#!/usr/bin/env python2.4
import os
import cgi
from PyLucene import QueryParser, IndexSearcher, StandardAnalyzer, FSDirectory
class MetaSearch(object):
def asXML(self, hits, maxhits):
'''
Output result from metasearch as XML
'''
out = "\t\n"
for i in xrange(min(hits.length(), maxhits)):
doc = hits.doc(i)
out += "\t\t\n"
out += "\t"
return out
def escape(self, text):
'''
Escape XML/HTML entities and convert output to utf-8
'''
return (text.replace(u'&', u'&').replace(u'<', u'<').replace(u'>', u'>')).encode('utf-8', 'replace')
if __name__ == "__main__":
maxhits = 50
input = cgi.parse_qs(os.environ["QUERY_STRING"])
if input.get("max", [""])[0].isdigit():
limit = int(input.get("max", [25])[0])
else:
limit = 25
if limit <= 0:
limit = 1
if limit > maxhits:
limit = maxhits
search = input.get("query", [""])[0]
if search == "":
print "Content-type: text/html\n"
print ""
print ""
print ""
print "Error"
print ""
print ""
print "Hi there, apparently you're new to this :)
"
print "You'll have to specify the parameters, the valid ones are:
"
print ""
print "- max (results returned, 1 to 50)
"
print "- query (your lucene formated query)
"
print "
"
print "Valid fields in the document:
"
print ""
print "- content (default field, contains what you need ;)
"
print "- albumid
"
print "- artistid
"
print "- trackid
"
print "- tracklengthsec
"
print "- tracknum
"
print "
"
print "Brief example:
"
print "http://mb.samfundet.no/?query=europe countdown
"
print ""
print ""
else:
directory = FSDirectory.getDirectory("index", False)
searcher = IndexSearcher(directory)
analyzer = StandardAnalyzer([])
query = QueryParser("content", analyzer).parse(search)
hits = searcher.search(query)
ms = MetaSearch()
print "Content-type: text/xml\n"
print "\n"
print ""
print ms.asXML(hits, limit)
print ""
searcher.close()