This is a fairly straighforward port of Martin Porter's C implementation of the Porter stemming algorithm. The C version this port is based on is available for download here: http://tartarus.org/~martin/PorterStemmer/c_thread_safe.txt
The original algorithm is described in the paper:
M.F. Porter, 1980, An algorithm for suffix stripping, Program, 14(3) pp
130-137.
While the internal implementation and interface is nearly identical to the original implementation, the Go interface is much simplified. The stemmer can be called as follows:
import "porter"
...
stemmed := porter.Stem(word_to_stem)
go get github.com/a2800276/porter
to use the stemmer when installed using goinstall, import:
import "github.com/a2800276/porter"
While the implementation is fairly robust, this is a work in progress.
In particular, a new interface will likely be provided to prevent
excessive conversions between string
s and []byte
. Currently, on
calling Stem
the string argument is converted to a byte slice which
the algorithm works on and is converted back into a string before
returning.
Also, the implementation is not particularly robust at handling Unicode input, currently, only bytes with the high bit set are ignored. It's up to the caller to make sure the string contains only ASCII characters. Since the algorithm itself operates on English words only, this doens't restrict the functionality, but it is nuisance.
- byte slice API to void roundtripping to string and back