import warnings
from sunburnt import search
import copy
-
+from httplib2 import socket
class TermVectorOptions(search.Options):
- option_name = "tv"
-
def __init__(self, schema, original=None):
self.schema = schema
if original is None:
def options(self):
opts = {}
- opts['tv'] = 'true'
+ if self.positions or self.fields:
+ opts['tv'] = 'true'
if self.positions:
opts['tv.positions'] = 'true'
if self.fields:
newself.term_vectorer.update(positions, fields)
return newself
setattr(search.SolrSearch, 'term_vector', __term_vector)
-__original__init_common_modules = search.SolrSearch._init_common_modules
def __patched__init_common_modules(self):
__original__init_common_modules(self)
self.term_vectorer = TermVectorOptions(self.schema)
+__original__init_common_modules = search.SolrSearch._init_common_modules
setattr(search.SolrSearch, '_init_common_modules', __patched__init_common_modules)
def __init__(self, url, schemadoc=None, http_connection=None, mode='', retry_timeout=-1, max_length_get_url=sunburnt.MAX_LENGTH_GET_URL):
self.conn = CustomSolrConnection(url, http_connection, retry_timeout, max_length_get_url)
self.schemadoc = schemadoc
- if mode == 'r':
+ if 'w' not in mode:
self.writeable = False
- elif mode == 'w':
+ elif 'r' not in mode:
self.readable = False
- self.init_schema()
+ try:
+ self.init_schema()
+ except socket.error, e:
+ raise socket.error, "Cannot connect to Solr server, and search indexing is enabled (%s)" % str(e)
+
def _analyze(self, **kwargs):
if not self.readable:
end = int(wrd.xpath("int[@name='end']")[0].text)
matches.add((start, end))
- print matches
if matches:
return self.substring(kwargs['text'], matches,
- margins=kwargs.get('margins', 30),
- mark=kwargs.get('mark', ("<b>", "</b>")))
+ margins=kwargs.get('margins', 30),
+ mark=kwargs.get('mark', ("<b>", "</b>")))
else:
return None
def analyze(self, **kwargs):
- doc = self._analyze(self, **kwargs)
- terms = doc.xpath("/lst[@name='index']/arr[last()]/lst/str[1]")
+ doc = self._analyze(**kwargs)
+ terms = doc.xpath("//lst[@name='index']/arr[last()]/lst/str[1]")
terms = map(lambda n: unicode(n.text), terms)
return terms
start = None
end = None
totlen = len(text)
- matches_margins = map(lambda (s, e): (max(0, s - margins), min(totlen, e + margins)), matches)
- (start, end) = matches_margins[0]
-
- for (s, e) in matches_margins[1:]:
+ matches_margins = map(lambda (s, e):
+ ((s, e),
+ (max(0, s - margins), min(totlen, e + margins))),
+ matches)
+ (start, end) = matches_margins[0][1]
+ matches = []
+ for (m, (s, e)) in matches_margins[1:]:
if end < s or start > e:
continue
start = min(start, s)
end = max(end, e)
-
+ matches.append(m)
+
snip = text[start:end]
- matches = list(matches)
matches.sort(lambda a, b: cmp(b[0], a[0]))
+
for (s, e) in matches:
off = - start
snip = snip[:e + off] + mark[1] + snip[e + off:]
snip = snip[:s + off] + mark[0] + snip[s + off:]
# maybe break on word boundaries
+
return snip
+