ptx(1)                                                     ptx(1)



     NAME
          ptx - permuted index

     SYNOPSIS
          ptx [-b break] [-f] [-g gap] [-i ignore] [-o only] [-r] [-t]
          [-w n] [input[output]]

     DESCRIPTION
          ptx generates the file output that can be processed with a
          text formatter to produce a permuted index of file input
          (standard input and output default).  It has three phases:
          the first does the permutation, generating one line for each
          keyword in an input line.  The keyword is rotated to the
          front.  The permuted file is then sorted.  Finally, the
          sorted lines are rotated so the keyword comes at the middle
          of each line.  ptx output is in the form:

               .xx "tail" "before-keyword" "keyword-and-after" "head"

          where .xx is assumed to be an nroff or troff macro provided
          by the user, or provided by the mptx(5) macro package.  The
          before-keyword and keyword-and-after fields incorporate as
          much of the line as will fit around the keyword when it is
          printed.  tail and head, at least one of which is always the
          empty string, are wrapped-around pieces small enough to fit
          in the unused space at the opposite end of the line.

          The following flag options may be applied:

          -b break   Use the characters in the break file to separate
                     words.  Tab, newline, and space characters are
                     always used as break characters.

          -f         Fold upper and lower case letters for sorting.

          -g gap     Use the next argument, gap, as the number of
                     characters that ptx will reserve in its
                     calculations for each gap among the four parts of
                     the line as finally printed.  The default gap is
                     3.

          -i ignore  Do not use as keywords any words given in the
                     ignore file.  If the -i and -o flag options are
                     missing, use /usr/lib/eign as the ignore file.

          -o only    Use as keywords only the words given in the only
                     file.

          -r         Take any leading non-blank characters of each
                     input line to be a reference identifier (as to a
                     page or chapter), separate from the text of the
                     line.  Attach that identifier as a 5th field on



     Page 1                                        (last mod. 1/16/87)

     ptx(1)                                                     ptx(1)



                     each output line.

          -t         Prepare troff output.

          -w n       Use the next argument, n, as the length of the
                     output line.  The default line length is 72
                     characters for nroff and 100 for troff.

          The index for this manual was generated using ptx.

     EXAMPLE
          If file1 contains:

               once upon a time
               in the middle of a large
               dark forest

          The command ptx file1

          responds with:

               .xx "" "" "dark forest" ""
               .xx "" "dark" "forest" ""
               .xx "" "in the middle of a" "large" ""
               .xx "" "in the" "middle of a large" ""
               .xx "" "" "once upon a time" ""
               .xx "" "once" "upon a time" ""

     FILES
          /usr/bin/ptx
          /bin/sort
          /usr/lib/eign
          /usr/lib/tmac/tmac.ptx

     SEE ALSO
          nroff(1), troff(1), mm(5), mptx(5),
          ``Other Text Processing Tools'' in Oreo Text Processing
          Tools.

     BUGS
          Line length counts do not account for overstriking or
          proportional spacing.
          Lines that contain tildes () are botched, because ptx uses
          that character internally.











     Page 2                                        (last mod. 1/16/87)