#!/bin/sh

exec awk -f - -- index.html << 'EOF'

BEGIN {
  found = 0
  section = 0
  images = 0
  title = "TITLE"
  header = "DESCRIPTION"
  description = ""
}

/"")/ { found=0 }
/^$/  { next }

found {
    if (substr($1,0,1) == "/") next
    gsub("^ +", "")
    split($0, x, ",")
    fn = substr(x[1], 2, length(x[1])-2)
    i = 4
    if (length(fn) == 0) i = 2
    str = ""; while (i in x) { str = str","x[i]; i++; }
    str = substr(str, 2); gsub("^ +", "", str); str = substr(str, 2, length(str)-3)
    while (index(str, "<") != 0) { j=index(str,"<"); k=index(str,">"); str=substr(str,1,j-1)""substr(str,k+1) }
    if (length(fn) == 0) {
      if (section > 0) print "  </Section>\n";
      section++
      print "  <Section ID='"section"'>\n    <Title>"str"</Title>\n"
      next
    }
    if ((images == 0) && (section == 0)) {
      section++
      print "  <Section ID='"section"'>\n    <Title></Title>\n"
    }

    images++
    width  = x[2]; gsub("^ +", "", width)
    height = x[3]; gsub("^ +", "", height)
    print "    <Image ID='"fn"'>\n      <filename>"fn"</filename>\n      <width>"width"</width>\n      <height>"height"</height>\n      <description>"str"</description>\n    </Image>\n"
  }

/<TITLE>.*<\/TITLE>/ {
  match($0, "<TITLE>.*</TITLE>");
  str=substr($0,RSTART+7,RLENGTH-15);
  while (index(str, "<") != 0) { j=index(str,"<"); k=index(str,">"); str=substr(str,1,j-1)""substr(str,k+1) }
  title=str
}
/<H1>.*<\/H1>/ {
  match($0, "<H1>.*</H1>");
  str=substr($0,RSTART+4,RLENGTH-9);
  while (index(str, "<") != 0) { j=index(str,"<"); k=index(str,">"); str=substr(str,1,j-1)""substr(str,k+1) }
  header=str
}

/ new Array/ {
  print "<Page ID='1'>\n  <Title>"title"</Title>\n  <Header>"header"</Header>\n  <Description>"description"</Description>\n"
  found = 1
}

END {
  print "  </Section>\n</Page>\n";
}

EOF