bhfutils 0.2.3__tar.gz → 0.2.6__tar.gz

{bhfutils-0.2.3 → bhfutils-0.2.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bhfutils
-Version: 0.2.3
+Version: 0.2.6
 Summary: Utilities that are used by any spider of Behoof project
 Home-page: https://behoof.app/
 Author: Teplygin Vladimir

{bhfutils-0.2.3 → bhfutils-0.2.6}/bhfutils/crawler/meta_passthrough_middleware.py

@@ -9,9 +9,9 @@ class MetaPassthroughMiddleware(object):
         self.setup(settings)
 
     def setup(self, settings):
-        '''
+        """
         Does the actual setup of the middleware
-        '''
+        """
         # set up the default sc logger
         my_level = settings.get('SC_LOG_LEVEL', 'INFO')
         my_name = settings.get('SC_LOGGER_NAME', 'sc-logger')
@@ -35,19 +35,31 @@ class MetaPassthroughMiddleware(object):
     def from_crawler(cls, crawler):
         return cls(crawler.settings)
 
+    @staticmethod
+    def _passthrough(response, x):
+        # only operate on requests: pass along all known meta fields, but only
+        # if they were not already set in the spider's new request
+        if isinstance(x, Request):
+            for key in list(response.meta.keys()):
+                if key not in x.meta and key != 'playwright_page_methods':
+                    x.meta[key] = response.meta[key]
+        return x
+
     def process_spider_output(self, response, result, spider):
-        '''
+        """
         Ensures the meta data from the response is passed
-        through in any Request's generated from the spider
-        '''
+        through in any Request's generated from the spider (sync path).
+        """
         self.logger.debug("processing meta passthrough middleware")
         for x in result:
-            # only operate on requests
-            if isinstance(x, Request):
-                self.logger.debug("found request")
-                # pass along all known meta fields, only if
-                # they were not already set in the spider's new request
-                for key in list(response.meta.keys()):
-                    if key not in x.meta and key != 'playwright_page_methods':
-                        x.meta[key] = response.meta[key]
-            yield x
+            yield self._passthrough(response, x)
+
+    async def process_spider_output_async(self, response, result, spider):
+        """
+        Async-generator variant required by Scrapy >= 2.13 when the spider
+        output is an async iterable (AsyncCrawlerProcess). Defining both makes
+        this a "universal" spider middleware compatible with old + new Scrapy.
+        """
+        self.logger.debug("processing meta passthrough middleware (async)")
+        async for x in result:
+            yield self._passthrough(response, x)

{bhfutils-0.2.3 → bhfutils-0.2.6}/bhfutils/crawler/pipelines.py

@@ -109,13 +109,30 @@ class KafkaPipeline(object):
                                          bytes=my_bytes,
                                          backups=my_backups)
 
+        # Security settings are optional: a spider running inside the VPC talks to the
+        # PLAINTEXT listener and defines none of them, while one running outside needs
+        # SASL_SSL. Defaults keep existing projects working without touching them.
+        producer_kwargs = dict(bootstrap_servers=settings['KAFKA_HOSTS'],
+                               retries=3,
+                               linger_ms=settings['KAFKA_PRODUCER_BATCH_LINGER_MS'],
+                               buffer_memory=settings['KAFKA_PRODUCER_BUFFER_BYTES'],
+                               value_serializer=lambda m: m.encode('utf-8'),
+                               max_request_size=settings['KAFKA_PRODUCER_MAX_REQUEST_SIZE'])
+
+        security_protocol = settings.get('KAFKA_SECURITY_PROTOCOL', 'PLAINTEXT')
+        if security_protocol != 'PLAINTEXT':
+            producer_kwargs['security_protocol'] = security_protocol
+        if 'SASL' in security_protocol:
+            producer_kwargs['sasl_mechanism'] = settings.get('KAFKA_SASL_MECHANISM', 'PLAIN')
+            producer_kwargs['sasl_plain_username'] = settings.get('KAFKA_SASL_USERNAME')
+            producer_kwargs['sasl_plain_password'] = settings.get('KAFKA_SASL_PASSWORD')
+        if 'SSL' in security_protocol:
+            # Left as None for a certificate issued by a public CA: kafka-python then
+            # falls back to the system trust store. Set it only for a private CA.
+            producer_kwargs['ssl_cafile'] = settings.get('KAFKA_SSL_CAFILE')
+
         try:
-            producer = KafkaProducer(bootstrap_servers=settings['KAFKA_HOSTS'],
-                                     retries=3,
-                                     linger_ms=settings['KAFKA_PRODUCER_BATCH_LINGER_MS'],
-                                     buffer_memory=settings['KAFKA_PRODUCER_BUFFER_BYTES'],
-                                     value_serializer=lambda m: m.encode('utf-8'),
-                                     max_request_size=settings['KAFKA_PRODUCER_MAX_REQUEST_SIZE'])
+            producer = KafkaProducer(**producer_kwargs)
         except Exception as e:
             logger.error("Unable to connect to Kafka in Pipeline"", raising exit flag.")
             # this is critical so we choose to exit.

{bhfutils-0.2.3 → bhfutils-0.2.6}/bhfutils/crawler/settings_template.py

@@ -22,6 +22,16 @@ KAFKA_PRODUCER_BATCH_LINGER_MS = 25  # 25 ms before flush
 KAFKA_PRODUCER_BUFFER_BYTES = 4 * 1024 * 1024  # 4MB before blocking
 KAFKA_PRODUCER_MAX_REQUEST_SIZE = 1024 * 1024 # 1MB
 
+# Security for the Kafka connection. PLAINTEXT is right for a spider running inside
+# the VPC, which reaches the broker's INTERNAL listener. A spider running anywhere
+# else must go through the EXTERNAL listener and set SASL_SSL plus credentials.
+KAFKA_SECURITY_PROTOCOL = 'PLAINTEXT'
+KAFKA_SASL_MECHANISM = 'PLAIN'
+KAFKA_SASL_USERNAME = None
+KAFKA_SASL_PASSWORD = None
+# Only needed for a certificate from a private CA; leave None for a public one.
+KAFKA_SSL_CAFILE = None
+
 ZOOKEEPER_ASSIGN_PATH = '/scrapy-cluster/crawler/'
 ZOOKEEPER_ID = 'all'
 ZOOKEEPER_HOSTS = 'localhost:2181'

{bhfutils-0.2.3 → bhfutils-0.2.6}/bhfutils.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bhfutils
-Version: 0.2.3
+Version: 0.2.6
 Summary: Utilities that are used by any spider of Behoof project
 Home-page: https://behoof.app/
 Author: Teplygin Vladimir

{bhfutils-0.2.3 → bhfutils-0.2.6}/setup.py

@@ -43,7 +43,7 @@ if 'nosetests' in sys.argv[1:]:
 
 setup(
     name='bhfutils',
-    version='0.2.3',
+    version='0.2.6',
     description='Utilities that are used by any spider of Behoof project',
     long_description=readme(),
     long_description_content_type='text/x-rst',