Draft: testcat

#!/bin/ash
outfile=~/testcat.in
#cat_file=/usr/local/petget/categories.dat
cat_file=~/cat_test_cat.dat
. ~/build_cannonical
            awk -v PKGOS="$distro" -v PKGOSVER="$distro_ver" -v CATFILE="$cat_file" -v ALIASFILE="$alias_file" \
'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 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
    }
  }
}'\
"$ALIAS_FNS"\
'BEGIN { init_CANONICAL_ARY() #MAPS a PKG NAME to a canonical name
      }
FILENAME = ALIASFILE {
FILENAME = CATFILE {parse_cats($0)
                    next} #https://www3.physnet.uni-hamburg.de/physnet/Tru64-Unix/HTML/APS32DTE/WKXXXXXX.HTM

/^PACKAGE NAME:/     { sub(/^PACKAGE NAME:\s*/,""); PKGFILE=$0; parse_pkg_name($0); }
/^PACKAGE LOCATION:/     { sub(/^PACKAGE LOCATION:\s*/,""); sub(/^(.\/)?/,""); PKGPATH=$0; }
/^PACKAGE SIZE \(uncompressed\):/ { sub(/^PACKAGE SIZE \(uncompressed\):\s*/,""); sub(/\s/,""); PKGSIZE=$0; }
/^PACKAGE DESCRIPTION:/ { PARSE_DEC_FLAG=1}
/^PACKAGE REQUIRED:/ { sub(/^PACKAGE REQUIRED:\s*/,"");     PKGDEP=fixdepends($0); }
{maybe_parse_dec($0)}
/^$/            {
    if (length(PKG)>1){
    print PKG "_" PKGVER "|" PKG "|" PKGVER "|"  BUILD_NO "|" PKGSECTION "|" PKGSIZE "|" PKGPATH "|" PKGFILE  "|" PKGDEP "|" PKGINFO "|" PKGOS "|" PKGOSVER  "|"
    };

                  PKG=""; PKGVER=""; PKGSECTION="CATEGORY"; PKGSIZE=""; PKGFILE=""; PKGPATH=""; PKGDEP=""; PKGINFO=""; PKGPRIO="";   BUILD_NO="";  ARCH=""; PARSE_DEC_FLAG=0; }
' "$alias_file" "$cat_file"
#cat "$cat_file"