Last active
February 1, 2017 02:10
-
-
Save dat-boris/cab29b158297cc491e6afebac9f5eb36 to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import re | |
| from collections import Counter | |
| RE_CHAR = re.compile('\w') | |
| def functional_counts(stream): | |
| """ | |
| How would we scale and distribute a wordcount operation | |
| """ | |
| datapipe = pipeline.Pipeline([ | |
| emit_words, | |
| filter_empty_word, | |
| count_words | |
| ]) | |
| return datapipe.apply(iter(lambda: stream.read(1), '')) | |
| assert emit_words("Hello, world🌏") | |
| == ["Hello", "world"] | |
| assert emit_words("世界你好") | |
| == ["世界", "你好"] | |
| # This test will fail in our simple example, but use to make a point in the presentation | |
| def emit_words(stream): | |
| """ | |
| Return a generator of words | |
| """ | |
| word = '' | |
| for char in stream: | |
| if char is None: | |
| break | |
| if not RE_CHAR.match(char): | |
| yield word | |
| word = '' | |
| else: | |
| word += char | |
| yield word | |
| def filter_empty_word(word_stream): | |
| """ | |
| Filter out empty words | |
| - principle one, edge case should be handled by separate function | |
| - Note: can be replace by ifilter | |
| """ | |
| for word in word_stream: | |
| if word: | |
| yield word | |
| def count_words(word_stream): | |
| """ | |
| Filter out empty words | |
| """ | |
| counter = Counter() | |
| for word in word_stream: | |
| counter[word] += 1 | |
| return dict(counter) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment