{pat2}
XA pair of patterns separated by a comma, specifying a range of records.
X(@xref{Ranges, , Specifying Record Ranges With Patterns}.)
X
X@item BEGIN
X@itemx END
XSpecial patterns to supply start-up or clean-up information to
X@code{awk}. (@xref{BEGIN/END}.)
X
X@item @var{null}
XThe empty pattern matches every input record. (@xref{Empty, , The Empty
XPattern}.)
X@end table
X
X@node Empty, Regexp, Kinds of Patterns, Patterns
X@section The Empty Pattern
X
X@cindex empty pattern
X@cindex pattern, empty
XAn empty pattern is considered to match @emph{every} input record. For
Xexample, the program:@refill
X
X@example
Xawk '@{ print $1 @}' BBS-list
X@end example
X
X@noindent
Xprints just the first field of every record.
X
X@node Regexp, Comparison Patterns, Empty, Patterns
X@section Regular Expressions as Patterns
X@cindex pattern, regular expressions
X@cindex regexp
X@cindex regular expressions as patterns
X
XA @dfn{regular expression}, or @dfn{regexp}, is a way of describing a
Xclass of strings. A regular expression enclosed in slashes (@samp{/})
Xis an @code{awk} pattern that matches every input record whose text
Xbelongs to that class.
X
XThe simplest regular expression is a sequence of letters, numbers, or
Xboth. Such a regexp matches any string that contains that sequence.
XThus, the regexp @samp{foo} matches any string containing @samp{foo}.
XTherefore, the pattern @code{/foo/} matches any input record containing
X@samp{foo}. Other kinds of regexps let you specify more complicated
Xclasses of strings.
X
X@menu
X* Usage: Regexp Usage. How regexps are used in patterns.
X* Operators: Regexp Operators. How to write a regexp.
X* Case-sensitivity:: How to do case-insensitive matching.
X@end menu
X
X@node Regexp Usage, Regexp Operators, Regexp, Regexp
X@subsection How to Use Regular Expressions
X
XA regular expression can be used as a pattern by enclosing it in
Xslashes. Then the regular expression is matched against the entire text
Xof each record. (Normally, it only needs to match some part of the text
Xin order to succeed.) For example, this prints the second field of each
Xrecord that contains @samp{foo} anywhere:
X
X@example
Xawk '/foo/ @{ print $2 @}' BBS-list
X@end example
X
X@cindex regular expression matching operators
X@cindex string-matching operators
X@cindex operators, string-matching
X@cindex operators, regular expression matching
X@cindex regexp search operators
XRegular expressions can also be used in comparison expressions. Then
Xyou can specify the string to match against; it need not be the entire
Xcurrent input record. These comparison expressions can be used as
Xpatterns or in @code{if} and @code{while} statements.
X
X@table @code
X@item @var{exp} ~ /@var{regexp}/
XThis is true if the expression @var{exp} (taken as a character string)
Xis matched by @var{regexp}. The following example matches, or selects,
Xall input records with the upper-case letter @samp{J} somewhere in the
Xfirst field:@refill
X
X@example
Xawk '$1 ~ /J/' inventory-shipped
X@end example
X
XSo does this:
X
X@example
Xawk '@{ if ($1 ~ /J/) print @}' inventory-shipped
X@end example
X
X@item @var{exp} !~ /@var{regexp}/
XThis is true if the expression @var{exp} (taken as a character string)
Xis @emph{not} matched by @var{regexp}. The following example matches,
Xor selects, all input records whose first field @emph{does not} contain
Xthe upper-case letter @samp{J}:@refill
X
X@example
Xawk '$1 !~ /J/' inventory-shipped
X@end example
X@end table
X
X@cindex computed regular expressions
X@cindex regular expressions, computed
X@cindex dynamic regular expressions
XThe right hand side of a @samp{~} or @samp{!~} operator need not be a
Xconstant regexp (i.e., a string of characters between slashes). It may
Xbe any expression. The expression is evaluated, and converted if
Xnecessary to a string; the contents of the string are used as the
Xregexp. A regexp that is computed in this way is called a @dfn{dynamic
Xregexp}. For example:
X
X@example
Xidentifier_regexp = "[A-Za-z_][A-Za-z_0-9]+"
X$0 ~ identifier_regexp
X@end example
X
X@noindent
Xsets @code{identifier_regexp} to a regexp that describes @code{awk}
Xvariable names, and tests if the input record matches this regexp.
X
X@node Regexp Operators, Case-sensitivity, Regexp Usage, Regexp
X@subsection Regular Expression Operators
X@cindex metacharacters
X@cindex regular expression metacharacters
X
XYou can combine regular expressions with the following characters,
Xcalled @dfn{regular expression operators}, or @dfn{metacharacters}, to
Xincrease the power and versatility of regular expressions.
X
XHere is a table of metacharacters. All characters not listed in the
Xtable stand for themselves.
X
X@table @code
X@item ^
XThis matches the beginning of the string or the beginning of a line
Xwithin the string. For example:
X
X@example
X^@@chapter
X@end example
X
X@noindent
Xmatches the @samp{@@chapter} at the beginning of a string, and can be used
Xto identify chapter beginnings in Texinfo source files.
X
X@item $
XThis is similar to @samp{^}, but it matches only at the end of a string
Xor the end of a line within the string. For example:
X
X@example
Xp$
X@end example
X
X@noindent
Xmatches a record that ends with a @samp{p}.
X
X@item .
XThis matches any single character except a newline. For example:
X
X@example
X.P
X@end example
X
X@noindent
Xmatches any single character followed by a @samp{P} in a string. Using
Xconcatenation we can make regular expressions like @samp{U.A}, which
Xmatches any three-character sequence that begins with @samp{U} and ends
Xwith @samp{A}.
X
X@item [@dots{}]
XThis is called a @dfn{character set}. It matches any one of the
Xcharacters that are enclosed in the square brackets. For example:
X
X@example
X[MVX]
X@end example
X
X@noindent
Xmatches any of the characters @samp{M}, @samp{V}, or @samp{X} in a
Xstring.@refill
X
XRanges of characters are indicated by using a hyphen between the beginning
Xand ending characters, and enclosing the whole thing in brackets. For
Xexample:@refill
X
X@example
X[0-9]
X@end example
X
X@noindent
Xmatches any digit.
X
XTo include the character @samp{\}, @samp{]}, @samp{-} or @samp{^} in a
Xcharacter set, put a @samp{\} in front of it. For example:
X
X@example
X[d\]]
X@end example
X
X@noindent
Xmatches either @samp{]}, or @samp{d}.@refill
X
XThis treatment of @samp{\} is compatible with other @code{awk}
Ximplementations but incompatible with the proposed POSIX specification
Xfor @code{awk}. The current draft specifies the use of the same syntax
Xused in @code{egrep}.
X
XWe may change @code{gawk} to fit the standard, once we are sure it will
Xno longer change. For the meanwhile, the @samp{-a} option specifies the
Xtraditional @code{awk} syntax described above (which is also the
Xdefault), while the @samp{-e} option specifies @code{egrep} syntax.
X@xref{Options}.
X
XIn @code{egrep} syntax, backslash is not syntactically special within
Xsquare brackets. This means that special tricks have to be used to
Xrepresent the characters @samp{]}, @samp{-} and @samp{^} as members of a
Xcharacter set.
X
XTo match @samp{-}, write it as @samp{---}, which is a range containing
Xonly @samp{-}. You may also give @samp{-} as the first or last
Xcharacter in the set. To match @samp{^}, put it anywhere except as the
Xfirst character of a set. To match a @samp{]}, make it the first
Xcharacter in the set. For example:
X
X@example
X[]d^]
X@end example
X
X@noindent
Xmatches either @samp{]}, @samp{d} or @samp{^}.@refill
X
X@item [^ @dots{}]
XThis is a @dfn{complemented character set}. The first character after
Xthe @samp{[} @emph{must} be a @samp{^}. It matches any characters
X@emph{except} those in the square brackets. For example:
X
X@example
X[^0-9]
X@end example
X
X@noindent
Xmatches any character that is not a digit.
X
X@item |
XThis is the @dfn{alternation operator} and it is used to specify
Xalternatives. For example:
X
X@example
X^P|[0-9]
X@end example
X
X@noindent
Xmatches any string that matches either @samp{^P} or @samp{[0-9]}. This
Xmeans it matches any string that contains a digit or starts with @samp{P}.
X
XThe alternation applies to the largest possible regexps on either side.
X@item (@dots{})
XParentheses are used for grouping in regular expressions as in
Xarithmetic. They can be used to concatenate regular expressions
Xcontaining the alternation operator, @samp{|}.
X
X@item *
XThis symbol means that the preceding regular expression is to be
Xrepeated as many times as possible to find a match. For example:
X
X@example
Xph*
X@end example
X
X@noindent
Xapplies the @samp{*} symbol to the preceding @samp{h} and looks for matches
Xto one @samp{p} followed by any number of @samp{h}s. This will also match
Xjust @samp{p} if no @samp{h}s are present.
X
XThe @samp{*} repeats the @emph{smallest} possible preceding expression.
X(Use parentheses if you wish to repeat a larger expression.) It finds
Xas many repetitions as possible. For example:
X
X@example
Xawk '/\(c[ad][ad]*r x\)/ @{ print @}' sample
X@end example
X
X@noindent
Xprints every record in the input containing a string of the form
X@samp{(car x)}, @samp{(cdr x)}, @samp{(cadr x)}, and so on.@refill
X
X@item +
XThis symbol is similar to @samp{*}, but the preceding expression must be
Xmatched at least once. This means that:
X
X@example
Xwh+y
X@end example
X
X@noindent
Xwould match @samp{why} and @samp{whhy} but not @samp{wy}, whereas
X@samp{wh*y} would match all three of these strings. This is a simpler
Xway of writing the last @samp{*} example:
X
X@example
Xawk '/\(c[ad]+r x\)/ @{ print @}' sample
X@end example
X
X@item ?
XThis symbol is similar to @samp{*}, but the preceding expression can be
Xmatched once or not at all. For example:
X
X@example
Xfe?d
X@end example
X
X@noindent
Xwill match @samp{fed} or @samp{fd}, but nothing else.@refill
X
X@item \
XThis is used to suppress the special meaning of a character when
Xmatching. For example:
X
X@example
X\$
X@end example
X
X@noindent
Xmatches the character @samp{$}.
X
XThe escape sequences used for string constants (@pxref{Constants}) are
Xvalid in regular expressions as well; they are also introduced by a
X@samp{\}.
X@end table
X
XIn regular expressions, the @samp{*}, @samp{+}, and @samp{?} operators have
Xthe highest precedence, followed by concatenation, and finally by @samp{|}.
XAs in arithmetic, parentheses can change how operators are grouped.@refill
X
X@node Case-sensitivity,, Regexp Operators, Regexp
X@subsection Case-sensitivity in Matching
X
XCase is normally significant in regular expressions, both when matching
Xordinary characters (i.e., not metacharacters), and inside character
Xsets. Thus a @samp{w} in a regular expression matches only a lower case
X@samp{w} and not an upper case @samp{W}.
X
XThe simplest way to do a case-independent match is to use a character
Xset: @samp{[Ww]}. However, this can be cumbersome if you need to use it
Xoften; and it can make the regular expressions harder for humans to
Xread. There are two other alternatives that you might prefer.
X
XOne way to do a case-insensitive match at a particular point in the
Xprogram is to convert the data to a single case, using the
X@code{tolower} or @code{toupper} built-in string functions (which we
Xhaven't discussed yet; @pxref{String Functions}). For example:
X
X@example
Xtolower($1) ~ /foo/ @{ @dots{} @}
X@end example
X
X@noindent
Xconverts the first field to lower case before matching against it.
X
XAnother method is to set the variable @code{IGNORECASE} to a nonzero
Xvalue (@pxref{Built-in Variables}). When @code{IGNORECASE} is not zero,
X@emph{all} regexp operations ignore case. Changing the value of
X@code{IGNORECASE} dynamically controls the case sensitivity of your
Xprogram as it runs. Case is significant by default because
X@code{IGNORECASE} (like most variables) is initialized to zero.
X
X@example
Xx = "aB"
Xif (x ~ /ab/) @dots{} # this test will fail
X
XIGNORECASE = 1
Xif (x ~ /ab/) @dots{} # now it will succeed
X@end example
X
XYou cannot generally use @code{IGNORECASE} to make certain rules
Xcase-insensitive and other rules case-sensitive, because there is no way
Xto set @code{IGNORECASE} just for the pattern of a particular rule. To
Xdo this, you must use character sets or @code{tolower}. However, one
Xthing you can do only with @code{IGNORECASE} is turn case-sensitivity on
Xor off dynamically for all the rules at once.
X
X@code{IGNORECASE} can be set on the command line, or in a @code{BEGIN}
Xrule. Setting @code{IGNORECASE} from the command line is a way to make
Xa program case-insensitive without having to edit it.
X
XThe value of @code{IGNORECASE} has no effect if @code{gawk} is in
Xcompatibility mode (@pxref{Command Line}). Case is always significant
Xin compatibility mode.
X
X@node Comparison Patterns, Boolean Patterns, Regexp, Patterns
X@section Comparison Expressions as Patterns
X@cindex comparison expressions as patterns
X@cindex pattern, comparison expressions
X@cindex relational operators
X@cindex operators, relational
X
X@dfn{Comparison patterns} test relationships such as equality between
Xtwo strings or numbers. They are a special case of expression patterns
X(@pxref{Expression Patterns}). They are written with @dfn{relational
Xoperators}, which are a superset of those in C. Here is a table of
Xthem:
X
X@table @code
X@item @var{x} < @var{y}
XTrue if @var{x} is less than @var{y}.
X
X@item @var{x} <= @var{y}
XTrue if @var{x} is less than or equal to @var{y}.
X
X@item @var{x} > @var{y}
XTrue if @var{x} is greater than @var{y}.
X
X@item @var{x} >= @var{y}
XTrue if @var{x} is greater than or equal to @var{y}.
X
X@item @var{x} == @var{y}
XTrue if @var{x} is equal to @var{y}.
X
X@item @var{x} != @var{y}
XTrue if @var{x} is not equal to @var{y}.
X
X@item @var{x} ~ @var{y}
XTrue if @var{x} matches the regular expression described by @var{y}.
X
X@item @var{x} !~ @var{y}
XTrue if @var{x} does not match the regular expression described by @var{y}.
X@end table
X
XThe operands of a relational operator are compared as numbers if they
Xare both numbers. Otherwise they are converted to, and compared as,
Xstrings (@pxref{Conversion}). Strings are compared by comparing the
Xfirst character of each, then the second character of each, and so on,
Xuntil there is a difference. If the two strings are equal until the
Xshorter one runs out, the shorter one is considered to be less than the
Xlonger one. Thus, @code{"10"} is less than @code{"9"}.
X
XThe left operand of the @samp{~} and @samp{!~} operators is a string.
XThe right operand is either a constant regular expression enclosed in
Xslashes (@code{/@var{regexp}/}), or any expression, whose string value
Xis used as a dynamic regular expression (@pxref{Regexp Usage}).
X
XThe following example prints the second field of each input record
Xwhose first field is precisely @samp{foo}.
X
X@example
Xawk '$1 == "foo" @{ print $2 @}' BBS-list
X@end example
X
X@noindent
XContrast this with the following regular expression match, which would
Xaccept any record with a first field that contains @samp{foo}:
X
X@example
Xawk '$1 ~ "foo" @{ print $2 @}' BBS-list
X@end example
X
X@noindent
Xor, equivalently, this one:
X
X@example
Xawk '$1 ~ /foo/ @{ print $2 @}' BBS-list
X@end example
X
X@node Boolean Patterns, Expression Patterns, Comparison Patterns, Patterns
X@section Boolean Operators and Patterns
X@cindex patterns, boolean
X@cindex boolean patterns
X
XA @dfn{boolean pattern} is an expression which combines other patterns
Xusing the @dfn{boolean operators} ``or'' (@samp{||}), ``and''
X(@samp{&&}), and ``not'' (@samp{!}). Whether the boolean pattern
Xmatches an input record depends on whether its subpatterns match.
X
XFor example, the following command prints all records in the input file
X@file{BBS-list} that contain both @samp{2400} and @samp{foo}.@refill
X
X@example
Xawk '/2400/ && /foo/' BBS-list
X@end example
X
XThe following command prints all records in the input file
X@file{BBS-list} that contain @emph{either} @samp{2400} or @samp{foo}, or
Xboth.@refill
X
X@example
Xawk '/2400/ || /foo/' BBS-list
X@end example
X
XThe following command prints all records in the input file
X@file{BBS-list} that do @emph{not} contain the string @samp{foo}.
X
X@example
Xawk '! /foo/' BBS-list
X@end example
X
XNote that boolean patterns are a special case of expression patterns
X(@pxref{Expression Patterns}); they are expressions that use the boolean
Xoperators. For complete information on the boolean operators, see
X@ref{Boolean Ops}.
X
XThe subpatterns of a boolean pattern can be constant regular
Xexpressions, comparisons, or any other @code{gawk} expressions. Range
Xpatterns are not expressions, so they cannot appear inside boolean
Xpatterns. Likewise, the special patterns @code{BEGIN} and @code{END},
Xwhich never match any input record, are not expressions and cannot
Xappear inside boolean patterns.
X
X@node Expression Patterns, Ranges, Boolean Patterns, Patterns
X@section Expressions as Patterns
X
XAny @code{awk} expression is valid also as a pattern in @code{gawk}.
XThen the pattern ``matches'' if the expression's value is nonzero (if a
Xnumber) or nonnull (if a string).
X
XThe expression is reevaluated each time the rule is tested against a new
Xinput record. If the expression uses fields such as @code{$1}, the
Xvalue depends directly on the new input record's text; otherwise, it
Xdepends only on what has happened so far in the execution of the
X@code{awk} program, but that may still be useful.
X
XComparison patterns are actually a special case of this. For
Xexample, the expression @code{$5 == "foo"} has the value 1 when the
Xvalue of @code{$5} equals @code{"foo"}, and 0 otherwise; therefore, this
Xexpression as a pattern matches when the two values are equal.
X
XBoolean patterns are also special cases of expression patterns.
X
XA constant regexp as a pattern is also a special case of an expression
Xpattern. @code{/foo/} as an expression has the value 1 if @samp{foo}
Xappears in the current input record; thus, as a pattern, @code{/foo/}
Xmatches any record containing @samp{foo}.
X
XOther implementations of @code{awk} are less general than @code{gawk}:
Xthey allow comparison expressions, and boolean combinations thereof
X(optionally with parentheses), but not necessarily other kinds of
Xexpressions.
X
X@node Ranges, BEGIN/END, Expression Patterns, Patterns
X@section Specifying Record Ranges With Patterns
X
X@cindex range pattern
X@cindex patterns, range
XA @dfn{range pattern} is made of two patterns separated by a comma, of
Xthe form @code{@var{begpat}, @var{endpat}}. It matches ranges of
Xconsecutive input records. The first pattern @var{begpat} controls
Xwhere the range begins, and the second one @var{endpat} controls where
Xit ends. For example,@refill
X
X@example
Xawk '$1 == "on", $1 == "off"'
X@end example
X
X@noindent
Xprints every record between @samp{on}/@samp{off} pairs, inclusive.
X
XIn more detail, a range pattern starts out by matching @var{begpat}
Xagainst every input record; when a record matches @var{begpat}, the
Xrange pattern becomes @dfn{turned on}. The range pattern matches this
Xrecord. As long as it stays turned on, it automatically matches every
Xinput record read. But meanwhile, it also matches @var{endpat} against
Xevery input record, and when that succeeds, the range pattern is turned
Xoff again for the following record. Now it goes back to checking
X@var{begpat} against each record.
X
XThe record that turns on the range pattern and the one that turns it
Xoff both match the range pattern. If you don't want to operate on
Xthese records, you can write @code{if} statements in the rule's action
Xto distinguish them.
X
XIt is possible for a pattern to be turned both on and off by the same
Xrecord, if both conditions are satisfied by that record. Then the action is
Xexecuted for just that record.
X
X@node BEGIN/END,, Ranges, Patterns
X@section @code{BEGIN} and @code{END} Special Patterns
X
X@cindex @code{BEGIN} special pattern
X@cindex patterns, @code{BEGIN}
X@cindex @code{END} special pattern
X@cindex patterns, @code{END}
X@code{BEGIN} and @code{END} are special patterns. They are not used to
Xmatch input records. Rather, they are used for supplying start-up or
Xclean-up information to your @code{awk} script. A @code{BEGIN} rule is
Xexecuted, once, before the first input record has been read. An @code{END}
Xrule is executed, once, after all the input has been read. For
Xexample:@refill
X
X@group
X@example
Xawk 'BEGIN @{ print "Analysis of `foo'" @}
X /foo/ @{ ++foobar @}
X END @{ print "`foo' appears " foobar " times." @}' BBS-list
X@end example
X@end group
X
XThis program finds out how many times the string @samp{foo} appears in
Xthe input file @file{BBS-list}. The @code{BEGIN} rule prints a title
Xfor the report. There is no need to use the @code{BEGIN} rule to
Xinitialize the counter @code{foobar} to zero, as @code{awk} does this
Xfor us automatically (@pxref{Variables}).
X
XThe second rule increments the variable @code{foobar} every time a
Xrecord containing the pattern @samp{foo} is read. The @code{END} rule
Xprints the value of @code{foobar} at the end of the run.@refill
X
XThe special patterns @code{BEGIN} and @code{END} cannot be used in ranges
Xor with boolean operators.
X
XAn @code{awk} program may have multiple @code{BEGIN} and/or @code{END}
Xrules. They are executed in the order they appear, all the @code{BEGIN}
Xrules at start-up and all the @code{END} rules at termination.
X
XMultiple @code{BEGIN} and @code{END} sections are useful for writing
Xlibrary functions, since each library can have its own @code{BEGIN} or
X@code{END} rule to do its own initialization and/or cleanup. Note that
Xthe order in which library functions are named on the command line
Xcontrols the order in which their @code{BEGIN} and @code{END} rules are
Xexecuted. Therefore you have to be careful to write such rules in
Xlibrary files so that it doesn't matter what order they are executed in.
X@xref{Command Line}, for more information on using library functions.
X
XIf an @code{awk} program only has a @code{BEGIN} rule, and no other
Xrules, then the program exits after the @code{BEGIN} rule has been run.
X(Older versions of @code{awk} used to keep reading and ignoring input
Xuntil end of file was seen.) However, if an @code{END} rule exists as
Xwell, then the input will be read, even if there are no other rules in
Xthe program. This is necessary in case the @code{END} rule checks the
X@code{NR} variable.
X
X@code{BEGIN} and @code{END} rules must have actions; there is no default
Xaction for these rules since there is no current record when they run.
X
X@node Actions, Expressions, Patterns, Top
X@chapter Actions: Overview
X@cindex action, definition of
X@cindex curly braces
X@cindex action, curly braces
X@cindex action, separating statements
X
XAn @code{awk} @dfn{program} or @dfn{script} consists of a series of
X@dfn{rules} and function definitions, interspersed. (Functions are
Xdescribed later; see @ref{User-defined}.)
X
XA rule contains a pattern and an @dfn{action}, either of which may be
Xomitted. The purpose of the action is to tell @code{awk} what to do
Xonce a match for the pattern is found. Thus, the entire program
Xlooks somewhat like this:
X
X@example
X@r{[}@var{pattern}@r{]} @r{[}@{ @var{action} @}@r{]}
X@r{[}@var{pattern}@r{]} @r{[}@{ @var{action} @}@r{]}
X@dots{}
Xfunction @var{name} (@var{args}) @{ @dots{} @}
X@dots{}
X@end example
X
XAn action consists of one or more @code{awk} @dfn{statements}, enclosed
Xin curly braces (@samp{@{} and @samp{@}}). Each statement specifies one
Xthing to be done. The statements are separated by newlines or
Xsemicolons.
X
XThe curly braces around an action must be used even if the action
Xcontains only one statement, or even if it contains no statements at
Xall. However, if you omit the action entirely, omit the curly braces as
Xwell. (An omitted action is equivalent to @samp{@{ print $0 @}}.)
X
XHere are the kinds of statement supported in @code{awk}:
X
X@itemize @bullet
X@item
XExpressions, which can call functions or assign values to variables
X(@pxref{Expressions}). Executing this kind of statement simply computes
Xthe value of the expression and then ignores it. This is useful when
Xthe expression has side effects (@pxref{Assignment Ops}).
X
X@item
XControl statements, which specify the control flow of @code{awk}
Xprograms. The @code{awk} language gives you C-like constructs
X(@code{if}, @code{for}, @code{while}, and so on) as well as a few
Xspecial ones (@pxref{Statements}).@refill
X
X@item
XCompound statements, which consist of one or more statements enclosed in
Xcurly braces. A compound statement is used in order to put several
Xstatements together in the body of an @code{if}, @code{while}, @code{do}
Xor @code{for} statement.
X
X@item
XInput control, using the @code{getline} function (@pxref{Getline}),
Xand the @code{next} statement (@pxref{Next Statement}).
X
X@item
XOutput statements, @code{print} and @code{printf}. @xref{Printing}.
X
X@item
XDeletion statements, for deleting array elements. @xref{Delete}.
X@end itemize
X
X@iftex
XThe next two chapters cover in detail expressions and control
Xstatements, respectively. We go on to treat arrays, and built-in
Xfunctions, both of which are used in expressions. Then we proceed
Xto discuss how to define your own functions.
X@end iftex
X
X@node Expressions, Statements, Actions, Top
X@chapter Actions: Expressions
X@cindex expression
X
XExpressions are the basic building block of @code{awk} actions. An
Xexpression evaluates to a value, which you can print, test, store in a
Xvariable or pass to a function.
X
XBut, beyond that, an expression can assign a new value to a variable
Xor a field, with an assignment operator.
X
XAn expression can serve as a statement on its own. Most other kinds of
Xstatement contain one or more expressions which specify data to be
Xoperated on. As in other languages, expressions in @code{awk} include
Xvariables, array references, constants, and function calls, as well as
Xcombinations of these with various operators.
X
X@menu
X* Constants:: String, numeric, and regexp constants.
X* Variables:: Variables give names to values for later use.
X* Arithmetic Ops:: Arithmetic operations (@samp{+}, @samp{-}, etc.)
X* Concatenation:: Concatenating strings.
X* Comparison Ops:: Comparison of numbers and strings with @samp{<}, etc.
X* Boolean Ops:: Combining comparison expressions using boolean operators
X @samp{||} (``or''), @samp{&&} (``and'') and @samp{!} (``not'').
X
X* Assignment Ops:: Changing the value of a variable or a field.
X* Increment Ops:: Incrementing the numeric value of a variable.
X
X* Conversion:: The conversion of strings to numbers and vice versa.
X* Conditional Exp:: Conditional expressions select between two subexpressions
X under control of a third subexpression.
X* Function Calls:: A function call is an expression.
X* Precedence:: How various operators nest.
X@end menu
X
X@node Constants, Variables, Expressions, Expressions
X@section Constant Expressions
X@cindex constants, types of
X@cindex string constants
X
XThe simplest type of expression is the @dfn{constant}, which always has
Xthe same value. There are three types of constant: numeric constants,
Xstring constants, and regular expression constants.
X
X@cindex numeric constant
X@cindex numeric value
XA @dfn{numeric constant} stands for a number. This number can be an
Xinteger, a decimal fraction, or a number in scientific (exponential)
Xnotation. Note that all numeric values are represented within
X@code{awk} in double-precision floating point. Here are some examples
Xof numeric constants, which all have the same value:
X
X@example
X105
X1.05e+2
X1050e-1
X@end example
X
XA string constant consists of a sequence of characters enclosed in
Xdouble-quote marks. For example:
X
X@example
X"parrot"
X@end example
X
X@noindent
X@c @cindex differences between @code{gawk} and @code{awk}
Xrepresents the string whose contents are @samp{parrot}. Strings in
X@code{gawk} can be of any length and they can contain all the possible
X8-bit ASCII characters including ASCII NUL. Other @code{awk}
Ximplementations may have difficulty with some character codes.@refill
X
X@cindex escape sequence notation
XSome characters cannot be included literally in a string constant. You
Xrepresent them instead with @dfn{escape sequences}, which are character
Xsequences beginning with a backslash (@samp{\}).
X
XOne use of an escape sequence is to include a double-quote character in
Xa string constant. Since a plain double-quote would end the string, you
Xmust use @samp{\"} to represent a single double-quote character as a
Xpart of the string. Backslash itself is another character that can't be
Xincluded normally; you write @samp{\\} to put one backslash in the
Xstring. Thus, the string whose contents are the two characters
X@samp{"\} must be written @code{"\"\\"}.
X
XAnother use of backslash is to represent unprintable characters
Xsuch as newline. While there is nothing to stop you from writing most
Xof these characters directly in a string constant, they may look ugly.
X
XHere is a table of all the escape sequences used in @code{awk}:
X
X@table @code
X@item \\
XRepresents a literal backslash, @samp{\}.
X
X@item \a
XRepresents the ``alert'' character, control-g, ASCII code 7.
X
X@item \b
XRepresents a backspace, control-h, ASCII code 8.
X
X@item \f
XRepresents a formfeed, control-l, ASCII code 12.
X
X@item \n
XRepresents a newline, control-j, ASCII code 10.
X
X@item \r
XRepresents a carriage return, control-m, ASCII code 13.
X
X@item \t
XRepresents a horizontal tab, control-i, ASCII code 9.
X
X@item \v
XRepresents a vertical tab, control-k, ASCII code 11.
X
X@item \@var{nnn}
XRepresents the octal value @var{nnn}, where @var{nnn} are one to three
Xdigits between 0 and 7. For example, the code for the ASCII ESC
X(escape) character is @samp{\033}.@refill
X
X@item \x@var{hh@dots{}}
XRepresents the hexadecimal value @var{hh}, where @var{hh} are hexadecimal
Xdigits (@samp{0} through @samp{9} and either @samp{A} through @samp{F} or
X@samp{a} through @samp{f}). Like the same construct in ANSI C, the escape
Xsequence continues until the first non-hexadecimal digit is seen. However,
Xusing more than two hexadecimal digits produces undefined results.@refill
X@end table
X
XA constant regexp is a regular expression description enclosed in
Xslashes, such as @code{/^beginning and end$/}. Most regexps used in
X@code{awk} programs are constant, but the @samp{~} and @samp{!~}
Xoperators can also match computed or ``dynamic'' regexps (@pxref{Regexp
XUsage}).
X
XConstant regexps are useful only with the @samp{~} and @samp{!~} operators;
Xyou cannot assign them to variables or print them. They are not truly
Xexpressions in the usual sense.
X
X@node Variables, Arithmetic Ops, Constants, Expressions
X@section Variables
X@cindex variables, user-defined
X@cindex user-defined variables
X
XVariables let you give names to values and refer to them later. You have
Xalready seen variables in many of the examples. The name of a variable
Xmust be a sequence of letters, digits and underscores, but it may not begin
Xwith a digit. Case is significant in variable names; @code{a} and @code{A}
Xare distinct variables.
X
XA variable name is a valid expression by itself; it represents the
Xvariable's current value. Variables are given new values with
X@dfn{assignment operators} and @dfn{increment operators}.
X@xref{Assignment Ops}.
X
XA few variables have special built-in meanings, such as @code{FS}, the
Xfield separator, and @code{NF}, the number of fields in the current
Xinput record. @xref{Built-in Variables}, for a list of them. These
Xbuilt-in variables can be used and assigned just like all other
Xvariables, but their values are also used or changed automatically by
X@code{awk}. Each built-in variable's name is made entirely of upper case
Xletters.
X
XVariables in @code{awk} can be assigned either numeric values or string
Xvalues. By default, variables are initialized to the null string, which
Xis effectively zero if converted to a number. So there is no need to
X``initialize'' each variable explicitly in @code{awk}, the way you would
Xneed to do in C or most other traditional programming languages.
X
X@menu
X* Assignment Options:: Setting variables on the command line and a summary
X of command line syntax. This is an advanced method
X of input.
X@end menu
X
X@node Assignment Options,, Variables, Variables
X@subsection Assigning Variables on the Command Line
X
XYou can set any @code{awk} variable by including a @dfn{variable assignment}
Xamong the arguments on the command line when you invoke @code{awk}
X(@pxref{Command Line}). Such an assignment has this form:
X
X@example
X@var{variable}=@var{text}
X@end example
X
X@noindent
XWith it, you can set a variable either at the beginning of the
X@code{awk} run or in between input files.
X
XIf you precede the assignment with the @samp{-v} option, like this:
X
X@example
X-v @var{variable}=@var{text}
X@end example
X
X@noindent
Xthen the variable is set at the very beginning, before even the
X@code{BEGIN} rules are run. The @samp{-v} option and its assignment
Xmust precede all the file name arguments.
X
XOtherwise, the variable assignment is performed at a time determined by
Xits position among the input file arguments: after the processing of the
Xpreceding input file argument. For example:
X
X@example
Xawk '@{ print $n @}' n=4 inventory-shipped n=2 BBS-list
X@end example
X
X@noindent
Xprints the value of field number @code{n} for all input records. Before
Xthe first file is read, the command line sets the variable @code{n}
Xequal to 4. This causes the fourth field to be printed in lines from
Xthe file @file{inventory-shipped}. After the first file has finished,
Xbut before the second file is started, @code{n} is set to 2, so that the
Xsecond field is printed in lines from @file{BBS-list}.
X
XCommand line arguments are made available for explicit examination by
Xthe @code{awk} program in an array named @code{ARGV} (@pxref{Built-in
XVariables}).
X
X@node Arithmetic Ops, Concatenation, Variables, Expressions
X@section Arithmetic Operators
X@cindex arithmetic operators
X@cindex operators, arithmetic
X@cindex addition
X@cindex subtraction
X@cindex multiplication
X@cindex division
X@cindex remainder
X@cindex quotient
X@cindex exponentiation
X
XThe @code{awk} language uses the common arithmetic operators when
Xevaluating expressions. All of these arithmetic operators follow normal
Xprecedence rules, and work as you would expect them to. This example
Xdivides field three by field four, adds field two, stores the result
Xinto field one, and prints the resulting altered input record:
X
X@example
Xawk '@{ $1 = $2 + $3 / $4; print @}' inventory-shipped
X@end example
X
XThe arithmetic operators in @code{awk} are:
X
X@table @code
X@item @var{x} + @var{y}
XAddition.
X
X@item @var{x} - @var{y}
XSubtraction.
X
X@item - @var{x}
XNegation.
X
X@item @var{x} * @var{y}
XMultiplication.
X
X@item @var{x} / @var{y}
XDivision. Since all numbers in @code{awk} are double-precision
Xfloating point, the result is not rounded to an integer: @code{3 / 4}
Xhas the value 0.75.
X
X@item @var{x} % @var{y}
X@c @cindex differences between @code{gawk} and @code{awk}
XRemainder. The quotient is rounded toward zero to an integer,
Xmultiplied by @var{y} and this result is subtracted from @var{x}.
XThis operation is sometimes known as ``trunc-mod''. The following
Xrelation always holds:
X
X@example
Xb * int(a / b) + (a % b) == a
X@end example
X
XOne undesirable effect of this definition of remainder is that
X@code{@var{x} % @var{y}} is negative if @var{x} is negative. Thus,
X
X@example
X-17 % 8 = -1
X@end example
X
XIn other @code{awk} implementations, the signedness of the remainder
Xmay be machine dependent.
X
X@item @var{x} ^ @var{y}
X@itemx @var{x} ** @var{y}
XExponentiation: @var{x} raised to the @var{y} power. @code{2 ^ 3} has
Xthe value 8. The character sequence @samp{**} is equivalent to
X@samp{^}.
X@end table
X
X@node Concatenation, Comparison Ops, Arithmetic Ops, Expressions
X@section String Concatenation
X
X@cindex string operators
X@cindex operators, string
X@cindex concatenation
XThere is only one string operation: concatenation. It does not have a
Xspecific operator to represent it. Instead, concatenation is performed by
Xwriting expressions next to one another, with no operator. For example:
X
X@example
Xawk '@{ print "Field number one: " $1 @}' BBS-list
X@end example
X
X@noindent
Xproduces, for the first record in @file{BBS-list}:
X
X@example
XField number one: aardvark
X@end example
X
XWithout the space in the string constant after the @samp{:}, the line
Xwould run together. For example:
X
X@example
Xawk '@{ print "Field number one:" $1 @}' BBS-list
X@end example
X
X@noindent
Xproduces, for the first record in @file{BBS-list}:
X
X@example
XField number one:aardvark
X@end example
X
XSince string concatenation does not have an explicit operator, it is
END_OF_FILE
if test 49625 -ne `wc -c <'./gawk.texinfo.03'`; then
echo shar: \"'./gawk.texinfo.03'\" unpacked with wrong size!
fi
# end of './gawk.texinfo.03'
fi
if test -f './missing.d/strchr.c' -a "${1}" != "-c" ; then
echo shar: Will not clobber existing file \"'./missing.d/strchr.c'\"
else
echo shar: Extracting \"'./missing.d/strchr.c'\" \(543 characters\)
sed "s/^X//" >'./missing.d/strchr.c' <<'END_OF_FILE'
X/*
X * strchr --- search a string for a character
X *
X * We supply this routine for those systems that aren't standard yet.
X */
X
Xchar *
Xstrchr (str, c)
Xregister char *str, c;
X{
X for (; *str; str++)
X if (*str == c)
X return str;
X
X return NULL;
X}
X
X/*
X * strrchr --- find the last occurrence of a character in a string
X *
X * We supply this routine for those systems that aren't standard yet.
X */
X
Xchar *
Xstrrchr (str, c)
Xregister char *str, c;
X{
X register char *save = NULL;
X
X for (; *str; str++)
X if (*str == c)
X save = str;
X
X return save;
X}
END_OF_FILE
if test 543 -ne `wc -c <'./missing.d/strchr.c'`; then
echo shar: \"'./missing.d/strchr.c'\" unpacked with wrong size!
fi
# end of './missing.d/strchr.c'
fi
if test -f './version.sh' -a "${1}" != "-c" ; then
echo shar: Will not clobber existing file \"'./version.sh'\"
else
echo shar: Extracting \"'./version.sh'\" \(1509 characters\)
sed "s/^X//" >'./version.sh' <<'END_OF_FILE'
X#! /bin/sh
X
X# version.sh --- create version.c
X
Xif [ "x$1" = "x" ]
Xthen
X echo you must specify a release number on the command line
X exit 1
Xfi
X
XRELEASE="$1"
X
Xcat << EOF
Xchar *version_string = "@(#)Gnu Awk (gawk) ${RELEASE}";
X
X/* 1.02 fixed /= += *= etc to return the new Left Hand Side instead
X of the Right Hand Side */
X
X/* 1.03 Fixed split() to treat strings of space and tab as FS if
X the split char is ' '.
X
X Added -v option to print version number
X
X Fixed bug that caused rounding when printing large numbers */
X
X/* 2.00beta Incorporated the functionality of the "new" awk as described
X the book (reference not handy). Extensively tested, but no
X doubt still buggy. Badly needs tuning and cleanup, in
X particular in memory management which is currently almost
X non-existent. */
X
X/* 2.01 JF: Modified to compile under GCC, and fixed a few
X bugs while I was at it. I hope I didn't add any more.
X I modified parse.y to reduce the number of reduce/reduce
X conflicts. There are still a few left. */
X
X/* 2.02 Fixed JF's bugs; improved memory management, still needs
X lots of work. */
X
X/* 2.10 Major grammar rework and lots of bug fixes from David.
X Major changes for performance enhancements from David.
X A number of minor bug fixes and new features from Arnold.
X Changes for MSDOS from Conrad Kwok and Scott Garfinkle.
X The gawk.texinfo and info files included! */
X
X/* 2.11 Bug fix release to 2.10. Lots of changes for portability,
X speed, and configurability. */
XEOF
Xexit 0
END_OF_FILE
if test 1509 -ne `wc -c <'./version.sh'`; then
echo shar: \"'./version.sh'\" unpacked with wrong size!
fi
# end of './version.sh'
fi
echo shar: End of archive 6 \(of 16\).
cp /dev/null ark6isdone
MISSING=""
for I in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 ; do
if test ! -f ark${I}isdone ; then
MISSING="${MISSING} ${I}"
fi
done
if test "${MISSING}" = "" ; then
echo You have unpacked all 16 archives.
rm -f ark[1-9]isdone ark[1-9][0-9]isdone
else
echo You still must unpack the following archives:
echo " " ${MISSING}
fi
exit 0
exit 0 # Just in case...
--
Please send comp.sources.unix-related mail to rsalz@uunet.uu.net.
Use a domain-based address or give alternate paths, or you may lose out.