C++ program that counts the number of words, characters that are not included already in words, special characters, and numbers in a string or file.
Analysis if Counter constructor with file is used The real workhorse of my version is the countMatches method. This method takes in a customized File object and a regex pattern. It loops through the file via the std library method getLine and delimiterizes the file by '\n' and '\0'. Since we need to include the '\n' in our final count, we used the istream seekg method to offset the current position of our pointer by -1; then added this char to the end of our line. This works great since getLine also finds the null termination character '\0' and will add it instead of '\n', if '\n' is not the delim character. This loop also keeps track of the _position Counter attribute.
We then call the overloaded countMatches method from the CountMatches loop. This method takes in a string from CountMatches loop and the regex pattern. It iterates through all the matches and calls countOccurrenceOf with each match passed in.
countOccurrenceOf converts each match to whatever the modification function was passed in. Unless it was the default modification function which does nothing. Then proceeds to either add the match to the count map as a new entry object or increments the _timesOccured property of that Entry object. Through the occured() method. The position of each Entry is kept track of by the _position attribute.
After all Entries are accounted for, we then call the rankValues method and pass a custom type called EntryCompareFunction which is a static function, functor, of the Entry Object. The Priorty Queue then inserts each Entry object from the count map via foreach loop. This allows us to dynamically and automatically sort our Entry objects by the defined functor.
countMatches: O(n) where n is the file size. countMatches: O(n) where n is the line size. countOccurrenceOf: O(n) where n is the matched string size that need to be converted to lower case. rankValues: O(nlog(n)) where n is the size of the count map and log(n) is the insertion
Time Complexity: O(n^2) Space Complexity: O(n)
Analysis if Counter constructor with string is used Same as above applies, execpt one BIG difference. In this scenerio we are reading the file into memory first and passing the contents as a string into the Counter constructor. By doing thing, we are skipping the call to the file version of countMatches and going directly to to the string version of countMatches.
The disadvantage here is that the std::string type can hold up to 4294967291 characters and it will take more memory to hold the file in memory. The advantage is reading the file in first, reduces run time to nlog(n).
countMatches: O(n) where n is the file size. countOccurrenceOf: O(n) where n is the matched string size that need to be converted to lower case. rankValues: O(nlog(n)) where n is the size of the count map and log(n) is the insertion
Time Complexity: O(nlog(n)) Space Complexity: O(n)