Rationale
This program counts the occurrence of the words in the file.
Synopsis
$ ./mapred file N file: a file name containing words (can be generated here:http://www.lipsum.com/feed/html ) N : a number of threads. The program will launch N threads (using pthreads) to share the work as equally as possible (map phase). The program will then collect the results of all the threads (reduce phase), aggregate all the results, and display them, sorted by word. A word is a sequence of characters between 2 separators. These separators are start and end of file, tabulations, spaces, punctuation, and end of line.
Example
Of the file is the following: foo bar qux bar bar baz And we launch the program with e.g. N=2, the program shall display: bar=3 baz=1 foo=1 qux=1
Build
Run ./waf configure to configure with default settings.
Run ./waf build to build in release mode or ./waf build --mode=debug with debug symbols and w/o CC optimizations.
The output in the build directory.
See test.sh for details.