From 22feb94f0b4e0d569ec4408a581a5c7d4144c5a0 Mon Sep 17 00:00:00 2001 From: luozhouyang Date: Wed, 20 Feb 2019 13:00:56 +0800 Subject: [PATCH] Revert "added regex prior to profiles calculation" --- similarity/cosine.py | 7 ------- similarity/jaccard.py | 6 ------ 2 files changed, 13 deletions(-) diff --git a/similarity/cosine.py b/similarity/cosine.py index 64f45e0..799e598 100644 --- a/similarity/cosine.py +++ b/similarity/cosine.py @@ -24,9 +24,6 @@ from .string_distance import NormalizedStringDistance from .string_similarity import NormalizedStringSimilarity -import re -_SPACE_PATTERN = re.compile("\\s+") - class Cosine(ShingleBased, NormalizedStringDistance, NormalizedStringSimilarity): @@ -44,10 +41,6 @@ def similarity(self, s0, s1): raise TypeError("Argument s1 is NoneType.") if s0 == s1: return 1.0 - - s0 = _SPACE_PATTERN.sub("", s0) - s1 = _SPACE_PATTERN.sub("", s1) - if len(s0) < self.get_k() or len(s1) < self.get_k(): return 0.0 profile0 = self.get_profile(s0) diff --git a/similarity/jaccard.py b/similarity/jaccard.py index ad6d832..a7db3c6 100644 --- a/similarity/jaccard.py +++ b/similarity/jaccard.py @@ -22,8 +22,6 @@ from .string_distance import NormalizedStringDistance, MetricStringDistance from .string_similarity import NormalizedStringSimilarity -import re -_SPACE_PATTERN = re.compile("\\s+") class Jaccard(ShingleBased, MetricStringDistance, NormalizedStringDistance, NormalizedStringSimilarity): @@ -40,10 +38,6 @@ def similarity(self, s0, s1): raise TypeError("Argument s1 is NoneType.") if s0 == s1: return 1.0 - - s0 = _SPACE_PATTERN.sub("", s0) - s1 = _SPACE_PATTERN.sub("", s1) - if len(s0) < self.get_k() or len(s1) < self.get_k(): return 0.0 profile0 = self.get_profile(s0)