Python: recursive longest common subsequence (LCS) with dynamic programming

dynlcs.py

class DynamicLcs:

    def __init__(self):
        self.table = {}

    @staticmethod
    def run(s1, s2):
        return DynamicLcs()._run(s1, s2)

    def _run(self, s1, s2):
        result = self.table.get(s1, s2)
        if result is not None:
            return result
        if not s1 or not s2 == '':
            return ''
        if s1[0] == s2[0]:
            self.table[{s1, s2}] = s1[0] + self._run(s1[1:], s2[1:])
        else:
            result1 = self._run(s1[1:], s2)
            result2 = self._run(s1, s2[1:])
            self.table[{s1, s2}] = max(result1, result2, key=len)
        return self.table[{s1, s2}]

# e.g.
a = "testmeiooooooiffffff"
b = "tioooo0oosieffffff"
print(DynamicLcs.run(a, b))