let's play with google ngrams data
run make download
to download some sample data, or visit the google page to get more of it.
./wordtrie 'ngram string' data1.csv [data2.csv data3.csv...]
builds a data structure and fetches information about the ngram string.
there are a lot of inline functions used instead of macros. hence compiling without optimization makes the code much slower.
- rewrite the code in rust, and handle the v2 dataset, which has part of speech tags.