Python SolrConnection.query方法代码示例

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
def word_matches_for_page(page_id, words):
    """
    Gets a list of pre-analyzed words for a list of words on a particular
    page. So if you pass in 'manufacturer' you can get back a list like
    ['Manufacturer', 'manufacturers', 'MANUFACTURER'] etc ...
    """
    solr = SolrConnection(settings.SOLR)

    # Make sure page_id is of type str, else the following string
    # operation may result in a UnicodeDecodeError. For example, see
    # ticket #493
    if not isinstance(page_id, str):
        page_id = str(page_id)

    ocr_list = ['ocr', ]
    ocr_list.extend(['ocr_%s' % l for l in settings.SOLR_LANGUAGES])
    ocrs = ' OR '.join([query_join(words, o) for o in ocr_list])
    q = 'id:%s AND (%s)' % (page_id, ocrs)
    params = {"hl.snippets": 100, "hl.requireFieldMatch": 'true', "hl.maxAnalyzedChars": '102400'}
    response = solr.query(q, fields=['id'], highlight=ocr_list, **params)

    if page_id not in response.highlighting:
        return []

    words = set()
    for ocr in ocr_list:
        if ocr in response.highlighting[page_id]:
            for context in response.highlighting[page_id][ocr]:
                words.update(find_words(context))
    return list(words)

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
 def _get_count(self):
     "Returns the total number of objects, across all pages."
     if self._count is None:
         solr = SolrConnection(settings.SOLR)  # TODO: maybe keep connection around?
         solr_response = solr.query(self._q, fields=['id'])
         self._count = int(solr_response.results.numFound)
     return self._count

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
def oralookup(pid=None, uuid=None, fields_to_return="f_name, f_subject, f_keyphrase, faculty, f_institution, thesis_type, content_type, collection", endpoint="http://ora.ouls.ox.ac.uk:8080/solr/select"):
  s = SolrConnection(endpoint)
  results = {}
  query = ""
  if pid:
    pid = "\:".join(pid.split(":"))
    query = "id:%s" % pid
  elif uuid:
    query = "id:uuid\:%s" % uuid
  else:
    return results
  # Running actual query (3 tries, failover)
  tries = 0
  while(tries != 3):
    try:
      r = s.query(q = query, fields = fields_to_return)
      logger.debug("Solr response: %s" % r.header)
      tries = 3
    except BadStatusLine:
      sleep(0.5)
      tries = tries + 1
  try:
    assert len(r.results) == 1
    return r.results[0]
  except ValueError:
    logger.warn("Couldn't parse json response from Solr endpoint: %s" % r)
    return {}
  except AssertionError:
    logger.warn("Couldn't assert that only a single result was fetched: %s" % results)
    return {}

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
def get_page_text(page):
    no_text = ["Text not available"]
    solr = SolrConnection(settings.SOLR)
    query = 'id:"%s"' % page.url
    solr_results = solr.query(query)
    results_attribute = getattr(solr_results, 'results', None)
    if isinstance(results_attribute, list) and len(results_attribute) > 0:
        return results_attribute[0].get('ocr', no_text)
    else:
        return no_text

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
def similar_pages(page):
    solr = SolrConnection(settings.SOLR)
    d = page.issue.date_issued
    year, month, day = '{0:02d}'.format(d.year), '{0:02d}'.format(d.month), '{0:02d}'.format(d.day) 
    date = ''.join(map(str, (year, month, day)))

    query = '+type:page AND date:%s AND %s AND NOT(lccn:%s)' % (date, query_join(map(lambda p: p.city, 
                                           page.issue.title.places.all()), 'city'), page.issue.title.lccn)
    response = solr.query(query, rows=25)
    results = response.results
    return map(lambda kwargs: utils.get_page(**kwargs), 
               map(lambda r: urlresolvers.resolve(r['id']).kwargs, results))

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
def execute_solr_query(query, fields, sort, sort_order, rows, start):
    solr = SolrConnection(settings.SOLR) # TODO: maybe keep connection around?
    solr_response = solr.query(query,
                               fields=['lccn', 'title',
                                       'edition',
                                       'place_of_publication',
                                       'start_year', 'end_year',
                                       'language'],
                               rows=rows,
                               sort=sort,
                               sort_order=sort_order,
                               start=start)
    return solr_response

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
def execute_solr_query(query, fields, sort, sort_order, rows, start):
    # default arg_separator - underscore wont work if fields to facet on
    # themselves have underscore in them
    solr = SolrConnection(settings.SOLR)  # TODO: maybe keep connection around?
    solr_response = solr.query(query,
                               fields=['lccn', 'title',
                                       'edition',
                                       'place_of_publication',
                                       'start_year', 'end_year',
                                       'language'],
                               rows=rows,
                               sort=sort,
                               sort_order=sort_order,
                               start=start)
    return solr_response

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
class QuoteResource:

	def __init__(self):
		self.solr = SolrConnection('http://localhost:8983/solr')

	@cherrypy.expose
	def index(self, callback=None, person=None, topic=None):

		filters = []
		filters.append('type:quote')

		if person:
			filters.append('person_t:%s' % person)
		if topic:
			filters.append('quote_t:%s' % topic)

		results = self.solr.query(q = ' AND '.join(filters), rows=100)
		docs = []

		timeline = { 
 	   		  'timeline':
    				{
				        'headline':'OnTheRecord',
				        'type':'default',
					'startDate':'2012,1,1',
					'text':'We help you track quotations from politicians over time',
				}
			}
		for result in results:
			doc = { 
                			"startDate":result['date'].strftime('%Y,%m,%d'),
                			"headline":result['person'],
                			"text":'<a href="' + result['url'] + '">'+result['title'] +'</a>',
                			"asset":
                			{
                    				"media":"<blockquote>\""+result['quote'] + "\"</blockquote>",
                    				"credit":"",
                    				"caption":""
                			}
            			}
			docs.append(doc)
		timeline['timeline']['date'] = docs;
		
		cherrypy.response.headers['Content-Type'] = 'application/json; charset=utf-8'
		return json.dumps(timeline, ensure_ascii=False, indent=4).encode('utf-8')

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
    def page(self, number):
        """
        Override the page method in Paginator since Solr has already
        paginated stuff for us.
        """

        number = self.validate_number(number)

        # figure out the solr query and execute it
        solr = SolrConnection(settings.SOLR) # TODO: maybe keep connection around?
        start = self.per_page * (number - 1)
        params = {"hl.snippets": 100, # TODO: make this unlimited
            "hl.requireFieldMatch": 'true', # limits highlighting slop
            "hl.maxAnalyzedChars": '102400', # increased from default 51200
            }
        sort_field, sort_order = _get_sort(self.query.get('sort'), in_pages=True)
        solr_response = solr.query(self._q,
                                   fields=['id', 'title', 'date', 'sequence',
                                           'edition_label', 'section_label'],
                                   highlight=self._ocr_list,
                                   rows=self.per_page,
                                   sort=sort_field,
                                   sort_order=sort_order,
                                   start=start,
                                   **params)

        pages = []
        for result in solr_response.results:
            page = models.Page.lookup(result['id'])
            if not page:
                continue
            words = set()
            coords = solr_response.highlighting[result['id']]
            for ocr in self._ocr_list:
                for s in coords.get(ocr) or []:
                    words.update(find_words(s))
            page.words = sorted(words, key=lambda v: v.lower())

            page.highlight_url = self.highlight_url(page.url,
                                                    page.words,
                                                    number, len(pages))
            pages.append(page)

        return Page(pages, number, self)

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
class PersonResource:

	def __init__(self):
		self.solr = SolrConnection('http://localhost:8983/solr')

	@cherrypy.expose
	def index(self):

		results = self.solr.query('*:*', facet='true', facet_field='person')

		for person in results.facet_counts[u'facet_fields'][u'person']:

			print person

		cherrypy.response.headers['Content-Type'] = 'application/json; charset=utf-8'
		return json.dumps({
			'test': 'test',
			'data': 'data'
		}, ensure_ascii=False, indent=4).encode('utf-8')

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
    def __init__(self, query):
        self.query = query.copy()

        # figure out the solr query
        q = title_search(self.query)

        try:
            page = int(self.query.get('page'))
        except:
            page = 1

        try:
            rows = int(self.query.get('rows'))
        except:
            rows = 50
        start = rows * (page - 1)

        # determine sort order
        sort_field, sort_order = _get_sort(self.query.get('sort'))

        # execute query
        solr = SolrConnection(settings.SOLR) # TODO: maybe keep connection around?
        solr_response = solr.query(q,
                                   fields=['lccn', 'title',
                                           'edition',
                                           'place_of_publication',
                                           'start_year', 'end_year',
                                           'language'],
                                   rows=rows,
                                   sort=sort_field,
                                   sort_order=sort_order,
                                   start=start)

        # convert the solr documents to Title models
        # could use solr doc instead of going to db, if performance requires it
        lccns = [d['lccn'] for d in solr_response.results]
        results = []
        for lccn in lccns:
            try:
                title = models.Title.objects.get(lccn=lccn)
                results.append(title)
            except models.Title.DoesNotExist, e:
                pass # TODO: log exception

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
    def setUpClass(cls):
        # First, add a folio to Solr so that the image_uri can be retrieved during the MEI conversion
        # Using curl here because it turned out to be easier than solrconn.add and gives better error messages
        os.system("curl {0}/update/?commit=true -H 'Content-Type: text/xml' -d '<add><doc>\
        <field name=\"id\">testid</field>\
        <field name=\"type\">cantusdata_folio</field>\
        <field name=\"manuscript_id\">{1}</field>\
        <field name=\"number\">{2}</field>\
        <field name=\"image_uri\">{3}</field>\
        </doc></add>'".format(settings.SOLR_SERVER, MEI_FIXTURE_ID, MEI_FIXTURE_FOLIO, MEI_FIXTURE_URI))

        docs = list(MEIConverter.process_file(MEI_FIXTURE, MEI_FIXTURE_SIGLUM, MEI_FIXTURE_ID))

        # Sanity check
        solrconn = SolrConnection(settings.SOLR_SERVER)
        prequery = solrconn.query('type:cantusdata_music_notation AND manuscript:' + MEI_FIXTURE_SIGLUM)
        assert prequery.numFound == 0, 'MEI was already in the database when loading the test fixture'

        solrconn.add_many(docs)
        solrconn.commit()

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
if options.wiki:
    query = 'host:%s' % (options.wiki)
elif not options.query:
    raise Exception('A wiki is required, passed as host name')

if options.query:
    query += ' '+options.query

specifier = options.wiki if options.wiki else str(os.getpid())

conn = SolrConnection('http://search-s10.prod.wikia.net:8983/solr')

print query

response = conn.query(query, fields=['html_en','nolang_txt','html', 'title', 'title_en', 'id'], rows=100)
paginator = SolrPaginator(response)

def initialize_dir(page):
    paths = [options.dest, specifier, str(page)]
    fullpath = ''
    for path in paths:
        fullpath += path + '/'
        if not os.path.exists(fullpath):
            os.mkdir(fullpath)
    return fullpath

for page in paginator.page_range:
    pagedir = initialize_dir(page)
    lockfilepath = pagedir+'/LOCK'
    with open(lockfilepath, 'w') as lockfile:

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
if not options.wiki:
    raise Exception('A wiki is required, passed as host name')

conn = SolrConnection('http://search-s10.prod.wikia.net:8983/solr')

query = ["host:'%s'" % (options.wiki)]

query += ['ns:%d ' % (int(options.namespace))]

if options.start_date or options.end_date:
    start = options.start_date + 'T00:00:00.000Z' if options.start_date else '*'
    end = options.end_date + 'T00:00:00.000Z' if options.end_date else '*'
    query += ['created:[%s TO %s]' % (start, end)]

response = conn.query(' AND '.join(query), fields=['html_en','nolang_txt','html'])
paginator = SolrPaginator(response)

print paginator.count, 'results to chomp through...'

polarities, subjectivities = [], []

for page in paginator.page_range:
    for doc in paginator.page(page).object_list:
        sent = sentiment.sentiment(doc.get('html_en', doc.get('nolang_txt', doc.get('html'))))
        if ( sent == (0,0)):
            continue
        polarities.append(sent[0])
        subjectivities.append(sent[1])
    if page % int(paginator.num_pages/10) == 0:
        print "========","On page", page, "of", paginator.num_pages, "======="

# 需要导入模块: from solr import SolrConnection [as 别名]
# 或者: from solr.SolrConnection import query [as 别名]
    "-w", "--wiki", dest="wiki", action="store", default=None, help="Specifies the wiki to perform calculations against"
)
parser.add_option(
    "-n", "--sents", dest="num_sents", action="store", default=5, help="Specifies the number of sentences to write"
)

(options, args) = parser.parse_args()

if options.id:
    query = "id:%s" % (options.id)
elif options.wiki:
    query = "host:'%s' AND ns:0" % (options.wiki)
else:
    raise Exception("A wiki  or ID is required, passed as host name")

conn = SolrConnection("http://search-s10.prod.wikia.net:8983/solr")

response = conn.query(query, fields=["html_en", "nolang_txt", "html", "title", "title_en", "id"])
paginator = SolrPaginator(response)

summarizer = SimpleSummarizer()

for page in paginator.page_range:
    for doc in paginator.page(page).object_list:
        text = doc.get("html_en", doc.get("nolang_txt", doc.get("html")))
        title = doc.get("title_en", doc.get("title", doc["id"]))
        summed = summarizer.get_summarized(text, options.num_sents)
        print "\t\t=======", title, "======="
        print "\t" + "\n\t".join([sent for sent in summed if not sent.startswith("Contents")])
        print "\t\t====================================="