github.com/apache/beam/sdks/v2@v2.48.2/python/apache_beam/examples/windowed_wordcount.py (about) 1 # 2 # Licensed to the Apache Software Foundation (ASF) under one or more 3 # contributor license agreements. See the NOTICE file distributed with 4 # this work for additional information regarding copyright ownership. 5 # The ASF licenses this file to You under the Apache License, Version 2.0 6 # (the "License"); you may not use this file except in compliance with 7 # the License. You may obtain a copy of the License at 8 # 9 # http://www.apache.org/licenses/LICENSE-2.0 10 # 11 # Unless required by applicable law or agreed to in writing, software 12 # distributed under the License is distributed on an "AS IS" BASIS, 13 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 14 # See the License for the specific language governing permissions and 15 # limitations under the License. 16 # 17 18 """A streaming word-counting workflow. 19 20 Important: streaming pipeline support in Python Dataflow is in development 21 and is not yet available for use. 22 """ 23 24 # pytype: skip-file 25 26 import argparse 27 import logging 28 29 import apache_beam as beam 30 from apache_beam.transforms import window 31 32 TABLE_SCHEMA = ( 33 'word:STRING, count:INTEGER, ' 34 'window_start:TIMESTAMP, window_end:TIMESTAMP') 35 36 37 def find_words(element): 38 import re 39 return re.findall(r'[A-Za-z\']+', element) 40 41 42 class FormatDoFn(beam.DoFn): 43 def process(self, element, window=beam.DoFn.WindowParam): 44 ts_format = '%Y-%m-%d %H:%M:%S.%f UTC' 45 window_start = window.start.to_utc_datetime().strftime(ts_format) 46 window_end = window.end.to_utc_datetime().strftime(ts_format) 47 return [{ 48 'word': element[0], 49 'count': element[1], 50 'window_start': window_start, 51 'window_end': window_end 52 }] 53 54 55 def main(argv=None): 56 """Build and run the pipeline.""" 57 58 parser = argparse.ArgumentParser() 59 parser.add_argument( 60 '--input_topic', 61 required=True, 62 help='Input PubSub topic of the form "/topics/<PROJECT>/<TOPIC>".') 63 parser.add_argument( 64 '--output_table', 65 required=True, 66 help=( 67 'Output BigQuery table for results specified as: ' 68 'PROJECT:DATASET.TABLE or DATASET.TABLE.')) 69 known_args, pipeline_args = parser.parse_known_args(argv) 70 71 with beam.Pipeline(argv=pipeline_args) as p: 72 73 # Read the text from PubSub messages. 74 lines = p | beam.io.ReadFromPubSub(known_args.input_topic) 75 76 # Get the number of appearances of a word. 77 def count_ones(word_ones): 78 (word, ones) = word_ones 79 return (word, sum(ones)) 80 81 transformed = ( 82 lines 83 | 'Split' >> (beam.FlatMap(find_words).with_output_types(str)) 84 | 'PairWithOne' >> beam.Map(lambda x: (x, 1)) 85 | beam.WindowInto(window.FixedWindows(2 * 60, 0)) 86 | 'Group' >> beam.GroupByKey() 87 | 'Count' >> beam.Map(count_ones) 88 | 'Format' >> beam.ParDo(FormatDoFn())) 89 90 # Write to BigQuery. 91 # pylint: disable=expression-not-assigned 92 transformed | 'Write' >> beam.io.WriteToBigQuery( 93 known_args.output_table, 94 schema=TABLE_SCHEMA, 95 create_disposition=beam.io.BigQueryDisposition.CREATE_IF_NEEDED, 96 write_disposition=beam.io.BigQueryDisposition.WRITE_APPEND) 97 98 99 if __name__ == '__main__': 100 logging.getLogger().setLevel(logging.INFO) 101 main()