#!/usr/bin/mawk -We
# *********************************************************************
#  Written by and copyright Carlo Strozzi <carlos@linux.it>.
# 
#  compute: applies arbitrary AWK expressions to table data, using
#	    column names.
#  Copyright (C) 2000-2001 Carlo Strozzi <carlos@linux.it>
#
#  This program is free software; you can redistribute it and/or modify
#  it under the terms of the GNU General Public License as published by
#  the Free Software Foundation; either version 2 of the License, or
#  (at your option) any later version.
#
#  This program is distributed in the hope that it will be useful,
#  but WITHOUT ANY WARRANTY; without even the implied warranty of
#  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
#  GNU General Public License for more details.
#  
#  You should have received a copy of the GNU General Public License
#  along with this program; if not, write to the Free Software
#  Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
#
#  2001-01-03 Ported to NoSQL v3
#  2001-02-02 Fixed a bug in parser() 
#  2001-04-17 Added inline help
#  2001-08-17 Added stdio portability
#
#  $Id$
# *********************************************************************

BEGIN {
  NULL = ""; FS = OFS = "\t"; srand()

  while (ARGV[++i] != NULL) {
    if (ARGV[i] == "-x" || ARGV[i] == "--debug") debug = 1
    else if (ARGV[i] == "-N" || ARGV[i] == "--no-header") no_hdr = 1
    else if (ARGV[i] == "-i" || ARGV[i] == "--input") i_file = ARGV[++i]
    else if (ARGV[i] == "-o" || ARGV[i] == "--output") o_file = ARGV[++i]
    else if (ARGV[i] == "-h" || ARGV[i] == "--help") {
       system("grep -v '^#' @NOSQLPATH@/nosql/help/compute.txt")
       rc = 1
       exit(rc)
    }
    else awk_program = awk_program " " ARGV[i]
  }
  if (awk_program == NULL)  exit

  ARGC = 1					# Fix argv[]

  if (o_file == NULL) o_file = "@STDOUT@"
  if (i_file != NULL) { ARGV[1] = i_file; ARGC = 2 }
}

# Table header
r == 0 {
  # Start building the back-end awk program.

  awkpgm = "'BEGIN{FS=OFS=\"\\t\";}{"

  # Do not let duplicated input columns fool us.

  for (col = 1; col <= NF; col++) {
    if (col_names[$(col)] == NULL) {
      col_names[$(col)] = $(col)
      awk_col[$(col)] = "$(" col ")"
    }
  }

  awkpgm = awkpgm parser(awk_program) " } {print}'"

  unix_cmd = "mawk " awkpgm

  # Print header.
  if (!no_hdr) {
    print > o_file ; gsub(/[^\t]/, "-"); print > o_file

    # Make sure we print the header before calling mawk(1) again.
    fflush()

    # Close output file before the back-end mawk(1) appends to it.
    if (o_file != "@STDOUT@") close(o_file)
  }

  r++ ; next
}

# Dashline.
r == 1 {
  # Let's save one concurrent process and a couple of msec.
  unix_cmd = "exec " unix_cmd
  if (o_file != "@STDOUT@") unix_cmd = unix_cmd " >> " o_file

  if (debug) {
    print "\n" awk_program "\n" unix_cmd "\n" > "@STDERR@"
  }

  r++ ; next
}

# Table body.
{ print | unix_cmd }

END {
   if (rc) exit(rc)
   exit(close(unix_cmd))	# Return back-end mawk(1) exit status.
}

#
# Function section.
#

function parser(in_pgm,		i, var_name, col_pos, p_length, \
				a, quoted, escaped, slashed, \
				tilde, out_pgm) {

  p_length = split(in_pgm, a, "")

  # The next statement is necessary, to make sure the last section 
  # of the parser is entered once even after the end of the input AWK
  # program.

  p_length++

  while (++i <= p_length) {

    # Do not remove the next two lines of code. They can be *very*
    # useful to debug the parser. Simply comment them out when
    # not in use.
    #
    # "q:" stands for "quoted", "e:" for "escaped, etc.
    #
    #print "q:" quoted " e:" escaped " t:" tilde " s:" slashed \
    #    " a:" a[i] > "@STDERR@"

    # The "/" character triggers regexp matching only after "~",
    # otherwise it stands for the divide operator.
    if (a[i] == "~" && !escaped && !quoted && !slashed) tilde = 1

    if (a[i] == "\\" && !escaped) {
      escaped = 1; out_pgm = out_pgm a[i] ; continue
    }
    if (a[i] == "/" && !escaped && !quoted && tilde) {
      if (slashed) tilde = slashed = 0
      else slashed = 1
    }
    if (a[i] == "\"" && !escaped && !slashed) {
      quoted ? quoted = 0 : quoted = 1
    }
    escaped = 0
    if (quoted || slashed) { out_pgm = out_pgm a[i] ; continue }
    if (a[i] ~ /[A-Za-z_]/) { var_name = var_name a[i] ; continue }
    if (var_name != NULL) {
      if (a[i] ~ /[0-9]/) { var_name = var_name a[i] ; continue }
      col_pos = awk_col[var_name]
      if (col_pos != NULL) { out_pgm = out_pgm col_pos }
      else out_pgm = out_pgm var_name
      var_name = NULL
    }
    out_pgm = out_pgm a[i]
  }
  gsub(/'/, "\\047", out_pgm)		# Escape single quotes for sh(1).
  return out_pgm
}

#
# End of program.
#
