grgar/cc-cw

🎓 University Project: ☁️ 18COC105 Cloud Computing

Anagrams MapReduce

University Project: ☁ 18COC105 Cloud Computing

cc-cw/src/Anagrams.kt

Lines 21 to 24 in a41745f

	/**
	* Performs parsing of text into tokens of words, then builds sets of words which anagram together.
	*/
	class Anagrams : Configured(), Tool {

cc-cw/src/Anagrams.kt

Lines 25 to 33 in a41745f

	/**
	* Mapper to extract words from text. Parses text, extracting all words which can be anagram of another word
	* (i.e. ignores single letter words and words with all the same letter, which can never be anagram'd).
	*
	* Normalises text by splitting on specific word boundaries, removing numbers and words with numbers.
	* • Input is line number of input text [LongWritable] and string in [Text].
	* • Output is word with letters in alphabetical order as [Text] and [SetWritable] containing original form of word.
	*/
	class AnagramMapper : Mapper<LongWritable, Text, Text, SetWritable>() {

cc-cw/src/Anagrams.kt

Lines 155 to 160 in a41745f

	/**
	* Partition based on length of word. This is for performance since words of different length will never anagram with
	* each other. The more partitions available, the more different word lengths can be spread out across them and the
	* less each partition will check keys for a word of different length.
	*/
	class AnagramPartitioner : Partitioner<Text, SetWritable>() {

cc-cw/src/Anagrams.kt

Lines 168 to 172 in a41745f

	/**
	* Reducer for performing anagram operation on words with matching alphabetically ordered letters (i.e. words which
	* are anagrams of each other or are the same word).
	*/
	class AnagramReducer : Reducer<Text, SetWritable, Text, SetWritable>() {