Introduction

This is a Suffix Tree data structure lib implemented in C++, wrapped with python.

Expect for contructing the tree, this lib also provides to construct the query tree for speeding up querying strings.

Install and Build

Install from Pypi

run pip install csuffixtree

Build

You can build this either on linux or windows.

Install Python Package

1. Build SuffixTreePyBinding solution in Windows
2. On linux you should go to linux directory, and run make python
3. under SuffixTreePy directory. run python setup.py install

How to use

You can construct a suffix tree in the following way.

from suffixtree import *
tree = SuffixTree(True,["abc","123","321"])
tree.addStrings(["xyz","abcd"])
print(tree.findString("1"))
print(tree.findString("a"))

If you finished inserting strings to a suffix tree, and you want the querying be very fast you can do:

# convert tree to query tree, this release a part of memory
qtree = tree.createQueryTree()
qtree.cacheNodes() # take some time to cache intermediate nodes
qtree.findString("a")

For simplicity you can also write

# do not preserve string
qtree = SuffixQueryTree(False,["abc","abcd","123","321"])
idx = qtree.findStringIdx("abc") # you cannot use findString now
print(idx)

Serialization

SuffixQueryTree can be serialized and deserialized to/from a file a bytes object. For example:

from suffixtree import *

qtree = SuffixQueryTree(True,["abc","abcd","123","321"])
path = "snapshot.bin"
qtree.serialize(path)
qtree2 = SuffixQueryTree(True)
qtree2.deserialize(path)
idx = qtree.findString("abc") 
idx2 = qtree2.findString("abc")
print(idx,idx2)

# it is also possible to serialize the object as a bytes object

qtree = SuffixQueryTree(True,["abc","abcd","123","321"])
path = "snapshot.bin"
data = qtree.serialize()

print(type(data),len(data)) # bytes, length of bytes

qtree2 = SuffixQueryTree(True)
qtree2.deserialize(data)
idx = qtree.findString("abc") 
idx2 = qtree2.findString("abc")
print(idx,idx2)