# Search entire Wikipedia database 
#  Download Wikipedia (English): https://en.wikipedia.org/wiki/Wikipedia:Database_download#English-language_Wikipedia
#  Install mawk. Binary for Cygwin: http://artfiles.org/cygwin.org/pub/cygwinports/x86/release/mawk/
#

BEGIN {

  r = "/f/t/wikipedia-dump/enwiki-20150515-pages-articles.xml"

  RS=("<page|</page>")

  while ((getline rawstr < r ) > 0) {

    gsub(/&lt;/,"<",rawstr);gsub(/&gt;/,">",rawstr);gsub(/&quot;/,"\"",rawstr);gsub(/&amp;/,"\\&",rawstr)
    
    # Article title
    if ( match(rawstr, "<title>.+</title>") ) {
      a = substr(rawstr, RSTART, RLENGTH)
      split(a, b, "(<title>|</title>)")
      title = b[2]
    }

    # Article body
    if ( match(rawstr, "<text xml:space=\"preserve\">.+</text>") ) {
      a = substr(rawstr, RSTART, RLENGTH)
      split(a, b, "(<text xml:space=\"preserve\">|</text>)")
      document = b[2]
    }

# ---------- Find all cases of Books and Writers at the Kuusankoski Public Library in Finland -----

    if ( match(document, "sportsillustrated.cnn.com") ) {
      print title
      #system("") # flush buffer
      continue
    }

  }
  close(r)
}