{"id":1128,"date":"2022-06-22T14:14:22","date_gmt":"2022-06-22T06:14:22","guid":{"rendered":"http:\/\/175.24.184.226\/?p=1128"},"modified":"2022-06-22T14:17:22","modified_gmt":"2022-06-22T06:17:22","slug":"python%e6%93%8d%e4%bd%9ckafka%e7%9a%84%e9%80%9a%e4%bf%97%e6%80%bb%e7%bb%93%ef%bc%88kafka-python%ef%bc%89","status":"publish","type":"post","link":"\/?p=1128","title":{"rendered":"Python\u64cd\u4f5cKafka\u7684\u901a\u4fd7\u603b\u7ed3\uff08kafka-python\uff09"},"content":{"rendered":"<p>kafka-python\u5b98\u65b9\u6587\u6863\uff1a<br \/>\n<a href=\"https:\/\/kafka-python.readthedocs.io\/en\/master\/apidoc\/KafkaConsumer.html\">https:\/\/kafka-python.readthedocs.io\/en\/master\/apidoc\/KafkaConsumer.html<\/a><\/p>\n<h2>\u4e00\u3001\u57fa\u672c\u6982\u5ff5<\/h2>\n<ul>\n<li>Topic\uff1a\u4e00\u7ec4\u6d88\u606f\u6570\u636e\u7684\u6807\u8bb0\u7b26\uff1b<\/li>\n<li>Producer\uff1a\u751f\u4ea7\u8005\uff0c\u7528\u4e8e\u751f\u4ea7\u6570\u636e\uff0c\u53ef\u5c06\u751f\u4ea7\u540e\u7684\u6d88\u606f\u9001\u5165\u6307\u5b9a\u7684Topic\uff1b<\/li>\n<li>Consumer\uff1a\u6d88\u8d39\u8005\uff0c\u83b7\u53d6\u6570\u636e\uff0c\u53ef\u6d88\u8d39\u6307\u5b9a\u7684Topic\uff1b<\/li>\n<li>Group\uff1a\u6d88\u8d39\u8005\u7ec4\uff0c\u540c\u4e00\u4e2agroup\u53ef\u4ee5\u6709\u591a\u4e2a\u6d88\u8d39\u8005\uff0c\u4e00\u6761\u6d88\u606f\u5728\u4e00\u4e2agroup\u4e2d\uff0c\u53ea\u4f1a\u88ab\u4e00\u4e2a\u6d88\u8d39\u8005\u83b7\u53d6\uff1b<\/li>\n<li>Partition\uff1a\u5206\u533a\uff0c\u4e3a\u4e86\u4fdd\u8bc1kafka\u7684\u541e\u5410\u91cf\uff0c\u4e00\u4e2aTopic\u53ef\u4ee5\u8bbe\u7f6e\u591a\u4e2a\u5206\u533a\u3002\u540c\u4e00\u5206\u533a\u53ea\u80fd\u88ab\u4e00\u4e2a\u6d88\u8d39\u8005\u8ba2\u9605\u3002<br \/>\n<h2>\u4e8c\u3001\u672c\u5730\u5b89\u88c5\u4e0e\u542f\u52a8\uff08\u57fa\u4e8eDocker\uff09<\/h2>\n<h3>1.\u4e0b\u8f7dzookeeper\u955c\u50cf\u4e0ekafka\u955c\u50cf\uff1a<\/h3>\n<pre><code class=\"language-bash\">docker pull wurstmeister\/zookeeper\ndocker pull wurstmeister\/kafka<\/code><\/pre>\n<h3>2. \u672c\u5730\u542f\u52a8zookeeper<\/h3>\n<pre><code class=\"language-bash\">docker run -d --name zookeeper -p 2181:2181 -t wurstmeister\/zookeeper  <\/code><\/pre>\n<h3>3. \u672c\u5730\u542f\u52a8kafka<\/h3>\n<pre><code class=\"language-bash\">docker run -d --name kafka --publish 9092:9092 --link zookeeper \\\n--env KAFKA_ZOOKEEPER_CONNECT=zookeeper:2181 \\\n--env KAFKA_ADVERTISED_HOST_NAME=localhost \\\n--env KAFKA_ADVERTISED_PORT=9092 \\\nwurstmeister\/kafka:latest <\/code><\/pre>\n<p>\u6ce8\u610f\uff1a\u4e0a\u8ff0\u4ee3\u7801\uff0c\u5c06kafka\u542f\u52a8\u57289092\u7aef\u53e3<\/p>\n<\/li>\n<\/ul>\n<h3>4. \u8fdb\u5165kafka bash<\/h3>\n<pre><code class=\"language-bash\">docker exec -it kafka bash\ncd \/opt\/kafka\/bin<\/code><\/pre>\n<h3>5. \u521b\u5efaTopic\uff0c\u5206\u533a\u4e3a2\uff0cTopic name\u4e3a'kafka_demo'<\/h3>\n<pre><code class=\"language-bash\">kafka-topics.sh --create --zookeeper zookeeper:2181 \\\n--replication-factor 1 --partitions 2 --topic kafka_demo<\/code><\/pre>\n<h3>6. \u67e5\u770b\u5f53\u524d\u6240\u6709topic<\/h3>\n<pre><code class=\"language-bash\">kafka-topics.sh --zookeeper zookeeper:2181 --list<\/code><\/pre>\n<h3>7. \u5b89\u88c5kafka-python<\/h3>\n<pre><code class=\"language-bash\">pip install kafka-python<\/code><\/pre>\n<h2>\u4e09\u3001\u751f\u4ea7\u8005\uff08Producer\uff09\u4e0e\u6d88\u8d39\u8005\uff08Consumer\uff09<\/h2>\n<p>\u751f\u4ea7\u8005\u548c\u6d88\u8d39\u8005\u7684\u7b80\u6613Demo\uff0c\u8fd9\u91cc\u4e00\u8d77\u6f14\u793a\uff1a<\/p>\n<pre><code class=\"language-python\">from kafka import KafkaProducer, KafkaConsumer\nfrom kafka.errors import kafka_errors\nimport traceback\nimport json\n\ndef producer_demo():\n    # \u5047\u8bbe\u751f\u4ea7\u7684\u6d88\u606f\u4e3a\u952e\u503c\u5bf9\uff08\u4e0d\u662f\u4e00\u5b9a\u8981\u952e\u503c\u5bf9\uff09\uff0c\u4e14\u5e8f\u5217\u5316\u65b9\u5f0f\u4e3ajson\n    producer = KafkaProducer(\n        bootstrap_servers=[&#039;localhost:9092&#039;], \n        key_serializer=lambda k: json.dumps(k).encode(),\n        value_serializer=lambda v: json.dumps(v).encode())\n    # \u53d1\u9001\u4e09\u6761\u6d88\u606f\n    for i in range(0, 3):\n        future = producer.send(\n            &#039;kafka_demo&#039;,\n            key=&#039;count_num&#039;,  # \u540c\u4e00\u4e2akey\u503c\uff0c\u4f1a\u88ab\u9001\u81f3\u540c\u4e00\u4e2a\u5206\u533a\n            value=str(i),\n            partition=1)  # \u5411\u5206\u533a1\u53d1\u9001\u6d88\u606f\n        print(&quot;send {}&quot;.format(str(i)))\n        try:\n            future.get(timeout=10) # \u76d1\u63a7\u662f\u5426\u53d1\u9001\u6210\u529f           \n        except kafka_errors:  # \u53d1\u9001\u5931\u8d25\u629b\u51fakafka_errors\n            traceback.format_exc()\n\ndef consumer_demo():\n    consumer = KafkaConsumer(\n        &#039;kafka_demo&#039;, \n        bootstrap_servers=&#039;:9092&#039;,\n        group_id=&#039;test&#039;\n    )\n    for message in consumer:\n        print(&quot;receive, key: {}, value: {}&quot;.format(\n            json.loads(message.key.decode()),\n            json.loads(message.value.decode())\n            )\n        )<\/code><\/pre>\n<p>\u8fd9\u91cc\u5efa\u8bae\u8d77\u4e24\u4e2aterminal\uff0c\u6216\u8005\u4e24\u4e2ajupyter notebook\u9875\u9762\u6765\u9a8c\u8bc1\u3002<\/p>\n<p>\u5148\u6267\u884c\u6d88\u8d39\u8005\uff1a<\/p>\n<p><code>consumer_demo()<\/code><br \/>\n\u518d\u6267\u884c\u751f\u4ea7\u8005\uff1a<\/p>\n<p><code>producer_demo()<\/code><br \/>\n\u4f1a\u770b\u5230\u5982\u4e0b\u8f93\u51fa\uff1a<\/p>\n<pre><code>>&gt;&gt; producer_demo()\n>&gt;&gt; send 0\n>&gt;&gt; send 1\n>&gt;&gt; send 2\n\n>&gt;&gt; consumer_demo()\n>&gt;&gt; receive, key: count_num, value: 0\n>&gt;&gt; receive, key: count_num, value: 1\n>&gt;&gt; receive, key: count_num, value: 2<\/code><\/pre>\n<h2>\u56db\u3001\u6d88\u8d39\u8005\u8fdb\u9636\u64cd\u4f5c<\/h2>\n<h3>\uff081\uff09\u521d\u59cb\u5316\u53c2\u6570\uff1a<\/h3>\n<p>\u5217\u4e3e\u4e00\u4e9bKafkaConsumer\u521d\u59cb\u5316\u65f6\u7684\u91cd\u8981\u53c2\u6570\uff1a<\/p>\n<ul>\n<li>\n<p>group_id<br \/>\n\u9ad8\u5e76\u53d1\u91cf\uff0c\u5219\u9700\u8981\u6709\u591a\u4e2a\u6d88\u8d39\u8005\u534f\u4f5c\uff0c\u6d88\u8d39\u8fdb\u5ea6\uff0c\u5219\u7531group_id\u7edf\u4e00\u3002\u4f8b\u5982\u6d88\u8d39\u8005A\u4e0e\u6d88\u8d39\u8005B\uff0c\u5728\u521d\u59cb\u5316\u65f6\u4f7f\u7528\u540c\u4e00\u4e2agroup_id\u3002\u5728\u8fdb\u884c\u6d88\u8d39\u65f6\uff0c\u4e00\u6761\u6d88\u606f\u88ab\u6d88\u8d39\u8005A\u6d88\u8d39\u540e\uff0c\u5728kafka\u4e2d\u4f1a\u88ab\u6807\u8bb0\uff0c\u8fd9\u6761\u6d88\u606f\u4e0d\u4f1a\u518d\u88abB\u6d88\u8d39\uff08\u524d\u63d0\u662fA\u6d88\u8d39\u540e\u6b63\u786ecommit\uff09\u3002<\/p>\n<\/li>\n<li>\n<p>key_deserializer\uff0c value_deserializer<br \/>\n\u4e0e\u751f\u4ea7\u8005\u4e2d\u7684\u53c2\u6570\u4e00\u81f4\uff0c\u81ea\u52a8\u89e3\u6790\u3002<\/p>\n<\/li>\n<li>\n<p>auto_offset_reset<br \/>\n\u6d88\u8d39\u8005\u542f\u52a8\u7684\u65f6\u523b\uff0c\u6d88\u606f\u961f\u5217\u4e2d\u6216\u8bb8\u5df2\u7ecf\u6709\u5806\u79ef\u7684\u672a\u6d88\u8d39\u6d88\u606f\uff0c\u6709\u65f6\u5019\u9700\u6c42\u662f\u4ece\u4e0a\u4e00\u6b21\u672a\u6d88\u8d39\u7684\u4f4d\u7f6e\u5f00\u59cb\u8bfb\uff08\u5219\u8be5\u53c2\u6570\u8bbe\u7f6e\u4e3aearliest\uff09\uff0c\u6709\u65f6\u5019\u7684\u9700\u6c42\u4e3a\u4ece\u5f53\u524d\u65f6\u523b\u5f00\u59cb\u8bfb\u4e4b\u540e\u4ea7\u751f\u7684\uff0c\u4e4b\u524d\u4ea7\u751f\u7684\u6570\u636e\u4e0d\u518d\u6d88\u8d39\uff08\u5219\u8be5\u53c2\u6570\u8bbe\u7f6e\u4e3alatest\uff09\u3002<\/p>\n<\/li>\n<li>\n<p>enable_auto_commit\uff0c auto_commit_interval_ms<br \/>\n\u662f\u5426\u81ea\u52a8commit\uff0c\u5f53\u524d\u6d88\u8d39\u8005\u6d88\u8d39\u5b8c\u8be5\u6570\u636e\u540e\uff0c\u9700\u8981commit\uff0c\u624d\u53ef\u4ee5\u5c06\u6d88\u8d39\u5b8c\u7684\u4fe1\u606f\u4f20\u56de\u6d88\u606f\u961f\u5217\u7684\u63a7\u5236\u4e2d\u5fc3\u3002enable_auto_commit\u8bbe\u7f6e\u4e3aTrue\u540e\uff0c\u6d88\u8d39\u8005\u5c06\u81ea\u52a8commit\uff0c\u5e76\u4e14\u4e24\u6b21commit\u7684\u65f6\u95f4\u95f4\u9694\u4e3aauto_commit_interval_ms\u3002<\/p>\n<\/li>\n<\/ul>\n<h3>\uff082\uff09\u624b\u52a8commit<\/h3>\n<pre><code class=\"language-python\">def consumer_demo():\n    consumer = KafkaConsumer(\n        &#039;kafka_demo&#039;, \n        bootstrap_servers=&#039;:9092&#039;,\n        group_id=&#039;test&#039;,\n        enable_auto_commit=False\n    )\n    for message in consumer:\n        print(&quot;receive, key: {}, value: {}&quot;.format(\n            json.loads(message.key.decode()),\n            json.loads(message.value.decode())\n            )\n        )\n        consumer.commit()<\/code><\/pre>\n<h3>\uff083\uff09\u67e5\u770bkafka\u5806\u79ef\u5269\u4f59\u91cf<\/h3>\n<p>\u5728\u7ebf\u73af\u5883\u4e2d\uff0c\u9700\u8981\u4fdd\u8bc1\u6d88\u8d39\u8005\u7684\u6d88\u8d39\u901f\u5ea6\u5927\u4e8e\u751f\u4ea7\u8005\u7684\u751f\u4ea7\u901f\u5ea6\uff0c\u6240\u4ee5\u9700\u8981\u68c0\u6d4bkafka\u4e2d\u7684\u5269\u4f59\u5806\u79ef\u91cf\u662f\u5728\u589e\u52a0\u8fd8\u662f\u51cf\u5c0f\u3002\u53ef\u4ee5\u7528\u5982\u4e0b\u4ee3\u7801\uff0c\u89c2\u6d4b\u961f\u5217\u6d88\u606f\u5269\u4f59\u91cf\uff1a<\/p>\n<pre><code class=\"language-python\">consumer = KafkaConsumer(topic, **kwargs)\npartitions = [TopicPartition(topic, p) for p in consumer.partitions_for_topic(topic)]\n\nprint(&quot;start to cal offset:&quot;)\n\n# total\ntoff = consumer.end_offsets(partitions)\ntoff = [(key.partition, toff[key]) for key in toff.keys()]\ntoff.sort()\nprint(&quot;total offset: {}&quot;.format(str(toff)))\n\n# current\ncoff = [(x.partition, consumer.committed(x)) for x in partitions]\ncoff.sort()\nprint(&quot;current offset: {}&quot;.format(str(coff)))\n\n# cal sum and left\ntoff_sum = sum([x[1] for x in toff])\ncur_sum = sum([x[1] for x in coff if x[1] is not None])\nleft_sum = toff_sum - cur_sum\nprint(&quot;kafka left: {}&quot;.format(left_sum))<\/code><\/pre>\n<p>\u5b8c\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>kafka-python\u5b98\u65b9\u6587\u6863\uff1a https:\/\/kafka-python.readthedocs.io\/en\/master\/apid   \u2026 &#8230;<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ngg_post_thumbnail":0},"categories":[63],"tags":[61,64,16],"_links":{"self":[{"href":"\/index.php?rest_route=\/wp\/v2\/posts\/1128"}],"collection":[{"href":"\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=1128"}],"version-history":[{"count":3,"href":"\/index.php?rest_route=\/wp\/v2\/posts\/1128\/revisions"}],"predecessor-version":[{"id":1131,"href":"\/index.php?rest_route=\/wp\/v2\/posts\/1128\/revisions\/1131"}],"wp:attachment":[{"href":"\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1128"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1128"},{"taxonomy":"post_tag","embeddable":true,"href":"\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1128"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}