AWK_gen_prg

#!/bin/gawk -f
function fixdepends(s,   p,a,sout) {
    split(s,a,",")
    for (p in a) {
        sout = sout ",+" a[p]
    }
    sub(/^,/,"",sout) ; return sout;
}
function parse_pkg_name(s,        txz_split,arch_build){
    match(s,/^([^\-]*)(\-.*)(\.([^\-]+))$/,txz_split)

    PKG=txz_split[1]
    split(txz_split[2],ver_arch_build,"-")
    PKGVER=ver_arch_build[2]

    len=length(ver_arch_build)
    if (len == 4){
        BUILD_NO=ver_arch_build[4]
        ARCH=ver_arch_build[3]
    }
    else if(len == 3){
      if (ver_arch_build[len-1] ~ /^[0-9]+$/){
        BUILD_NO=ver_arch_build[2]
       }
       else{
           ARCH=ver_arch_build[3]
      }
    }

}
function maybe_parse_dec(s,        line_split){
  if( PARSE_DEC_FLAG == 1){
    match($0,/(\S+):\s+(.*)$/,line_split)
    if(length(line_split) > 0){
      if (line_split[1] == PKG){
        PKGINFO=substr(line_split[2],1,200)
        PARSE_DEC_FLAG=0
      }
    }
  }
}function cat_fallback(s){
    switch(s){
        case /vcs.*/:
          return "Utility;development"
          break
        case /doc.*/:
          return "Help"
          break
        case /admin.*/:
          return "Setup"
          break
        case /games.*/:
          return "Fun"
          break
        case /.{7,}/:
          return "science"
          break
       default:
          return "BuildingBlock"
          break

    }
}
function get_category(s1,     out,s2,sc1,sc2){
  if ( s1 in pkg_categories){
    return pkg_categories[s1]
  }
  sc1=get_canonical_name(s1)
  if (sc1 in pkg_categories_canonical){
      return pkg_categories_canonical[sc1]
  }
  s2=gensub(/^(.+[^0-9])([0-9]*)$/,"\\1","g",s1)
  if ( s2 in pkg_categories){
    return pkg_categories[s2]
  }
  sc2=get_canonical_name(s2)
  if (sc2 in pkg_categories_canonical){
      return pkg_categories_canonica[sc2]
  }
  #TODO: consider also trying to look up category based on the full name.
}
function parse_cats(s){
  i=index(s,"=")
  if (i>0){
    #catagory=( package1 package2 ... packageN)
    category=substr(s,i-1)
    sub(/PKGCAT_/,"",category) #e.g. category=PKGCAT_Utility => category=Utility
    sub(/_/,";",category) #e.g. category=Filesystem_filemanager => category=Filesystem;filemanager

    #e.g catagory=( package1 package2 ... packageN) => package1 package2 ... packageN
    gensub(/^([^=]+="\s)(.*)(\s*"\s*)$/,"\\2","g",s)

    split(s,pkg_names" ")
    for (pkg in pkg_names){
      pkg_categories[pkg]=category
      pkg_categories_canonical=get_canonical_name(category)
    }
  }
}function get_canonical_name(s){
  if (s in CANONICAL_ARY){
   return CANONICAL_ARY[s]
  } else {
    switch(s){
      case /gtk\+2.*/:
        return "gtk+"
      case /libdbus-glib.*/:
        return "dbus"
      case /libdbus.*/:
        return "dbus"
      case /dbus.*/:
        return "dbus"
      case /mesa-common.*/:
        return "mesa"
      case /libgl1-mesa.*/:
        return "mesa"
      case /mesa_.*/:
        return "mesa"
      case /samba_.*/:
        return "samba"
      case /libgudev.*/:
        return "udev"
      case /libudev.*/:
        return "udev"
      case /udev_.*/:
        return "udev"
      case /perlapi.*/:
        return "perl_tiny"
    }
  }
  return s
}
function init_CANONICAL_ARY(){
  CANONICAL_ARY["perl_tiny"]="perl_tiny"
  CANONICAL_ARY["perl-base"]="perl_tiny"
  CANONICAL_ARY["samba-tng"]="samba"
  CANONICAL_ARY["sane-backends"]="sane"
  CANONICAL_ARY["gtkdialog3"]="gtkdialog"
  CANONICAL_ARY["xdg_puppy"]="xdg_puppy"
  CANONICAL_ARY["mountcifs"]="samba"
  CANONICAL_ARY["mesa"]="mesa"
  CANONICAL_ARY["urxvt-unicode"]="rxvt-unicode"
  CANONICAL_ARY["gtkdialog"]="gtkdialog"
  CANONICAL_ARY["perl-modules"]="perl_tiny"
  CANONICAL_ARY["urxvt"]="rxvt-unicode"
  CANONICAL_ARY["rxvt-unicode"]="rxvt-unicode"
  CANONICAL_ARY["xdg-utils"]="xdg_puppy"
  CANONICAL_ARY["gtk+"]="gtk+"
  CANONICAL_ARY["samba"]="samba"
  CANONICAL_ARY["sane"]="sane"
  CANONICAL_ARY["udev"]="udev"
}BEGIN { init_CANONICAL_ARY() #MAPS a PKG NAME to a canonical name
      }
#FILENAME = ALIASFILE { }