def search(dbpath, querystring, offset=0, pagesize=10):
# offset - defines starting point within result set
# pagesize - defines number of records to retrieve
# Open the database we're going to search.
db = xapian.Database(dbpath)
# Set up a QueryParser with a stemmer and suitable prefixes
queryparser = xapian.QueryParser()
queryparser.set_stemmer(xapian.Stem("en"))
queryparser.set_stem_strategy(queryparser.STEM_SOME)
queryparser.add_prefix("title", "S")
queryparser.add_prefix("description", "XD")
# And parse the query
query = queryparser.parse_query(querystring)
# Use an Enquire object on the database to run the query
enquire = xapian.Enquire(db)
enquire.set_query(query)
# Start of example code.
enquire.set_sort_by_value_then_relevance(1, False)
# End of example code.
# And print out something about each match
matches = []
for index, match in enumerate(enquire.get_mset(offset, pagesize)):
fields = json.loads(match.document.get_data())
print u"%(rank)i: #%(docid)3.3i %(name)s %(date)s\n Population %(pop)s" % {
'rank': offset + index + 1,
'docid': match.docid,
'name': fields.get('name', u''),
'date': fields.get('admitted', u''),
'pop': fields.get('population', u''),
'lat': fields.get('latitude', u''),
'lon': fields.get('longitude', u''),
}
matches.append(match.docid)
# Finally, make sure we log the query and displayed results
support.log_matches(querystring, offset, pagesize, matches)
def search(dbpath, querystring, offset=0, pagesize=10):
# offset - defines starting point within result set
# pagesize - defines number of records to retrieve
# Open the database we're going to search.
db = xapian.Database(dbpath)
# Set up a QueryParser with a stemmer and suitable prefixes
queryparser = xapian.QueryParser()
queryparser.set_stemmer(xapian.Stem("en"))
queryparser.set_stem_strategy(queryparser.STEM_SOME)
queryparser.add_prefix("title", "S")
queryparser.add_prefix("description", "XD")
# and add in value range processors
queryparser.add_valuerangeprocessor(
xapian.NumberValueRangeProcessor(0, 'mm', False))
queryparser.add_valuerangeprocessor(xapian.NumberValueRangeProcessor(
1, ''))
# And parse the query
query = queryparser.parse_query(querystring)
# Use an Enquire object on the database to run the query
enquire = xapian.Enquire(db)
enquire.set_query(query)
# And print out something about each match
matches = []
for match in enquire.get_mset(offset, pagesize):
fields = json.loads(match.document.get_data())
print u"%(rank)i: #%(docid)3.3i (%(date)s) %(measurements)s\n %(title)s" % {
'rank': match.rank + 1,
'docid': match.docid,
'measurements': fields.get('MEASUREMENTS', u''),
'date': fields.get('DATE_MADE', u''),
'title': fields.get('TITLE', u''),
}
matches.append(match.docid)
# Finally, make sure we log the query and displayed results
support.log_matches(querystring, offset, pagesize, matches)
def reload_database(self): # {{{
# create the xapian handlers
self.db = xapian.WritableDatabase(self.database,
xapian.DB_CREATE_OR_OPEN)
self.qp = xapian.QueryParser()
self.qp.set_stemmer(xapian.Stem(self.language))
self.qp.set_stemming_strategy(self.qp.STEM_SOME)
self.qp.add_prefix("title", "S")
self.tg = xapian.TermGenerator()
self.tg.set_stemmer(xapian.Stem(self.language))
try:
self.tg.set_stemming_strategy(self.tg.STEM_SOME)
except AttributeError:
pass
self.e = xapian.Enquire(self.db)
self.sorted_e = xapian.Enquire(self.db)
# Value 2 is the lowercase form of the title
self.sorted_e.set_sort_by_value(2, False)
def _search(self, text, pagenum=1, limit=10, isPath=False):
database = self._index.index
# Start an enquire session.
enquire = xapian.Enquire(database)
# Parse the query string to produce a Xapian::Query object.
self.parser = xapian.QueryParser()
self.parser.set_stemmer(xapian.Stem("english"))
self.parser.set_database(database)
self.parser.set_stemming_strategy(xapian.QueryParser.STEM_SOME)
self.query = self.parser.parse_query(text, DEFAULT_SEARCH_FLAGS)
# find using the parsed query
enquire.set_query(self.query)
offset = pagenum * limit - limit
matches = enquire.get_mset(offset, limit)
# build results
results = XapianResults(self,
matches,
total_count=matches.get_matches_estimated(),
pagenum=pagenum,
limit=limit)
return results
def __init__(self, dbpath='/home/mirage/Downloads/weiboxa/simple'):
database = xapian.Database(dbpath)
enquire = xapian.Enquire(database)
qp = xapian.QueryParser()
stemmer = xapian.Stem("english")
qp.set_stemmer(stemmer)
qp.set_database(database)
qp.set_stemming_strategy(xapian.QueryParser.STEM_SOME)
self.qp = qp
self.enquire = enquire
self.emotionvi = 0
self.keywordsvi = 1
self.timestampvi = 2
self.loctvi = 3
self.reploctvi = 4
self.emotiononlyvi = 5
#usernamevi = 6
self.hashtagsvi = 7
#uidvi = 8
#repnameslistvi = 9
self.maxitems = 1000000000
def enquire(querystring, versions=None):
try:
db = xapian.Database(settings.XAPIAN_DB_PATH)
except xapian.DatabaseOpeningError:
return None
qp = xapian.QueryParser()
qp.set_stemming_strategy(qp.STEM_SOME)
qp.set_stemmer(xapian.Stem("en"))
qp.set_database(db)
query = qp.parse_query(querystring, qp.FLAG_PARTIAL)
if versions:
query = xapian.Query(xapian.Query.OP_FILTER,
query,
make_version_queries(versions))
enquiry = xapian.Enquire(db)
enquiry.set_query(query)
return db, enquiry
def search(dbpath, querystring, option=0, offset=0, pagesize=10):
# offset - defines starting point within result set
# pagesize - defines number of records to retrive
db = xapian.Database(dbpath)
queryparser = xapian.QueryParser()
# choose a language
queryparser.set_stemmer(xapian.Stem("en"))
queryparser.set_stemming_strategy(queryparser.STEM_SOME)
queryparser.add_prefix("title", "S")
queryparser.add_prefix("description", "XD")
query = queryparser.parse_query(querystring)
enquire = xapian.Enquire(db)
#select different weighting schema
bm = select_weight(option)
enquire.set_weighting_scheme(bm)
enquire.set_query(query)
matches = []
ret = ""
for match in enquire.get_mset(offset, pagesize):
fields = json.loads(match.document.get_data())
tmp = u"%(rank)i: #%(docid)3.3i %(title)s" % {
'rank': match.rank + 1,
'docid': match.docid,
'title': fields.get('TITLE', u''),
}
ret += tmp
ret += '\n'
matches.append(match.docid)
support.log_matches(querystring, offset, pagesize, matches)
return ret
### END of function
def xapian_get_ranked_index(index, pattern, params, hitset,
ranked_result_amount):
"""
Queries a Xapian index.
Returns: a list of ranked record ids [(recid, score), ...) contained in hitset
and an intbitset of record ids contained in hitset.
"""
result = []
matched_recs = intbitset()
database = DATABASES[index]
enquire = xapian.Enquire(database)
qp = xapian.QueryParser()
stemmer = xapian.Stem("english")
qp.set_stemmer(stemmer)
qp.set_database(database)
qp.set_stemming_strategy(xapian.QueryParser.STEM_SOME)
# Avoids phrase search to increase performance
if "avoid_phrase_search_threshold" in params and len(hitset) >= params[
"avoid_phrase_search_threshold"] and pattern.startswith('"'):
pattern = pattern[1:-1]
query_string = ' AND '.join(pattern.split(' '))
pattern = qp.parse_query(query_string)
else:
query_string = pattern
pattern = qp.parse_query(query_string, xapian.QueryParser.FLAG_PHRASE)
enquire.set_query(pattern)
matches = enquire.get_mset(0, ranked_result_amount, None,
MatchDecider(hitset))
weight = params["weight"]
for match in matches:
recid = match.docid
if recid in hitset:
score = int(match.percent) * weight
result.append((recid, score))
matched_recs.add(recid)
return (result, matched_recs)
def search(query, active_element, numresults=10):
# Put active_element at end of query
qfields = sorted(query, key=lambda k: k == active_element)
# XXX There should be a way to do this without going through an
# intermediate string, and without adding prefixes.
qvalues = [(k, e) for k in qfields for e in query[k].split()]
qstring = ['%s:%s' % (field, value) for field, value in qvalues]
querystring = ' AND '.join(qstring)
db = xapian.Database(dbpath)
queryparser = xapian.QueryParser()
queryparser.set_database(db)
for field, abbrev in fields.items():
queryparser.add_prefix(field, abbrev)
query = queryparser.parse_query(querystring,
queryparser.FLAG_BOOLEAN |
queryparser.FLAG_PARTIAL |
queryparser.FLAG_WILDCARD)
enquire = xapian.Enquire(db)
enquire.set_weighting_scheme(xapian.BoolWeight())
enquire.set_query(query)
return [json.loads(r.document.get_data())
for r in enquire.get_mset(0, numresults)]
def query_parser(self, database=None):
"""Return a xapian.QueryParser object set up for this fieldmap.
`database` should be supplied if you want to use spelling correction etc.
"""
qp = xapian.QueryParser()
if database:
qp.set_database(database)
if self.language:
qp.set_stemmer(xapian.Stem(self.language))
# add fieldname prefixes
for name, (prefix, valnum, isfilter) in self._fieldmap.iteritems():
if isfilter:
qp.add_boolean_prefix(name, prefix)
else:
qp.add_prefix(name, prefix)
return qp
def process(self, queryString, content):
# Parse query
queryParser = xapian.QueryParser()
queryParser.set_stemmer(xapian.Stem('english'))
queryParser.set_stemming_strategy(xapian.QueryParser.STEM_SOME)
query = queryParser.parse_query(queryString)
# Parse content after replacing non-alphanumeric characters with spaces
queryParser.parse_query(re.sub('\W', ' ', content).lower())
# Create search pattern
documentTerms = sum(
[list(queryParser.unstemlist(x)) for x in set(query)], [])
if not documentTerms:
documentTerms = set(query)
pattern = re.compile(
r'\b(%s)\b' % '|'.join(re.escape(x) for x in documentTerms),
re.IGNORECASE)
# If the user does not want to extract text or there is no queryString,
if not self.extractLength or not queryString:
extract = content
# If the user wants to extract text and there is a queryString,
else:
# Initialize
extractIntervals = []
extractLengthHalved = self.extractLength / 2
# For each matchInterval,
for match in pattern.finditer(content):
# Prepare
mStart = max(0, match.start() - extractLengthHalved)
mEnd = min(len(content), match.end() + extractLengthHalved)
# Absorb it
absorbInterval((mStart, mEnd), extractIntervals)
# Load extracts
extract = self.joinText.join(content[eStart:eEnd].strip()
for eStart, eEnd in extractIntervals)
# If the user wants to highlight relevant terms and there is a queryString,
if self.highlightTemplate and queryString:
extract = pattern.sub(self.highlightTemplate % r'\1', extract)
# Return
return extract
def _create_query_for_string(self, text, require_all=True, analyzer=None):
"""generate a query for a plain term of a string query
basically this function parses the string and returns the resulting
query
@param text: the query string
@type text: str
@param require_all: boolean operator
(True -> AND (default) / False -> OR)
@type require_all: bool
@param analyzer: Define query options (partial matching, exact matching,
tokenizing, ...) as bitwise combinations of
CommonIndexer.ANALYZER_???.
This can override previously defined field analyzer settings.
If analyzer is None (default), then the configured analyzer for the
field is used.
@type analyzer: int
@return: resulting query object
@rtype: xapian.Query
"""
qp = xapian.QueryParser()
qp.set_database(self.reader)
if require_all:
qp.set_default_op(xapian.Query.OP_AND)
else:
qp.set_default_op(xapian.Query.OP_OR)
if analyzer is None:
analyzer = self.analyzer
if analyzer & self.ANALYZER_PARTIAL > 0:
match_flags = xapian.QueryParser.FLAG_PARTIAL
return qp.parse_query(text, match_flags)
elif analyzer == self.ANALYZER_EXACT:
# exact matching -
return xapian.Query(text)
else:
# everything else (not partial and not exact)
match_flags = 0
return qp.parse_query(text, match_flags)
def _create_query_for_field(self, field, value, analyzer=None):
"""generate a field query
this functions creates a field->value query
:param field: the fieldname to be used
:type field: str
:param value: the wanted value of the field
:type value: str
:param analyzer: Define query options (partial matching, exact
matching, tokenizing, ...) as bitwise combinations of
*CommonIndexer.ANALYZER_???*.
This can override previously defined field
analyzer settings.
If analyzer is None (default), then the configured
analyzer for the field is used.
:type analyzer: int
:return: the resulting query object
:rtype: xapian.Query
"""
if analyzer is None:
analyzer = self.analyzer
if analyzer == self.ANALYZER_EXACT:
# exact matching -> keep special characters
return xapian.Query("%s%s" % (field.upper(), value))
# other queries need a parser object
qp = xapian.QueryParser()
qp.set_database(self.reader)
if (analyzer & self.ANALYZER_PARTIAL > 0):
# partial matching
match_flags = xapian.QueryParser.FLAG_PARTIAL
return qp.parse_query(value, match_flags, field.upper())
else:
# everything else (not partial and not exact)
match_flags = 0
return qp.parse_query(value, match_flags, field.upper())
def search(dbpath, querystring, offset=0, pagesize=10):
# offset - defines starting point within result set
# pagesize - defines number of records to retrieve
# Open the database we're going to search.
db = xapian.Database(dbpath)
# Set up a QueryParser with a stemmer and suitable prefixes
queryparser = xapian.QueryParser()
queryparser.set_stemmer(xapian.Stem("en"))
queryparser.set_stemming_strategy(queryparser.STEM_SOME)
# Start of prefix configuration.
queryparser.add_prefix("title", "S")
queryparser.add_prefix("description", "XD")
# End of prefix configuration.
# And parse the query
query = queryparser.parse_query(querystring)
# Use an Enquire object on the database to run the query
enquire = xapian.Enquire(db)
enquire.set_query(query)
# And print out something about each match
matches = []
for match in enquire.get_mset(offset, pagesize):
fields = json.loads(match.document.get_data())
print(
u"%(rank)i: #%(docid)3.3i %(title)s" % {
'rank': match.rank + 1,
'docid': match.docid,
'title': fields.get('TITLE', u''),
})
matches.append(match.docid)
# Finally, make sure we log the query and displayed results
# support.log_matches(querystring, offset, pagesize, matches)
print(querystring, offset, pagesize, matches)
def find(database, query, sortfield=None, reverse=False, maxnum=100):
"""sort_direction is False for ascending order, True for descending order.
"""
# Set up a QueryParser with a stemmer and suitable prefixes
queryparser = xapian.QueryParser()
queryparser.set_stemmer(xapian.Stem("en"))
queryparser.set_stemming_strategy(queryparser.STEM_SOME)
# Add in the termprefixes that are in use
for tp in iota.TERMPREFIXES:
queryparser.add_prefix(*tp)
# And parse the query
xquery = queryparser.parse_query(query)
# Use an Enquire object on the database to run the query
enquire = xapian.Enquire(database)
enquire.set_query(xquery)
# Determine how we're going to sort the results.
if sortfield is not None:
try:
slot = iota.SLOTS[sortfield]
except KeyError:
pass
else:
enquire.set_sort_by_value_then_relevance(slot, not reverse)
# And print out something about each match
logging.info("execute query: {}".format(query))
mset = enquire.get_mset(0, maxnum)
# make sexps
sexps = [match_as_sexp(match) for match in mset]
sexps.append('(:found {})\n'.format(len(mset)))
return sexps
def searchDB(queryStr, withContent=False, extractLength=32):
# Parse Query
queryParser = xapian.QueryParser()
queryParser.set_stemmer(xapian.Stem('english'))
queryParser.set_database(database)
queryParser.set_stemming_strategy(xapian.QueryParser.STEM_SOME)
query = queryParser.parse_query(queryStr)
offset, limit = 0, 5
# Start Query Session
enquire = xapian.Enquire(database)
enquire.set_query(query)
docIds = list()
# Display Matches
matches = enquire.get_mset(offset, limit)
print('*' * 50)
for match in matches:
print('-' * 50)
pmId = match.document.get_value(0)
docIds.append(match.document.get_value(0))
print 'Rank/ID: %s, docID: %s' % (match.rank, pmId)
print('-' * 50)
# Process
content = match.document.get_data()
extract = TextMachine(extractLength, '*%s*').process(queryStr, content)
print extract.replace('\n', ' ')
print('-' * 50)
print('*' * 50)
print 'No. of Docs matching Query: %s' % matches.get_matches_estimated()
print 'No. of Docs Returned: %s' % matches.size()
return docIds
def test_value_range_processor():
vp = MultipleValueRangeProcessor(dict(foo=1, bar=2), str.upper)
assert vp('foo:abc', 'def') == (1, 'ABC', 'DEF')
assert vp('bar:news', 'def') == (2, 'NEWS', 'DEF')
assert vp('bar:', 'def') == (2, '', 'DEF')
assert vp('bar', 'def') == (xapian.BAD_VALUENO, 'bar', 'def')
assert vp('baz:foo', 'def') == (xapian.BAD_VALUENO, 'baz:foo', 'def')
qp = xapian.QueryParser()
db = xodb.temp()
qp.set_database(db.backend)
qp.add_valuerangeprocessor(vp)
query = qp.parse_query('foo:abc..def')
assert str(query) == 'Xapian::Query(VALUE_RANGE 1 ABC DEF)'
query = qp.parse_query('bar:abc..def')
assert str(query) == 'Xapian::Query(VALUE_RANGE 2 ABC DEF)'
query = qp.parse_query('bar:3..4')
assert str(query) == 'Xapian::Query(VALUE_RANGE 2 3 4)'
assert_raises(xapian.QueryParserError, qp.parse_query, 'baz:abc..def')
def handle_query(self, q):
database = xapian.Database(self.db_path)
enquire = xapian.Enquire(database)
qp = xapian.QueryParser()
stemmer = xapian.Stem("english")
qp.set_stemmer(stemmer)
qp.set_database(database)
qp.set_stemming_strategy(xapian.QueryParser.STEM_SOME)
query = qp.parse_query(q)
# Find the top 100 results for the query.
enquire.set_query(query)
matches = enquire.get_mset(0, 100)
results = []
for m in matches:
context = self.extract_context(m.document.get_data())
results.append(
(m.document.get_value(self.DOC_PATH),
m.document.get_value(self.DOC_TITLE), ''.join(context)))
return results
def search(self, server_guid, store_guid, folder_ids, fields_terms, query, limit_results, log):
""" handle query; see links in the top for a description of the Xapian API """
db = self.open_db(server_guid, store_guid, log=log)
if not db:
return []
qp = xapian.QueryParser()
qp.add_prefix("sourcekey", "XK:")
qp.add_prefix("folderid", "XF:")
for fields, terms in fields_terms:
for field in fields:
qp.add_prefix('mapi%d' % field, "XM%d:" % field)
log.info('performing query: %s', query)
qp.set_database(db)
query = qp.parse_query(query, xapian.QueryParser.FLAG_BOOLEAN|xapian.QueryParser.FLAG_PHRASE|xapian.QueryParser.FLAG_WILDCARD)
enquire = xapian.Enquire(db)
enquire.set_query(query)
matches = []
for match in enquire.get_mset(0, limit_results or db.get_doccount()): # XXX catch exception if database is being updated?
matches.append(match.document.get_value(0).decode('ascii'))
db.close()
return matches
def get_latest_builds(self, package_name):
enquire = xapian.Enquire(self._versionmap_db)
qp = xapian.QueryParser()
qp.set_database(self._versionmap_db)
qp.add_boolean_prefix('key', 'XA')
query = qp.parse_query('key:%s' % utils.filter_search_string(package_name))
enquire.set_query(query)
matches = enquire.get_mset(0, 1)
if len(matches) == 0:
return None
results = json.loads(matches[0].document.get_data())
latest_builds = OrderedDict()
lastdistname = ""
for dist in distmappings.tags:
distname = dist['name']
if lastdistname != distname and distname in results:
latest_builds[distname] = results[distname]
lastdistname = distname
return latest_builds
def search(queryString, byDate=False, ownerID=None, extractLength=32):
# Parse query string
queryParser = xapian.QueryParser()
queryParser.set_stemmer(xapian.Stem('english'))
queryParser.set_database(database)
queryParser.set_stemming_strategy(xapian.QueryParser.STEM_SOME)
query = queryParser.parse_query(queryString)
# Set offset and limit for pagination
offset, limit = 0, database.get_doccount()
# Start query session
enquire = xapian.Enquire(database)
enquire.set_query(query)
# Sort by date
if byDate:
enquire.set_sort_by_value(xapian_when)
if ownerID == None:
matches = enquire.get_mset(offset, limit)
else:
# Filter by ownerID
matches = enquire.get_mset(offset, limit, None, MatchDecider(ownerID))
# Display matches
for match in matches:
# Load
documentWhen = match.document.get_value(xapian_when)
# documentOwnerID = int(xapian.sortable_unserialise(match.document.get_value(xapian_owner_id)))
# Display
print '==================='
print 'rank=%s, documentID=%s' % (match.rank, match.docid)
# Process
content = match.document.get_data()
extract = TextMachine(extractLength,
'*%s*').process(queryString, content)
print extract.replace('\n', ' ')
print '==================='
print 'Number of documents matching query: %s' % matches.get_matches_estimated(
)
print 'Number of documents returned: %s' % matches.size()
def crossquery(querystring, lgs):
url = 'http://www.galoes.org/grammars/%s/%s'
width = 0
#querystring = sys.argv[1]
d = '/var/wiki'
#lgs = glob.glob('/var/wiki/*')
results = []
for lg in lgs:
qp = xapian.QueryParser()
try:
database = xapian.Database('/var/wiki/xapian/%s/index' % lg)
except:
#print "no index found for ", lg
continue
qp.set_database(database)
msc = MoinSearchConnection(database)
qresults = msc.get_all_documents(query=qp.parse_query(querystring))
for r in qresults:
wikiname = r.data['title'][0]
wikinamefs = wikiname.replace(' ', '(20)').replace(',', '(2c)')
try:
revnr = open('%s/%s/pasges/%s/current' %
(d, lg, wikinamefs)).read().strip()
content = open('%s/%s/pages/%s/revisions/%s' %
(d, lg, wikinamefs, revnr)).read()
except IOError:
print "File not Found", wikinamefs
continue
matches = re.findall(
'%s%s%s' % (width * '.', querystring.lower(), width * '.'),
content.lower())
results.append({
'link': url % (lg, wikinamefs),
'matchcount': len(matches),
'matches': matches,
})
return result
def search(xap_db_path, querystring, offset=0, pagesize=10):
# offset - defines starting point within result set
# pagesize - defines number of records to retrieve
# Open the database we're going to search.
db = xapian.Database(xap_db_path)
# Set up a QueryParser with a stemmer and suitable prefixes
queryparser = xapian.QueryParser()
queryparser.set_stemmer(xapian.Stem("swedish"))
queryparser.set_stemming_strategy(queryparser.STEM_SOME)
queryparser.add_prefix("food", "XFOOD")
xapian_flags = xapian.QueryParser.FLAG_WILDCARD | xapian.QueryParser.FLAG_BOOLEAN | \
xapian.QueryParser.FLAG_SYNONYM | xapian.QueryParser.FLAG_BOOLEAN_ANY_CASE | \
xapian.QueryParser.FLAG_LOVEHATE | xapian.Query.OP_PHRASE
# And parse the query
queryparser.set_database(db)
query = queryparser.parse_query(querystring, xapian_flags)
# Use an Enquire object on the database to run the query
enquire = xapian.Enquire(db)
enquire.set_query(query)
matches = []
for match in enquire.get_mset(offset, pagesize):
s_fields = match.document.get_data()
fields = unicode(s_fields, 'utf-8')
# print u"%(rank)i: #%(docid)3.3i %(data)s" % {
# 'rank': match.rank + 1,
# 'docid': match.docid,
# 'data': fields
# }
# part = [match.rank, match.docid, fields]
matches.append(fields)
return matches
def get_full_text_matches(database, full_text_query, offset, limit,
match_decider):
u"""Does the actual full-text search with Xapian.
:Parameters:
- `database`: the name of the RefDB database
- `full_text_query`: the raw query string for the full text search; must
not be empty
- `offset`: offset of the returned hits within the complete hits list
- `limit`: maximal number of returned hits
- `match_decider`: Xapian match decider object, e.g. for taking the other
search parameters into account
:type database: unicode
:type full_text_query: unicode
:type offset: int
:type limit: int
:type match_decider: `MatchDecider`
:Return:
the found matches
:rtype: ``Xapian.MSet``
"""
database = xapian.Database(
os.path.join("/var/lib/django_refdb_indices", database))
enquire = xapian.Enquire(database)
# enquire.set_collapse_key(0)
query_parser = xapian.QueryParser()
stemmer = xapian.Stem("english")
query_parser.set_stemmer(stemmer)
query_parser.set_database(database)
query_parser.set_stemming_strategy(xapian.QueryParser.STEM_SOME)
query = query_parser.parse_query(full_text_query)
enquire.set_query(query)
return enquire.get_mset(offset, limit, None, match_decider)
def crossquery(querystring, lg):
url = 'http://www.galoes.org/grammars/%s/%s'
width = 20
d = '/var/wiki'
results = []
qp = xapian.QueryParser()
database = xapian.Database('/var/wiki/xapian/%s/index' % lg)
qp.set_database(database)
msc = MoinSearchConnection(database)
qresults = msc.get_all_documents(query=qp.parse_query(querystring))
for r in qresults:
wikiname = r.data['title'][0]
wikinamefs = wikiutil.quoteWikinameFS(wikiname)
try:
refv = '%s/%s/pages/%s/current' % (d, lg, wikinamefs)
revnr = open(refv).read().strip()
contentf = '%s/%s/pages/%s/revisions/%s' % (d, lg, wikinamefs,
revnr)
content = open(contentf).read().decode('utf8')
except IOError:
print "File not Found", wikinamefs
continue
matches = re.findall(
u'%s%s%s' % ('.' + '{,%i}' % width, querystring.lower(),
'.' + '{,%i}' % width), content.lower())
print matches
results.append({
'link': url % (lg, wikinamefs),
'totalmatches': len(matches),
'matches': matches,
'name': wikiname,
})
return results
def test_userstem():
mystem = MyStemmer()
stem = xapian.Stem(mystem)
expect(stem(b'test'), b'tst')
stem2 = xapian.Stem(mystem)
expect(stem2(b'toastie'), b'tst')
indexer = xapian.TermGenerator()
indexer.set_stemmer(xapian.Stem(MyStemmer()))
doc = xapian.Document()
indexer.set_document(doc)
indexer.index_text(b'hello world')
s = '/'
for t in doc.termlist():
s += t.term.decode('utf-8')
s += '/'
expect(s, '/Zhll/Zwrld/hello/world/')
parser = xapian.QueryParser()
parser.set_stemmer(xapian.Stem(MyStemmer()))
parser.set_stemming_strategy(xapian.QueryParser.STEM_ALL)
expect_query(parser.parse_query(b'color television'), '(clr@1 OR tlvsn@2)')
from softwarecenter.enums import XapianValues
import softwarecenter.paths
# this is not a test as such, more a example of how xapian search
# work and useful features around them
if __name__ == "__main__":
if len(sys.argv) > 1:
search_term = sys.argv[1]
else:
search_term = "app"
db = xapian.Database(softwarecenter.paths.XAPIAN_PATH)
parser = xapian.QueryParser()
#parser.set_stemmer(xapian.Stem("english"))
#parser.set_stemming_strategy(xapian.QueryParser.STEM_ALL)
parser.set_database(db)
#parser.add_prefix("pkg", "AP")
query = parser.parse_query(
search_term,
xapian.QueryParser.FLAG_PARTIAL | xapian.QueryParser.FLAG_WILDCARD)
enquire = xapian.Enquire(db)
enquire.set_sort_by_value_then_relevance(XapianValues.POPCON)
enquire.set_query(query)
matches = enquire.get_mset(0, db.get_doccount())
print "Matches:"
for m in matches:
doc = m.document
def get_query_from_search_entry(search_term):
if not search_term:
return xapian.Query("")
parser = xapian.QueryParser()
user_query = parser.parse_query(search_term)
return user_query
while index < len(sys.argv):
arg = sys.argv[index]
index += 1
if arg == '--':
# Passed marker, move to parsing relevant docids.
break
query_string += ' '
query_string += arg
# Create an RSet with the listed docids in.
reldocs = xapian.RSet()
for index in range(index, len(sys.argv)):
reldocs.add_document(int(sys.argv[index]))
# Parse the query string to produce a Xapian::Query object.
qp = xapian.QueryParser()
stemmer = xapian.Stem("english")
qp.set_stemmer(stemmer)
qp.set_database(database)
qp.set_stemming_strategy(xapian.QueryParser.STEM_SOME)
query = qp.parse_query(query_string)
if not query.empty():
print("Parsed query is: %s" % str(query))
# Find the top 10 results for the query.
enquire.set_query(query)
matches = enquire.get_mset(0, 10, reldocs)
# Display the results.
print("%i results found." % matches.get_matches_estimated())
def test_all():
# Test the version number reporting functions give plausible results.
v = "%d.%d.%d" % (xapian.major_version(), xapian.minor_version(),
xapian.revision())
v2 = xapian.version_string()
expect(v2, v, "Unexpected version output")
# A regexp check would be better, but seems to create a bogus "leak" of -1
# objects in Python 3.
expect(len(xapian.__version__.split('.')), 3,
'xapian.__version__ not X.Y.Z')
expect((xapian.__version__.split('.'))[0], '1',
'xapian.__version__ not "1.Y.Z"')
def access_cvar():
res = xapian.cvar
print("Unhandled constants: ", res)
return res
# Check that SWIG isn't generating cvar (regression test for ticket#297).
#
# Python 3.5 generates a different exception message here to earlier
# versions, so we need a check which matches both.
expect_exception(AttributeError,
lambda msg: msg.find("has no attribute 'cvar'") != -1,
access_cvar)
stem = xapian.Stem(b"english")
expect(str(stem), "Xapian::Stem(english)", "Unexpected str(stem)")
doc = xapian.Document()
doc.set_data(b"a\0b")
if doc.get_data() == b"a":
raise TestFail("get_data+set_data truncates at a zero byte")
expect(doc.get_data(), b"a\0b",
"get_data+set_data doesn't transparently handle a zero byte")
doc.set_data(b"is there anybody out there?")
doc.add_term(b"XYzzy")
doc.add_posting(stem(b"is"), 1)
doc.add_posting(stem(b"there"), 2)
doc.add_posting(stem(b"anybody"), 3)
doc.add_posting(stem(b"out"), 4)
doc.add_posting(stem(b"there"), 5)
db = xapian.WritableDatabase('', xapian.DB_BACKEND_INMEMORY)
db.add_document(doc)
expect(db.get_doccount(), 1, "Unexpected db.get_doccount()")
terms = ["smoke", "test", "terms"]
expect_query(
xapian.Query(xapian.Query.OP_OR, [t.encode('utf-8') for t in terms]),
"(smoke OR test OR terms)")
query1 = xapian.Query(xapian.Query.OP_PHRASE,
(b"smoke", b"test", b"tuple"))
query2 = xapian.Query(xapian.Query.OP_XOR,
(xapian.Query(b"smoke"), query1, b"string"))
expect_query(query1, "(smoke PHRASE 3 test PHRASE 3 tuple)")
expect_query(
query2, "(smoke XOR (smoke PHRASE 3 test PHRASE 3 tuple) XOR string)")
subqs = ["a", "b"]
expect_query(
xapian.Query(xapian.Query.OP_OR, [s.encode('utf-8') for s in subqs]),
"(a OR b)")
expect_query(xapian.Query(xapian.Query.OP_VALUE_RANGE, 0, b'1', b'4'),
"VALUE_RANGE 0 1 4")
# Check database factory functions are wrapped as expected (or not wrapped
# in the first cases):
expect_exception(
AttributeError,
lambda msg: msg.find("has no attribute 'open_stub'") != -1,
lambda: xapian.open_stub(b"nosuchdir/nosuchdb"))
expect_exception(
AttributeError,
lambda msg: msg.find("has no attribute 'open_stub'") != -1,
lambda: xapian.open_stub(b"nosuchdir/nosuchdb", xapian.DB_OPEN))
expect_exception(
xapian.DatabaseOpeningError, None,
lambda: xapian.Database(b"nosuchdir/nosuchdb", xapian.DB_BACKEND_STUB))
expect_exception(
xapian.DatabaseOpeningError, None, lambda: xapian.WritableDatabase(
b"nosuchdir/nosuchdb", xapian.DB_OPEN | xapian.DB_BACKEND_STUB))
expect_exception(
xapian.DatabaseOpeningError, None, lambda: xapian.Database(
b"nosuchdir/nosuchdb", xapian.DB_BACKEND_GLASS))
expect_exception(
xapian.DatabaseCreateError, None, lambda: xapian.WritableDatabase(
b"nosuchdir/nosuchdb", xapian.DB_CREATE | xapian.DB_BACKEND_GLASS))
expect_exception(
xapian.FeatureUnavailableError, None, lambda: xapian.Database(
b"nosuchdir/nosuchdb", xapian.DB_BACKEND_CHERT))
expect_exception(
xapian.FeatureUnavailableError, None, lambda: xapian.WritableDatabase(
b"nosuchdir/nosuchdb", xapian.DB_CREATE | xapian.DB_BACKEND_CHERT))
expect_exception(xapian.NetworkError, None, xapian.remote_open,
b"/bin/false", b"")
expect_exception(xapian.NetworkError, None, xapian.remote_open_writable,
b"/bin/false", b"")
expect_exception(xapian.NetworkError, None, xapian.remote_open,
b"127.0.0.1", 0, 1)
expect_exception(xapian.NetworkError, None, xapian.remote_open_writable,
b"127.0.0.1", 0, 1)
# Check wrapping of MatchAll and MatchNothing:
expect_query(xapian.Query.MatchAll, "<alldocuments>")
expect_query(xapian.Query.MatchNothing, "")
# Feature test for Query.__iter__
term_count = 0
for term in query2:
term_count += 1
expect(term_count, 4, "Unexpected number of terms in query2")
enq = xapian.Enquire(db)
enq.set_query(xapian.Query(xapian.Query.OP_OR, b"there", b"is"))
mset = enq.get_mset(0, 10)
expect(mset.size(), 1, "Unexpected mset.size()")
expect(len(mset), 1, "Unexpected mset.size()")
# Feature test for Enquire.matching_terms(docid)
term_count = 0
for term in enq.matching_terms(mset.get_hit(0)):
term_count += 1
expect(term_count, 2, "Unexpected number of matching terms")
# Feature test for MSet.__iter__
msize = 0
for match in mset:
msize += 1
expect(msize, mset.size(), "Unexpected number of entries in mset")
terms = b" ".join(enq.matching_terms(mset.get_hit(0)))
expect(terms, b"is there", "Unexpected terms")
# Feature test for ESet.__iter__
rset = xapian.RSet()
rset.add_document(1)
eset = enq.get_eset(10, rset)
term_count = 0
for term in eset:
term_count += 1
expect(term_count, 3, "Unexpected number of expand terms")
# Feature test for Database.__iter__
term_count = 0
for term in db:
term_count += 1
expect(term_count, 5, "Unexpected number of terms in db")
# Feature test for Database.allterms
term_count = 0
for term in db.allterms():
term_count += 1
expect(term_count, 5, "Unexpected number of terms in db.allterms")
# Feature test for Database.postlist
count = 0
for posting in db.postlist(b"there"):
count += 1
expect(count, 1, "Unexpected number of entries in db.postlist('there')")
# Feature test for Database.postlist with empty term (alldocspostlist)
count = 0
for posting in db.postlist(b""):
count += 1
expect(count, 1, "Unexpected number of entries in db.postlist('')")
# Feature test for Database.termlist
count = 0
for term in db.termlist(1):
count += 1
expect(count, 5, "Unexpected number of entries in db.termlist(1)")
# Feature test for Database.positionlist
count = 0
for term in db.positionlist(1, b"there"):
count += 1
expect(count, 2,
"Unexpected number of entries in db.positionlist(1, 'there')")
# Feature test for Document.termlist
count = 0
for term in doc.termlist():
count += 1
expect(count, 5, "Unexpected number of entries in doc.termlist()")
# Feature test for TermIter.skip_to
term = doc.termlist()
term.skip_to(b'n')
while True:
try:
x = next(term)
except StopIteration:
break
if x.term < b'n':
raise TestFail("TermIter.skip_to didn't skip term '%s'" %
x.term.decode('utf-8'))
# Feature test for Document.values
count = 0
for term in list(doc.values()):
count += 1
expect(count, 0, "Unexpected number of entries in doc.values")
# Check exception handling for Xapian::DocNotFoundError
expect_exception(xapian.DocNotFoundError, "Docid 3 not found",
db.get_document, 3)
# Check value of OP_ELITE_SET
expect(xapian.Query.OP_ELITE_SET, 10, "Unexpected value for OP_ELITE_SET")
# Feature test for MatchDecider
doc = xapian.Document()
doc.set_data(b"Two")
doc.add_posting(stem(b"out"), 1)
doc.add_posting(stem(b"outside"), 1)
doc.add_posting(stem(b"source"), 2)
doc.add_value(0, b"yes")
db.add_document(doc)
class testmatchdecider(xapian.MatchDecider):
def __call__(self, doc):
return doc.get_value(0) == b"yes"
query = xapian.Query(stem(b"out"))
enquire = xapian.Enquire(db)
enquire.set_query(query)
mset = enquire.get_mset(0, 10, None, testmatchdecider())
expect(mset.size(), 1,
"Unexpected number of documents returned by match decider")
expect(mset.get_docid(0), 2, "MatchDecider mset has wrong docid in")
# Feature test for ExpandDecider
class testexpanddecider(xapian.ExpandDecider):
def __call__(self, term):
return (not term.startswith(b'a'))
enquire = xapian.Enquire(db)
rset = xapian.RSet()
rset.add_document(1)
eset = enquire.get_eset(10, rset, xapian.Enquire.USE_EXACT_TERMFREQ, 1.0,
testexpanddecider())
eset_terms = [item.term for item in eset]
expect(len(eset_terms), eset.size(),
"Unexpected number of terms returned by expand")
if [t for t in eset_terms if t.startswith(b'a')]:
raise TestFail("ExpandDecider was not used")
# Check min_wt argument to get_eset() works (new in 1.2.5).
eset = enquire.get_eset(100, rset, xapian.Enquire.USE_EXACT_TERMFREQ)
expect([i.weight for i in eset][-1] < 1.9, True,
"test get_eset() without min_wt")
eset = enquire.get_eset(100, rset, xapian.Enquire.USE_EXACT_TERMFREQ, 1.0,
None, 1.9)
expect([i.weight for i in eset][-1] >= 1.9, True, "test get_eset() min_wt")
# Check QueryParser parsing error.
qp = xapian.QueryParser()
expect_exception(xapian.QueryParserError,
"Syntax: <expression> AND <expression>", qp.parse_query,
b"test AND")
# Check QueryParser pure NOT option
qp = xapian.QueryParser()
expect_query(
qp.parse_query(b"NOT test", qp.FLAG_BOOLEAN + qp.FLAG_PURE_NOT),
"(<alldocuments> AND_NOT test@1)")
# Check QueryParser partial option
qp = xapian.QueryParser()
qp.set_database(db)
qp.set_default_op(xapian.Query.OP_AND)
qp.set_stemming_strategy(qp.STEM_SOME)
qp.set_stemmer(xapian.Stem(b'en'))
expect_query(qp.parse_query(b"foo o", qp.FLAG_PARTIAL),
"(Zfoo@1 AND ((SYNONYM WILDCARD OR o) OR Zo@2))")
expect_query(qp.parse_query(b"foo outside", qp.FLAG_PARTIAL),
"(Zfoo@1 AND ((SYNONYM WILDCARD OR outside) OR Zoutsid@2))")
# Test supplying unicode strings
expect_query(xapian.Query(xapian.Query.OP_OR, (b'foo', b'bar')),
'(foo OR bar)')
expect_query(xapian.Query(xapian.Query.OP_OR, (b'foo', b'bar\xa3')),
'(foo OR bar\\xa3)')
expect_query(xapian.Query(xapian.Query.OP_OR, (b'foo', b'bar\xc2\xa3')),
'(foo OR bar\u00a3)')
expect_query(xapian.Query(xapian.Query.OP_OR, b'foo', b'bar'),
'(foo OR bar)')
expect_query(
qp.parse_query(b"NOT t\xe9st", qp.FLAG_BOOLEAN + qp.FLAG_PURE_NOT),
"(<alldocuments> AND_NOT Zt\u00e9st@1)")
doc = xapian.Document()
doc.set_data(b"Unicode with an acc\xe9nt")
doc.add_posting(stem(b"out\xe9r"), 1)
expect(doc.get_data(), b"Unicode with an acc\xe9nt")
term = next(doc.termlist()).term
expect(term, b"out\xe9r")
# Check simple stopper
stop = xapian.SimpleStopper()
qp.set_stopper(stop)
expect(stop(b'a'), False)
expect_query(qp.parse_query(b"foo bar a", qp.FLAG_BOOLEAN),
"(Zfoo@1 AND Zbar@2 AND Za@3)")
stop.add(b'a')
expect(stop(b'a'), True)
expect_query(qp.parse_query(b"foo bar a", qp.FLAG_BOOLEAN),
"(Zfoo@1 AND Zbar@2)")
# Feature test for custom Stopper
class my_b_stopper(xapian.Stopper):
def __call__(self, term):
return term == b"b"
def get_description(self):
return "my_b_stopper"
stop = my_b_stopper()
expect(stop.get_description(), "my_b_stopper")
qp.set_stopper(stop)
expect(stop(b'a'), False)
expect_query(qp.parse_query(b"foo bar a", qp.FLAG_BOOLEAN),
"(Zfoo@1 AND Zbar@2 AND Za@3)")
expect(stop(b'b'), True)
expect_query(qp.parse_query(b"foo bar b", qp.FLAG_BOOLEAN),
"(Zfoo@1 AND Zbar@2)")
# Test TermGenerator
termgen = xapian.TermGenerator()
doc = xapian.Document()
termgen.set_document(doc)
termgen.index_text(b'foo bar baz foo')
expect([(item.term, item.wdf, [pos for pos in item.positer])
for item in doc.termlist()], [(b'bar', 1, [2]), (b'baz', 1, [3]),
(b'foo', 2, [1, 4])])
# Check DateValueRangeProcessor works
context("checking that DateValueRangeProcessor works")
qp = xapian.QueryParser()
vrpdate = xapian.DateValueRangeProcessor(1, 1, 1960)
qp.add_valuerangeprocessor(vrpdate)
query = qp.parse_query(b'12/03/99..12/04/01')
expect(str(query), 'Query(VALUE_RANGE 1 19991203 20011204)')
# Regression test for bug#193, fixed in 1.0.3.
context("running regression test for bug#193")
vrp = xapian.NumberValueRangeProcessor(0, b'$', True)
a = '$10'
b = '20'
slot, a, b = vrp(a, b.encode('utf-8'))
expect(slot, 0)
expect(xapian.sortable_unserialise(a), 10)
expect(xapian.sortable_unserialise(b), 20)
# Feature test for xapian.FieldProcessor
context("running feature test for xapian.FieldProcessor")
class testfieldprocessor(xapian.FieldProcessor):
def __call__(self, s):
if s == 'spam':
raise Exception('already spam')
return xapian.Query("spam")
qp.add_prefix('spam', testfieldprocessor())
qp.add_boolean_prefix('boolspam', testfieldprocessor())
query = qp.parse_query('spam:ignored')
expect(str(query), 'Query(spam)')
# FIXME: This doesn't currently work:
# expect_exception(Exception, 'already spam', qp.parse_query, 'spam:spam')
# Regression tests copied from PHP (probably always worked in python, but
# let's check...)
context("running regression tests for issues which were found in PHP")
# PHP overload resolution involving boolean types failed.
enq.set_sort_by_value(1, True)
# Regression test - fixed in 0.9.10.1.
oqparser = xapian.QueryParser()
oquery = oqparser.parse_query(b"I like tea")
# Regression test for bug#192 - fixed in 1.0.3.
enq.set_cutoff(100)
# Test setting and getting metadata
expect(db.get_metadata(b'Foo'), b'')
db.set_metadata(b'Foo', b'Foo')
expect(db.get_metadata(b'Foo'), b'Foo')
expect_exception(xapian.InvalidArgumentError,
"Empty metadata keys are invalid", db.get_metadata, b'')
expect_exception(xapian.InvalidArgumentError,
"Empty metadata keys are invalid", db.set_metadata, b'',
b'Foo')
expect_exception(xapian.InvalidArgumentError,
"Empty metadata keys are invalid", db.get_metadata, b'')
# Test OP_SCALE_WEIGHT and corresponding constructor
expect_query(
xapian.Query(xapian.Query.OP_SCALE_WEIGHT, xapian.Query(b'foo'), 5),
"5 * foo")
