def main(args):
# search corpus index
db = datastore.corpus_db()
sconn = xappy.SearchConnection(settings.XAPIAN_DB)
query = ' '.join(args.query)
print "Search {} documents for '{}'".format(
sconn.get_doccount(), args.query
)
q = sconn.query_parse(query, default_op=sconn.OP_AND)
if args.category:
qc = q.compose(q.OP_OR, [
sconn.query_field('category', c) for c in args.category
])
q = q & qc
if args.date:
qd = q.compose(q.OP_OR, [
sconn.query_field('date', d) for d in args.date
])
q = q & qd
if args.date_start and args.date_end:
qr = sconn.query_range('date', args.date_start, args.date_end)
q = q.filter(qr)
if args.sort:
sortby = [tuple(args.sort.split(','))]
else:
sortby = None
print 'Query: {!r}'.format(q)
results = execute_query(sconn, q, args.offset, args.limit,
getfacets=args.facet,
allowfacets=('category',),
sortby=sortby)
if results.estimate_is_exact:
print "Found {} results".format(results.matches_estimated)
else:
print "Found approximately {} results".format(results.matches_estimated)
for i, result in enumerate(results, 1):
doc = db[result.id]
try:
cat = result.get_terms('category').next()
except StopIteration:
cat = 'none'
try:
date = result.get_terms('date').next()
except StopIteration:
date = 'none'
print "{:2}. {} -- {} -- {}\n\t{}\n\t{}\n".format(
i, cat, doc['headline'], date, doc['url'], result.id)
from IPython import embed; embed()
def _query(self, sconn, q, offset, limit):
if q:
q = sconn.query_parse(q)
else:
q = sconn.query_all()
categories = self.get_args('category')
if categories:
qc = sconn.query_composite(
sconn.OP_OR,
[sconn.query_field('category', value) for value in categories])
#q = sconn.query_composite(
# sconn.OP_AND, [q, qc])
q = q.filter(qc)
date = self.get_arg('date')
if date:
qd = sconn.query_field('date', date)
q = q.filter(qd)
sortby = self.get_arg('sortby')
if sortby:
sortby = [sortby]
else:
sortby = None
self.log('Query: {!r}'.format(q))
results = execute_query(sconn, q, offset, limit, sortby=sortby)
self.success({
'total':
results.matches_estimated,
'is_exact':
results.estimate_is_exact,
'docs': [
dict(id=r.id, category=r.get_terms('category').next())
for r in results
]
})
def _query(self, sconn, q, offset, limit):
if q:
q = sconn.query_parse(q)
else:
q = sconn.query_all()
categories = self.get_args('category')
if categories:
qc = sconn.query_composite(
sconn.OP_OR, [
sconn.query_field('category', value)
for value in categories
])
#q = sconn.query_composite(
# sconn.OP_AND, [q, qc])
q = q.filter(qc)
date = self.get_arg('date')
if date:
qd = sconn.query_field('date', date)
q = q.filter(qd)
sortby = self.get_arg('sortby')
if sortby:
sortby = [sortby]
else:
sortby = None
self.log('Query: {!r}'.format(q))
results = execute_query(sconn, q, offset, limit, sortby=sortby)
self.success({
'total': results.matches_estimated,
'is_exact': results.estimate_is_exact,
'docs': [
dict(id=r.id, category=r.get_terms('category').next()) for r in results
]
})
def main(args):
# search corpus index
db = datastore.corpus_db()
sconn = xappy.SearchConnection(settings.XAPIAN_DB)
query = ' '.join(args.query)
print "Search {} documents for '{}'".format(sconn.get_doccount(),
args.query)
q = sconn.query_parse(query, default_op=sconn.OP_AND)
if args.category:
qc = q.compose(
q.OP_OR, [sconn.query_field('category', c) for c in args.category])
q = q & qc
if args.date:
qd = q.compose(q.OP_OR,
[sconn.query_field('date', d) for d in args.date])
q = q & qd
if args.date_start and args.date_end:
qr = sconn.query_range('date', args.date_start, args.date_end)
q = q.filter(qr)
if args.sort:
sortby = [tuple(args.sort.split(','))]
else:
sortby = None
print 'Query: {!r}'.format(q)
results = execute_query(sconn,
q,
args.offset,
args.limit,
getfacets=args.facet,
allowfacets=('category', ),
sortby=sortby)
if results.estimate_is_exact:
print "Found {} results".format(results.matches_estimated)
else:
print "Found approximately {} results".format(
results.matches_estimated)
for i, result in enumerate(results, 1):
doc = db[result.id]
try:
cat = result.get_terms('category').next()
except StopIteration:
cat = 'none'
try:
date = result.get_terms('date').next()
except StopIteration:
date = 'none'
print "{:2}. {} -- {} -- {}\n\t{}\n\t{}\n".format(
i, cat, doc['headline'], date, doc['url'], result.id)
from IPython import embed
embed()
