class JsonEncoderTestCase(unittest.TestCase):
def setUp(self):
self.encoder = ScrapyJSONEncoder()
def test_encode_decode(self):
dt = datetime.datetime(2010, 1, 2, 10, 11, 12)
dts = "2010-01-02 10:11:12"
d = datetime.date(2010, 1, 2)
ds = "2010-01-02"
t = datetime.time(10, 11, 12)
ts = "10:11:12"
dec = Decimal("1000.12")
decs = "1000.12"
for input, output in [('foo', 'foo'), (d, ds), (t, ts), (dt, dts),
(dec, decs), (['foo', d], ['foo', ds])]:
self.assertEqual(self.encoder.encode(input), json.dumps(output))
def test_encode_deferred(self):
self.assertIn('Deferred', self.encoder.encode(defer.Deferred()))
def test_encode_request(self):
r = Request("http://www.example.com/lala")
rs = self.encoder.encode(r)
self.assertIn(r.method, rs)
self.assertIn(r.url, rs)
def test_encode_response(self):
r = Response("http://www.example.com/lala")
rs = self.encoder.encode(r)
self.assertIn(r.url, rs)
self.assertIn(str(r.status), rs)
def process_item(self, item, spider):
url = "http://localhost:9200/articles/%s" % (item["publication"].lower())
encoder = ScrapyJSONEncoder()
json_body = encoder.encode(item)
resp = requests.post(url, data=json_body)
log.msg("Item added to elasticSearch node. Response: " + resp.text)
return item
class HadoopExporter(BaseItemExporter):
def __init__(self, hadoop, **kwargs):
#self.con = file_write.Connection()
#self.con.connect(hadoop.ip, hadoop.port)
self.encoder = ScrapyJSONEncoder(**kwargs)
#self.seq = file_write.SeqFileSaver(self.con, '/common/crawler/%s/' % hadoop.username.replace(".", "/"),
# 1, '%s' % hadoop.username.replace(".", "_"))
self.encoding = 'utf-8'
self.fields_to_export = None
self.export_empty_fields = False
self.writer = SeqWriter(os.path.join(Utils.settings['SEQFILE_DIR'], hadoop.username.replace(".", "/")),
hadoop.username.replace(".", "_"))
def close_file(self):
print "close"
self.writer.close()
#self.seq.set_is_end()
#self.con.close()
def start_exporting(self):
pass
def finish_exporting(self):
pass
def export_item(self, item):
value = self.encoder.encode(dict(self._get_serialized_fields(item)))
self.writer.writeData(
item['key'] if 'key' in item else item['url'],
value
)
class RedisPipeline(object):
"""Pushes serialized item into a redis list/queue"""
def __init__(self, host, port, queue_type):
self.server = redis.Redis(host, port)
self.encoder = ScrapyJSONEncoder()
self.queue_type = queue_type
@classmethod
def from_settings(cls, settings):
host = settings.get('REDIS_HOST', 'localhost')
port = settings.get('REDIS_PORT', 6379)
queue_type = settings.get('QUEUE_TYPE', 'FIFO')
return cls(host, port)
def process_item(self, item, spider):
return deferToThread(self._process_item, item, spider)
def _process_item(self, item, spider):
key = self.item_key(item, spider)
data = self.encoder.encode(dict(item))
self.server.lpush(key, data)
if (self.queue_type == 'LIFO'):
self.server.lpush(key, data)
else:
self.server.rpush(key, data)
return item
def item_key(self, item, spider):
"""Returns redis key based on given spider"""
return "%s:items" % spider.name
def __init__(self, host, port, db, queue_name, store_id):
self.encoder = ScrapyJSONEncoder()
self.store_id = store_id
self.queue_name = queue_name
self.server = redis.Redis(host, port, db)
self.queue = rq.Queue(queue_name, connection=self.server)
class RabbitMQPipeline(object):
"""Pushes serialized item into a RabbitMQ list/queue"""
def __init__(self, server, exchange_name):
self.server = server
self.exchange_name = exchange_name
self.encoder = ScrapyJSONEncoder()
@classmethod
def from_settings(cls, settings):
server, redis_server = connection.from_settings(settings)
exchange_name = settings.get('RABBITMQ_EXCHANGE_NAME', EXCHANGE_NAME)
return cls(server, exchange_name)
@classmethod
def from_crawler(cls, crawler):
return cls.from_settings(crawler.settings)
def process_item(self, item, spider):
key = self.item_key(item, spider)
data = self.encoder.encode(item)
self.server.basic_publish(exchange=self.exchange_name,
routing_key=key,
body=data)
return item
def item_key(self, item, spider):
"""Returns RabbitMQ key based on given spider"""
return "%s:items" % spider.name
class RedisPipeline(object):
def __init__(self, server):
self.server = server
self.encoder = ScrapyJSONEncoder()
@classmethod
def from_settings(cls, settings):
server = connection.from_settings(settings)
return cls(server)
@classmethod
def from_crawler(cls, crawler):
return cls.from_settings(crawler.settings)
def process_item(self, item, spider):
return deferToThread(self._process_item, item, spider)
def _process_item(self, item, spider):
key = self.item_key(item, spider)
data = self.encoder.encode(item)
self.server.rpush(key, data)
return item
def item_key(self, item, spider):
return "%s:items" % spider.name
class RedisPipeline(object):
"""
Pushes serialized item into a redis.
Specific for SocialSpiders
"""
def __init__(self, server):
self.server = server
self.encoder = ScrapyJSONEncoder()
@classmethod
def from_settings(cls, settings):
server = connection.from_settings(settings)
return cls(server)
@classmethod
def from_crawler(cls, crawler):
return cls.from_settings(crawler.settings)
def process_item(self, item, spider):
return deferToThread(self._process_item, item, spider)
def _process_item(self, item, spider):
key = self.item_key(item, spider)
data = self.encoder.encode(item)
self.server.set(key, data.decode('utf-8'))
return item
def item_key(self, item, spider):
"""Returns redis key based on given spider"""
return "{}_{}".format(spider.name, item['search_name'])
class RedisPipeline(object):
"""Pushes serialized item into a redis list/queue"""
def __init__(self, server):
self.server = server
self.encoder = ScrapyJSONEncoder()
@classmethod
def from_settings(cls, settings):
server = connection.from_settings(settings)
return cls(server)
@classmethod
def from_crawler(cls, crawler):
return cls.from_settings(crawler.settings)
def process_item(self, item, spider):
return deferToThread(self._process_item, item, spider)
def _process_item(self, item, spider):
key = self.item_key(item, spider)
data = self.encoder.encode(item)
self.server.rpush(key, data)
return item
def item_key(self, item, spider):
"""Returns redis key based on given spider"""
return "%s:items" % spider.name
class JsonItemExporter(BaseItemExporter):
def __init__(self, file, **kwargs):
self._configure(kwargs, dont_fail=True)
self.file = file
# there is a small difference between the behaviour or JsonItemExporter.indent
# and ScrapyJSONEncoder.indent. ScrapyJSONEncoder.indent=None is needed to prevent
# the addition of newlines everywhere
json_indent = self.indent if self.indent is not None and self.indent > 0 else None
kwargs.setdefault('indent', json_indent)
kwargs.setdefault('ensure_ascii', not self.encoding)
self.encoder = ScrapyJSONEncoder(**kwargs)
self.first_item = True
def _beautify_newline(self):
if self.indent is not None:
self.file.write(b'\n')
def start_exporting(self):
self.file.write(b"[")
self._beautify_newline()
def finish_exporting(self):
self._beautify_newline()
self.file.write(b"]")
def export_item(self, item):
if self.first_item:
self.first_item = False
else:
self.file.write(b',')
self._beautify_newline()
itemdict = dict(self._get_serialized_fields(item))
data = self.encoder.encode(itemdict)
self.file.write(to_bytes(data, self.encoding))
class RedisPipeline(object):
"""Pushes serialized item into a scrapy_redis list/queue"""
def __init__(self, host, port):
self.server = redis.Redis(host, port)
self.encoder = ScrapyJSONEncoder()
@classmethod
def from_settings(cls, settings):
host = settings.get("REDIS_HOST", "localhost")
port = settings.get("REDIS_PORT", 6379)
return cls(host, port)
def process_item(self, item, spider):
return deferToThread(self._process_item, item, spider)
def _process_item(self, item, spider):
key = self.item_key(item, spider)
data = self.encoder.encode(dict(item))
self.server.rpush(key, data)
return item
def item_key(self, item, spider):
"""Returns scrapy_redis key based on given spider"""
return "%s:items" % spider.name
class RedisStoragePipeline(object):
def __init__(self, server):
self.server = server
self.encoder = ScrapyJSONEncoder()
@classmethod
def from_settings(cls, settings):
server = connection.from_settings(settings)
return cls(server)
@classmethod
def from_crawler(cls, crawler):
return cls.from_settings(crawler.settings)
def process_item(self, item, spider):
return deferToThread(self._process_item, item, spider)
def _process_item(self, item, spider):
data = self.encoder.encode(item)
if isinstance(item, GubaPostListItem):
key = self.item_key_list(item, spider)
if isinstance(item, GubaPostDetailItem):
key = self.item_key_detail(item, spider)
self.server.rpush(key, data)
return item
def item_key_list(self, item, spider):
stock_id = item['stock_id']
return "%s:list_items" % stock_id
def item_key_detail(self, item, spider):
stock_id = item['stock_id']
return "%s:detail_items" % stock_id
class DockerhubExtension(object):
@classmethod
def from_crawler(cls, crawler):
return cls(crawler)
def __init__(self, crawler):
self.crawler = crawler
self.job_path = crawler.settings.get('JOB_PATH')
if not self.job_path:
raise NotConfigured('no JOB_PATH set')
self.json_encoder = ScrapyJSONEncoder()
self.looping_call = LoopingCall(self.store_job_info)
self.looping_call.start(5)
crawler.signals.connect(self.store_job_info,
signal=signals.spider_closed)
def store_job_info(self):
with open(self.job_path, 'w') as f:
stats = self.crawler.stats.get_stats()
job_info = {
'stats': stats
}
job_info_json = self.json_encoder.encode(job_info)
f.write(job_info_json)
class RabbitMQPipeline(object):
"""Pushes serialized item into a RabbitMQ list/queue"""
def __init__(self, server):
self.server = server
self.encoder = ScrapyJSONEncoder()
@classmethod
def from_settings(cls, settings):
server = connection.from_settings(settings)
return cls(server)
@classmethod
def from_crawler(cls, crawler):
return cls.from_settings(crawler.settings)
def process_item(self, item, spider):
return deferToThread(self._process_item, item, spider)
def _process_item(self, item, spider):
key = self.item_key(item, spider)
data = self.encoder.encode(item)
self.server.basic_publish(exchange='',
routing_key=key,
body=data)
return item
def item_key(self, item, spider):
"""Returns RabbitMQ key based on given spider"""
return "%s:items" % spider.name
def __init__(self, recipients, mail, compressor, crawler):
self.recipients = recipients
self.mail = mail
self.encoder = ScrapyJSONEncoder(crawler=crawler)
self.files = defaultdict(compressor)
self.num_items = 0
self.num_errors = 0
class JsonLinesItemExporter(BaseItemExporter):
def __init__(self, file, **kwargs):
self._configure(kwargs, dont_fail=True)
self.file = file
self.encoder = ScrapyJSONEncoder(**kwargs)
def export_item(self, item):
itemdict = dict(self._get_serialized_fields(item))
self.file.write(to_bytes(self.encoder.encode(itemdict) + "\n"))
def __init__(self, producer, topic):
"""
:type producer: kafka.producer.Producer
:type topic: str or unicode
"""
self.producer = producer
self.topic = topic
self.encoder = ScrapyJSONEncoder()
def __init__(self, outputAddress, **kwargs):
self._configure(kwargs, dont_fail=True)
# self.file = file
# needs to be updated when migrating to proto-buff
self.encoder = ScrapyJSONEncoder(**kwargs)
self.first_item = True
# creating ZMQ context and socket
self.context = zmq.Context()
self.socket = self.context.socket(zmq.PUB)
# self.socket.setsockopt(zmq.LINGER, 100)
self.outputAddress = outputAddress
def __init__(self, hadoop, **kwargs):
#self.con = file_write.Connection()
#self.con.connect(hadoop.ip, hadoop.port)
self.encoder = ScrapyJSONEncoder(**kwargs)
#self.seq = file_write.SeqFileSaver(self.con, '/common/crawler/%s/' % hadoop.username.replace(".", "/"),
# 1, '%s' % hadoop.username.replace(".", "_"))
self.encoding = 'utf-8'
self.fields_to_export = None
self.export_empty_fields = False
self.writer = SeqWriter(os.path.join(Utils.settings['SEQFILE_DIR'], hadoop.username.replace(".", "/")),
hadoop.username.replace(".", "_"))
def __init__(self, crawler):
self.crawler = crawler
self.job_path = crawler.settings.get('JOB_PATH')
if not self.job_path:
raise NotConfigured('no JOB_PATH set')
self.json_encoder = ScrapyJSONEncoder()
self.looping_call = LoopingCall(self.store_job_info)
self.looping_call.start(5)
crawler.signals.connect(self.store_job_info,
signal=signals.spider_closed)
class JsonLinesItemExporter(BaseItemExporter):
def __init__(self, file, **kwargs):
self._configure(kwargs, dont_fail=True)
self.file = file
kwargs.setdefault('ensure_ascii', not self.encoding)
self.encoder = ScrapyJSONEncoder(**kwargs)
def export_item(self, item):
itemdict = dict(self._get_serialized_fields(item))
data = self.encoder.encode(itemdict) + '\n'
self.file.write(to_bytes(data, self.encoding))
def __init__(self, file, **kwargs):
self._configure(kwargs, dont_fail=True)
self.file = file
# there is a small difference between the behaviour or JsonItemExporter.indent
# and ScrapyJSONEncoder.indent. ScrapyJSONEncoder.indent=None is needed to prevent
# the addition of newlines everywhere
json_indent = self.indent if self.indent is not None and self.indent > 0 else None
kwargs.setdefault('indent', json_indent)
kwargs.setdefault('ensure_ascii', not self.encoding)
self.encoder = ScrapyJSONEncoder(**kwargs)
self.first_item = True
class JsonLinesItemExporter(BaseItemExporter):
def __init__(self, file, **kwargs):
self._configure(kwargs, dont_fail=True)
self.file = file
kwargs.setdefault('ensure_ascii', not self.encoding)
self.encoder = ScrapyJSONEncoder(**kwargs)
def export_item(self, item):
itemdict = dict(self._get_serialized_fields(item))
data = self.encoder.encode(itemdict) + '\n'
self.file.write(to_bytes(data, self.encoding))
def run(output_path, download_workers):
crawl(download_workers)
reactor.run() # the script will block here until the last crawl call is finished
encoder = ScrapyJSONEncoder()
with open("results/publications.jl", "rb+") as publications_info_file:
with open(output_path, "wb+") as output_file:
output_file.write(b"[")
first = True
while True:
line = publications_info_file.readline()
if not line:
break
if first:
output_file.write(b"\n")
first = False
else:
output_file.write(b",\n")
company_info_data = json.loads(line)
data = encoder.encode(company_info_data)
output_file.write(to_bytes(data))
output_file.write(b"\n]\n")
class LetMeShopApiExporter(BaseItemExporter):
api_end_point = ''
method = 'POST'
def __init__(self, api_base_url, auth_token, *args, **kwargs):
super(LetMeShopApiExporter, self).__init__(*args, export_empty_fields=True, **kwargs)
self.api_base_url = api_base_url
self.encoder = ScrapyJSONEncoder(**kwargs)
self.headers = {'Authorization': 'Token %s' % auth_token}
def _fill_missing_fields(self, item, default_value=None):
if self.fields_to_export is None:
missing_keys = frozenset(item.fields.iterkeys()).difference(item.iterkeys())
else:
missing_keys = frozenset(self.fields_to_export).difference(item.iterkeys())
for missing_key in missing_keys:
item[missing_key] = item.fields[missing_key].get('default_value', default_value)
return item
def _get_serialized_fields(self, item, default_value=None, include_empty=None):
if include_empty is None:
include_empty = self.export_empty_fields
if include_empty:
item = self._fill_missing_fields(item, default_value)
return super(LetMeShopApiExporter, self)._get_serialized_fields(item, default_value, include_empty)
@property
def request_url(self):
return urljoin(self.api_base_url, self.api_end_point)
def export_item(self, item_or_items):
if isinstance(item_or_items, (list, tuple)):
item_list = item_or_items
serialized = [dict(self._get_serialized_fields(item)) for item in item_list]
else:
item = item_or_items
serialized = dict(self._get_serialized_fields(item))
serialized = snake_case_to_camel_case(serialized)
payload = self.encoder.encode(serialized)
r = requests.request(self.method, self.request_url, data=payload, headers=self.headers)
r.raise_for_status()
def start_exporting(self):
pass
def finish_exporting(self):
pass
class KafkaPipeline(object):
"""
Publishes a serialized item into a Kafka topic
:param producer: The Kafka producer
:type producer: kafka.producer.Producer
:param topic: The Kafka topic being used
:type topic: str or unicode
"""
def __init__(self, producer, topic):
"""
:type producer: kafka.producer.Producer
:type topic: str or unicode
"""
self.producer = producer
self.topic = topic
self.encoder = ScrapyJSONEncoder()
def process_item(self, item, spider):
"""
Overriden method to process the item
:param item: Item being passed
:type item: scrapy.item.Item
:param spider: The current spider being used
:type spider: scrapy.spider.Spider
"""
# put spider name in item
item = dict(item)
item['spider'] = spider.name
msg = self.encoder.encode(item)
self.producer.send_messages(self.topic, msg)
@classmethod
def from_settings(cls, settings):
"""
:param settings: the current Scrapy settings
:type settings: scrapy.settings.Settings
:rtype: A :class:`~KafkaPipeline` instance
"""
k_hosts = settings.get('SCRAPY_KAFKA_HOSTS', ['localhost:9092'])
topic = settings.get('SCRAPY_KAFKA_ITEM_PIPELINE_TOPIC', 'scrapy_kafka_item')
kafka = KafkaClient(k_hosts)
conn = SimpleProducer(kafka)
return cls(conn, topic)
class AdSpiderBase(RedisSpider):
task_encoder = ScrapyJSONEncoder().encode
task_decoder = ScrapyJSONDecoder().decode
def next_request(self):
serialized_task = self.server.rpop(self.redis_key)
if serialized_task:
self.logger.info("Got task {}".format(serialized_task))
return self.make_request_from_task(serialized_task, callback=self.parse, dont_filter=False)
@staticmethod
def tagfilter(tag):
return isinstance(tag, Tag)
def parse(self, response):
items = []
for row in response.xpath('//table[@class="TblDataRecs"]/tr'):
item = mylanguageexchangeItem()
name = row.xpath('td[@class="userdata"]//a//b/text()').extract()
item["name"] = [x.strip() for x in name]
country = row.xpath('td[@class="userdata"][@data-th="Country(City)"]//tr[1]/td/text()').extract()
item["country"] = [x.strip() for x in country]
city = row.xpath('td[@class="userdata"][@data-th="Country(City)"]//tr[2]/td/text()').extract()
item["city"] = [x.strip().strip('()') for x in city]
native = row.xpath('td[@class="userdata"][@data-th="Native Language"]//td/text()').extract()
item["native"] = [x.strip() for x in native]
practicing = row.xpath('td[@class="userdata"][@data-th="Practicing Language"]//td/text()').extract()
item["practicing"] = [x.strip() for x in practicing]
desc = row.xpath('td[@class="userdata"][@data-th="Description"]//td/text()').extract()
item["desc"] = [x.strip() for x in desc]
items.append(item)
_encoder = ScrapyJSONEncoder()
with open('mylanguageexchange_crawled.json', 'w') as outfile:
outfile.write(_encoder.encode(items))
def __init__(self):
"""
Connect to Shopware REST Api using HTTP digest authentication.
We need an ADMIN role with sufficient access to insert articles.
Shopware4 (german) API Guide: http://wiki.shopware.de/_detail_861_487.html
"""
self.name = settings['SHOPWARE_SERVICE_NAME']
self.api_url = settings['SHOPWARE_API_BASE']
self.access_token = settings['SHOPWARE_TOKEN_KEY']
self.request_headers = {'Content-Type': 'application/json; charset=utf-8', 'Accept': 'application/json'}
self.encoder = ScrapyJSONEncoder()
self.node = {}
# shopware minimal default item
self.default_item = RestItem({
'taxId': 1,
#'tax': 19,
'name': 'nexus',
'mainDetail': {
'number': 'nex24',
'prices': [{
'customerGroupKey': 'EK',
'basePrice': 16,
'price': 20, # shop will add VAT (if configured that way)
}],
# 'attribute': {
# 'supplier_url': 'http://example.net',
# 'supplierUrl': 'http://example.net',
# # 'attr19': 'http://example.net',
# },
},
'active': True,
'supplier': 'example com',
'categories': [
{'id': 5,},
{'id': 3,},
],
'images': [{
#'id': '1', ## this one is bugged in shopware (doesnt add image to article)
#'mediaId': '1',
# needs deduplication on update
'link': 'http://shopware.local/templates/_emotion/frontend/_resources/images/logo.jpg',
}],
'attribute': {
'attr19': 'http://example.net',
},
'description': 'Some Article',
'descriptionLong': 'Some Article Description',
})
class KafkaPipeline(object):
"""
Publishes a serialized item into a Kafka topic
:param producer: The Kafka producer
:type producer: kafka.producer.Producer
:param topic: The Kafka topic being used
:type topic: str or unicode
"""
def __init__(self, producer, topic):
"""
:type producer: kafka.producer.Producer
:type topic: str or unicode
"""
self.producer = producer
self.topic = topic
self.encoder = ScrapyJSONEncoder()
def process_item(self, item, spider):
"""
Overriden method to process the item
:param item: Item being passed
:type item: scrapy.item.Item
:param spider: The current spider being used
:type spider: scrapy.spider.Spider
"""
# put spider name in item
item = dict(item)
item['spider'] = spider.name
msg = self.encoder.encode(item)
self.producer.send_messages(self.topic, msg)
@classmethod
def from_settings(cls, settings):
"""
:param settings: the current Scrapy settings
:type settings: scrapy.settings.Settings
:rtype: A :class:`~KafkaPipeline` instance
"""
k_hosts = settings.get('SCRAPY_KAFKA_HOSTS', ['localhost:9092'])
topic = settings.get('SCRAPY_KAFKA_ITEM_PIPELINE_TOPIC',
'scrapy_kafka_item')
kafka = KafkaClient(k_hosts)
conn = SimpleProducer(kafka)
return cls(conn, topic)
def __init__(self, host, port, user, password, virtual_host, exchange,
routing_key, queue):
self.host = host
self.port = port
self.user = user
self.password = password
self.virtual_host = virtual_host
credentials = pika.PlainCredentials(self.user, self.password)
parameters = pika.ConnectionParameters(self.host, self.port,
self.virtual_host, credentials)
self.connection = pika.BlockingConnection(parameters=parameters)
self.channel = self.connection.channel()
self.exchange = exchange
self.routing_key = routing_key
self.queue = queue
self.channel.exchange_declare(exchange=exchange,
exchange_type="direct",
durable=True)
self.channel.queue_declare(queue=queue, durable=True)
self.channel.queue_bind(exchange=exchange,
routing_key=routing_key,
queue=queue)
self.encoder = ScrapyJSONEncoder()
class RabbitMQItemPublisherPipeline(object):
def __init__(self, host, port, user, password, virtual_host, exchange,
routing_key, queue):
self.host = host
self.port = port
self.user = user
self.password = password
self.virtual_host = virtual_host
credentials = pika.PlainCredentials(self.user, self.password)
parameters = pika.ConnectionParameters(self.host, self.port,
self.virtual_host, credentials)
self.connection = pika.BlockingConnection(parameters=parameters)
self.channel = self.connection.channel()
self.exchange = exchange
self.routing_key = routing_key
self.queue = queue
self.channel.exchange_declare(exchange=exchange,
exchange_type="direct",
durable=True)
self.channel.queue_declare(queue=queue, durable=True)
self.channel.queue_bind(exchange=exchange,
routing_key=routing_key,
queue=queue)
self.encoder = ScrapyJSONEncoder()
@classmethod
def from_crawler(cls, crawler):
return cls(
host=crawler.settings.get("RABBITMQ_HOST"),
port=crawler.settings.get("RABBITMQ_PORT"),
user=crawler.settings.get("RABBITMQ_USER"),
password=crawler.settings.get("RABBITMQ_PASSWORD"),
virtual_host=crawler.settings.get("RABBITMQ_VIRTUAL_HOST"),
exchange=crawler.settings.get("RABBITMQ_EXCHANGE"),
routing_key=crawler.settings.get("RABBITMQ_ROUTING_KEY"),
queue=crawler.settings.get("RABBITMQ_QUEUE"),
)
def close_spider(self, spider):
self.channel.close()
self.connection.close()
def process_item(self, item, spider):
data = self.encoder.encode(item)
self.channel.basic_publish(
exchange=self.exchange,
routing_key=self.routing_key,
body=data,
)
return item
class UnicodeJsonLinesItemExporter(BaseItemExporter):
""" Allows exporting to JSON directly as Unicode. """
def __init__(self, file, **kwargs):
self._configure(kwargs)
self.file = file
kwargs["ensure_ascii"] = False
self.encoder = ScrapyJSONEncoder(**kwargs)
def export_item(self, item):
itemdict = dict(self._get_serialized_fields(item))
self.file.write(self.encoder.encode(itemdict) + u"\n")
def serialize_field(self, field, name, value):
return value # DON'T call super version, this encodes the Unicode.
class JsonLinesItemExporter(BaseItemExporter):
def __init__(self, file, **kwargs):
super().__init__(dont_fail=True, **kwargs)
self.file = file
self._kwargs.setdefault('ensure_ascii', not self.encoding)
self.encoder = ScrapyJSONEncoder(
**self._kwargs) #就是处理各种数据类型的 比如时间 集合这种json自己处理不了的
def export_item(self, item):
#简单解释 将选出的项用serializer序列化后变成dict
itemdict = dict(self._get_serialized_fields(item))
data = self.encoder.encode(itemdict) + '\n'
#写到文件里
self.file.write(to_bytes(data, self.encoding))
def process_item(self, item, spider):
due_date = item['due_date']
last_update = item['last_update']
data_file_name = './data/' + due_date.strftime('%y%m%d') + '_' + \
last_update.strftime('%y%m%d') + '.json'
os.makedirs(os.path.dirname(data_file_name), exist_ok=True)
with open(data_file_name, 'w+') as f:
f.write(ScrapyJSONEncoder().encode(item['provinces']))
print('已保存到数据文件:', data_file_name)
return item
def process_item(self, item, spider):
""" Handle items. Send items to RabbitMQ.
:param item:
:param spider:
:return:
"""
self.rbmq_conn.send_message(channel=self.publisher,
message=ScrapyJSONEncoder().encode(
dict(item)),
exchange_name=self.exchange_name,
routing_key=self.routing_key)
logger.debug(f'Item scraped')
return item
class JsonEncoderTestCase(unittest.TestCase):
def setUp(self):
self.encoder = ScrapyJSONEncoder()
def test_encode_decode(self):
dt = datetime.datetime(2010, 1, 2, 10, 11, 12)
dts = "2010-01-02 10:11:12"
d = datetime.date(2010, 1, 2)
ds = "2010-01-02"
t = datetime.time(10, 11, 12)
ts = "10:11:12"
dec = Decimal("1000.12")
decs = "1000.12"
s = {'foo'}
ss = ['foo']
dt_set = {dt}
dt_sets = [dts]
for input, output in [('foo', 'foo'), (d, ds), (t, ts), (dt, dts),
(dec, decs), (['foo', d], ['foo', ds]), (s, ss),
(dt_set, dt_sets)]:
self.assertEqual(self.encoder.encode(input), json.dumps(output))
def test_encode_deferred(self):
self.assertIn('Deferred', self.encoder.encode(defer.Deferred()))
def test_encode_request(self):
r = Request("http://www.example.com/lala")
rs = self.encoder.encode(r)
self.assertIn(r.method, rs)
self.assertIn(r.url, rs)
def test_encode_response(self):
r = Response("http://www.example.com/lala")
rs = self.encoder.encode(r)
self.assertIn(r.url, rs)
self.assertIn(str(r.status), rs)
class JsonLinesItemSplitFileExporter(BaseItemExporter):
"""An item exporter to organize json lines into separate folders.
Attributes:
_configure (func): Uses to configure the Item Exporter by setting the options dictionary.
encoder (ScrapyJSONEncoder): Encoder used to convert scrapy items into a json format line.
"""
def __init__(self, **kwargs):
"""Initialize the configuration dictionary and encoder.
Args:
**kwargs: Arbitrary keyword arguments for the options dictionary.
"""
# If dont_fail is set, it won't raise an exception on unexpected options
self._configure(kwargs, dont_fail=True)
kwargs.setdefault('ensure_ascii', not self.encoding)
self.encoder = ScrapyJSONEncoder()
super(JsonLinesItemSplitFileExporter, self).__init__()
def export_item(self, item):
"""Export Scrapy items to specific files based on the article_type.
Args:
item (scrapy.Item): A Scrapy item that contains a complete scraped information for an article/product.
"""
# Serialize the item, and perform encoding to create a python dictionary
item_dict = dict(self._get_serialized_fields(item))
data = self.encoder.encode(item_dict) + os.linesep
# If there is only one item in article_type, then the path (folders) would just be
# scraped_data/spider.name/article_type. Otherwise we would combine all the article_type list except the last
# item into a path, such as scraped_data/spider.name/article_type[0]/article_type[1], then the item would be
# a json line placed in scraped_data/spider.name/article_type[0]/article_type[1]/article_type[2].jl.
if len(item['article_type']) == 1:
path = os.path.join("scraped_data", item["spider_name"])
item_path = os.path.join(path, item['article_type'][0]) + ".jl"
else:
path = os.path.join(os.path.join("scraped_data", item["spider_name"]),
(os.path.join(*item['article_type'][:-1])))
item_path = os.path.join(path, item['article_type'][-1]) + ".jl"
if not os.path.exists(path):
os.makedirs(path)
# Write in append and byte mode
open(item_path, 'a+b').write(to_bytes(data, self.encoding))
class SortedJsonItemExporter(BaseItemExporter):
def __init__(self, file, **kwargs):
super().__init__(dont_fail=True, **kwargs)
self.file = file
self._kwargs.setdefault('indent', 4)
self._kwargs.setdefault('ensure_ascii', not self.encoding)
self.encoder = ScrapyJSONEncoder(**self._kwargs)
self.items = []
def export_item(self, item):
self.items.append(dict(self._get_serialized_fields(item)))
def finish_exporting(self):
data = self.encoder.encode(sorted(self.items, key=sort_key))
self.file.write(to_bytes(data, self.encoding))
class KafkaPipeline(object):
def __init__(self, producer, topic):
self.producer = producer
self.topic = topic
self.encoder = ScrapyJSONEncoder()
def process_item(self, item, spider):
item = dict(item)
item['spider'] = spider.name
msg = self.encoder.encode(item)
self.producer.send_message(self.topic, msg)
@classmethod
def from_settings(cls, settings)
class KafkaProducerPipeline(object):
def __init__(self, kafka_bootstrap_server):
self.kafka_bootstrap_server = []
self.kafka_bootstrap_server.append(kafka_bootstrap_server)
self.collection_name = 'articles'
self.encoder = ScrapyJSONEncoder()
@classmethod
def from_crawler(cls, crawler):
# pull in information from settings.py
return cls(kafka_bootstrap_server=crawler.settings.get(
'KAFKA_BOOTSTRAP_SERVER'), )
def open_spider(self, spider):
print("spider name: ", spider.name)
# initializing py-Kafka producer
self.producer = KafkaProducer(
bootstrap_servers=self.kafka_bootstrap_server)
print("kafka_bootstrap_server: ", self.kafka_bootstrap_server)
if hasattr(spider, 'collection_name'):
print("spider collection_name: ", spider.collection_name)
self.collection_name = spider.collection_name
def close_spider(self, spider):
# clean up when spider is closed
self.producer.flush(timeout=60)
self.producer.close(timeout=60)
def process_item(self, item, spider):
valid = True
for data in item:
if not data:
valid = False
raise DropItem("Missing {0}!".format(data))
if valid:
print("valid - inside process_item...", item['source'], ': ',
item['headline'])
# self.producer.send('articles', self.encoder.encode(item).encode())
key = str(ord(item['source'][0])) + str(ord(item['source'][1]))
self.producer.send('articles',
value=self.encoder.encode(item).encode(),
key=key.encode())
self.index += 1
logging.debug("News item sent by Kafka Producer!")
return item
class ScrapyKafkaTopicWriter(KafkaTopicWriter):
""" Kafka writer which knows how to handle Scrapy items: they are
serialized to JSON, and "_id" field is used as Kafka key if present.
"""
def __init__(self, *args, **kwargs):
self._encoder = ScrapyJSONEncoder()
self._exporter = PythonItemExporter(binary=False)
kwargs.setdefault('value_serializer', self._serialize_value)
super(ScrapyKafkaTopicWriter, self).__init__(*args, **kwargs)
def write_item(self, item):
key = item.get('_id', None)
msg = self._exporter.export_item(item)
return self.write(key, msg)
def _serialize_value(self, value):
return self._encoder.encode(value).encode('utf8')
class AddItemPipeline(object):
""" Pushes serialized item into a RQ """
def __init__(self, host, port, db, queue_name, store_id):
self.encoder = ScrapyJSONEncoder()
self.store_id = store_id
self.queue_name = queue_name
self.server = redis.Redis(host, port, db)
self.queue = rq.Queue(queue_name, connection=self.server)
@classmethod
def from_settings(cls, settings):
host = settings.get('REDIS_HOST', 'localhost')
port = settings.get('REDIS_PORT', 6379)
db = settings.get('REDIS_DB', 0)
queue_name = settings.get('RQ_QUEUE', 'default')
store_id = int(settings.get('STORE', 0))
return cls(host, port, db, queue_name, store_id)
@classmethod
def from_crawler(cls, crawler):
return cls.from_settings(crawler.settings)
def process_item(self, item, spider):
return deferToThread(self._process_item, item, spider)
def _process_item(self, item, spider):
''' ''' # {{{
## get global Store URL mapping
store_id = self.store_id
if store_id is 0:
raise DropItem('Not set the store and no Store URL mapping')
## assign queue parameters
item['store_id'] = store_id
callback = 'worker.save_product_to_db'
event = self.encoder.encode(
dict(queue=self.queue_name, value=item, time=time.time()))
## push item to redis queue
self.queue.enqueue(callback, event)
return item
class AddItemPipeline(object):
""" Pushes serialized item into a RQ """
def __init__(self, host, port, db, queue_name, store_id):
self.encoder = ScrapyJSONEncoder()
self.store_id = store_id
self.queue_name = queue_name
self.server = redis.Redis(host, port, db)
self.queue = rq.Queue(queue_name, connection=self.server)
@classmethod
def from_settings(cls, settings):
host = settings.get('REDIS_HOST', 'localhost')
port = settings.get('REDIS_PORT', 6379)
db = settings.get('REDIS_DB', 0)
queue_name = settings.get('RQ_QUEUE', 'default')
store_id = int(settings.get('STORE', 0))
return cls(host, port, db, queue_name, store_id)
@classmethod
def from_crawler(cls, crawler):
return cls.from_settings(crawler.settings)
def process_item(self, item, spider):
return deferToThread(self._process_item, item, spider)
def _process_item(self, item, spider):
''' ''' # {{{
## get global Store URL mapping
store_id = self.store_id
if store_id is 0:
raise DropItem('Not set the store and no Store URL mapping')
## assign queue parameters
item['store_id'] = store_id
callback = 'worker.save_product_to_db'
event = self.encoder.encode(dict(queue=self.queue_name, value=item, time=time.time()))
## push item to redis queue
self.queue.enqueue(callback, event)
return item
class RedisListPipeline:
DEFAULT_QUEUE = 'queue'
DEFAULT_MAX_RETRY = 5
serializer = ScrapyJSONEncoder().encode
def __init__(self, conn_url: str, queue: str, max_retry=None):
try:
import redis
except ImportError:
raise NotConfigured('missing redis library')
self._conn = redis.from_url(conn_url)
self.queue = queue or self.DEFAULT_QUEUE
self.max_retry = max_retry or self.DEFAULT_MAX_RETRY
@classmethod
def from_crawler(cls, crawler):
if hasattr(crawler.spider, 'queue'):
queue = crawler.spider.queue
else:
queue = crawler.settings.get('REDIS_DEFAULT_QUEUE')
return cls(
conn_url=crawler.settings.get('REDIS_CONNECTION_URL'),
queue=queue,
max_retry=crawler.settings.get('REDIS_MAX_RETRY')
)
def process_item(self, item, spider):
data = self.serializer(item)
try_time = 0
while try_time < self.max_retry:
try:
self._conn.rpush(self.queue, data)
return item
except Exception:
spider.logger.error('process item failed {}'.format(item))
try_time += 1
spider.logger.error('Give up item for failed {} times {}'.format(try_time, item))
return item
def close(self):
self._conn.close()
class CustomJsonItemExporter(BaseItemExporter):
def __init__(self, file, **kwargs):
self._configure(kwargs, dont_fail=True)
self.file = file
kwargs.setdefault("ensure_ascii", False)
self.encoder = ScrapyJSONEncoder(**kwargs)
self.first_item = True
def start_exporting(self):
self.file.write(b'')
def finish_exporting(self):
self.file.write(b'\n')
def export_item(self, item):
itemdict = dict(self._get_serialized_fields(item))
bytes = to_bytes(self.encoder.encode(itemdict))
self.file.write(bytes)
self.file.write(b"\n")
class SendToBrokerPipeline(object):
def __init__(self):
self.publisher = Publisher('data_distributor')
self.encoder = ScrapyJSONEncoder()
def process_item(self, item, spider):
#Runs sending broker in separate thread to prevent it from blocking
#on single items
return deferToThread(self._process_item, item, spider)
def _process_item(self, item, spider):
item_dict = dict(item)
data = self.encoder.encode(item_dict)
self.publisher.send_message(data,'articles')
return item
class UnicodeJsonLinesItemExporter(BaseItemExporter):
logger = logging.getLogger('UnicodeJsonLinesLogging')
def __init__(self, file, **kwargs):
self._configure(kwargs, dont_fail=True)
self.file = file
self.encoder = ScrapyJSONEncoder(ensure_ascii=False,
encoding='UTF-8',
**kwargs)
def export_item(self, item):
itemdict = dict(self._get_serialized_fields(item))
data = self.encoder.encode(itemdict) + '\n'
self.logger.info('==============')
self.file.write(to_bytes(data, self.encoding))
def finish_exporting(self):
pass
class BasePipeline(object):
"""
"""
def __init__(self):
self.headers = {'Content-Type': 'application/json'}
self._encoder = ScrapyJSONEncoder()
def _item_to_json(self, obj):
"""
Utility that takes a object, serializes it using ScrapyJSONEncoder
and returns a deserialized version of the data.
We use this to convert Scrapy models to JSON objects we can to requests.
:param obj:
:return dict:
"""
return json.loads(self._encoder.encode(obj))
class JsonItemExporter(JsonLinesItemExporter):
def __init__(self, file, **kwargs):
self._configure(kwargs, dont_fail=True)
self.file = file
self.encoder = ScrapyJSONEncoder(**kwargs)
self.first_item = True
def start_exporting(self):
self.file.write("[")
def finish_exporting(self):
self.file.write("]")
def export_item(self, item):
if self.first_item:
self.first_item = False
else:
self.file.write(',\n')
itemdict = dict(self._get_serialized_fields(item))
self.file.write(self.encoder.encode(itemdict))
class JsonItemExporter(BaseItemExporter):
def __init__(self, file, **kwargs):
self._configure(kwargs, dont_fail=True)
self.file = file
self.encoder = ScrapyJSONEncoder(**kwargs)
self.first_item = True
def start_exporting(self):
self.file.write(b"[")
def finish_exporting(self):
self.file.write(b"]")
def export_item(self, item):
if self.first_item:
self.first_item = False
else:
self.file.write(b",\n")
itemdict = dict(self._get_serialized_fields(item))
self.file.write(to_bytes(self.encoder.encode(itemdict)))
class GeoJsonItemExporter(BaseItemExporter):
def __init__(self, file, **kwargs):
self._configure(kwargs, dont_fail=True)
self.file = file
kwargs.setdefault('ensure_ascii', not self.encoding)
self.encoder = ScrapyJSONEncoder(**kwargs)
self.first_item = True
def start_exporting(self):
self.file.write(b'{ "type": "FeatureCollection","features":[\n')
def finish_exporting(self):
self.file.write(b"\n]}")
def export_item(self, item):
if self.first_item:
self.first_item = False
else:
self.file.write(b',\n')
data = self.encoder.encode(item)
self.file.write(to_bytes(data, self.encoding))
class JsonItemExporter(JsonLinesItemExporter):
def __init__(self, file, **kwargs):
self._configure(kwargs)
self.file = file
self.encoder = ScrapyJSONEncoder(**kwargs)
self.first_item = True
def start_exporting(self):
self.file.write("[")
def finish_exporting(self):
self.file.write("]")
def export_item(self, item):
if self.first_item:
self.first_item = False
else:
self.file.write(',\n')
itemdict = dict(self._get_serialized_fields(item))
self.file.write(self.encoder.encode(itemdict))
class FanItemExporter(BaseItemExporter):
def __init__(self, file, **kwargs):
self._configure(kwargs, dont_fail=True)
self.file = file
self.encoder = ScrapyJSONEncoder(**kwargs)
self.first_item = True
def start_exporting(self):
self.file.write(b'{"product": [')
def finish_exporting(self):
self.file.write(b'\n]}')
def export_item(self, item):
if self.first_item:
self.first_item = False
else:
self.file.write(b',\n')
itemdict = dict(self._get_serialized_fields(item))
self.file.write(to_bytes(self.encoder.encode(itemdict)))
class JsonItemExporter(BaseItemExporter):
def __init__(self, file, **kwargs):
self._configure(kwargs, dont_fail=True)
self.file = file
kwargs.setdefault('ensure_ascii', not self.encoding)
self.encoder = ScrapyJSONEncoder(**kwargs)
self.first_item = True
def start_exporting(self):
self.file.write(b"[\n")
def finish_exporting(self):
self.file.write(b"\n]")
def export_item(self, item):
if self.first_item:
self.first_item = False
else:
self.file.write(b',\n')
itemdict = dict(self._get_serialized_fields(item))
data = self.encoder.encode(itemdict)
self.file.write(to_bytes(data, self.encoding))
class DockerhubExtension(object):
@classmethod
def from_crawler(cls, crawler):
return cls(crawler)
def __init__(self, crawler):
self.crawler = crawler
self.job_path = crawler.settings.get('JOB_PATH')
if not self.job_path:
raise NotConfigured('no JOB_PATH set')
self.json_encoder = ScrapyJSONEncoder()
self.looping_call = LoopingCall(self.store_job_info)
self.looping_call.start(5)
crawler.signals.connect(self.store_job_info,
signal=signals.spider_closed)
def store_job_info(self):
with open(self.job_path, 'w') as f:
stats = self.crawler.stats.get_stats()
job_info = {'stats': stats}
job_info_json = self.json_encoder.encode(job_info)
f.write(job_info_json)
class JsonItemExporter(BaseItemExporter):
def __init__(self, file, **kwargs):
self._configure(kwargs, dont_fail=True)
self.file = file
kwargs.setdefault('ensure_ascii', not self.encoding)
self.encoder = ScrapyJSONEncoder(**kwargs)
self.first_item = True
def start_exporting(self):
self.file.write(b"[\n")
def finish_exporting(self):
self.file.write(b"\n]")
def export_item(self, item):
if self.first_item:
self.first_item = False
else:
self.file.write(b',\n')
itemdict = dict(self._get_serialized_fields(item))
data = self.encoder.encode(itemdict)
self.file.write(to_bytes(data, self.encoding))
class KafkaPipeline:
# https://github.com/dfdeshom/scrapy-kafka/blob/master/scrapy_kafka/pipelines.py
def __init__(self, producer, topic):
self.producer = producer
self.topic = topic
self.encoder = ScrapyJSONEncoder()
def process_item(self, item, spider):
item = dict(item)
item['spider'] = spider.name
msg = self.encoder.encode(item)
spider.log(msg)
self.producer.send(self.topic, msg.encode('utf-8'))
spider.log("Sent to kafka.")
@classmethod
def from_settings(cls, settings):
k_hosts = settings.get('KAFKA_HOSTS', ['localhost:9092'])
topic = settings.get('KAFKA_ITEM_PIPELINE_TOPIC', 'scrapy_kafka_item')
prod = KafkaProducer(bootstrap_servers=k_hosts)
return cls(prod, topic)
class KafkaPipeline:
def __init__(self, producer, topic):
self.p = producer
self.topic = topic
self.encoder = ScrapyJSONEncoder()
@classmethod
def from_settings(cls, settings):
"""
:param settings: the current Scrapy settings
:type settings: scrapy.settings.Settings
:rtype: A :class:`~KafkaPipeline` instance
"""
brokers = settings.get('KAFKA_PRODUCER_BROKERS')
topic = settings.get('KAFKA_PRODUCER_TOPIC')
producer = Producer({
'bootstrap.servers': brokers,
})
return cls(producer, topic)
def process_item(self, item, spider):
msg = self.encoder.encode(item)
self.p.produce(self.topic, msg, callback=delivery_report)
