simhash 0.2.4 → 0.2.5
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- data/lib/simhash.rb +5 -1
- metadata +5 -7
data/lib/simhash.rb
CHANGED
|
@@ -14,7 +14,11 @@ end
|
|
|
14
14
|
|
|
15
15
|
module Simhash
|
|
16
16
|
DEFAULT_STRING_HASH_METHOD = String.public_instance_methods.include?("hash_vl") ? :hash_vl : :hash_vl_rb
|
|
17
|
-
PUNCTUATION_REGEXP =
|
|
17
|
+
PUNCTUATION_REGEXP = if RUBY_VERSION >= "1.9"
|
|
18
|
+
/(\s|\d|[^\p{L}]|\302\240| *— *|[«»…\-–—]| )+/u
|
|
19
|
+
else
|
|
20
|
+
/(\s|\d|\W|\302\240| *— *|[«»…\-–—]| )+/u
|
|
21
|
+
end
|
|
18
22
|
|
|
19
23
|
|
|
20
24
|
def self.hash(tokens, options={})
|
metadata
CHANGED
|
@@ -1,13 +1,13 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: simhash
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
hash:
|
|
4
|
+
hash: 29
|
|
5
5
|
prerelease:
|
|
6
6
|
segments:
|
|
7
7
|
- 0
|
|
8
8
|
- 2
|
|
9
|
-
-
|
|
10
|
-
version: 0.2.
|
|
9
|
+
- 5
|
|
10
|
+
version: 0.2.5
|
|
11
11
|
platform: ruby
|
|
12
12
|
authors:
|
|
13
13
|
- Alex Gusev
|
|
@@ -15,8 +15,7 @@ autorequire:
|
|
|
15
15
|
bindir: bin
|
|
16
16
|
cert_chain: []
|
|
17
17
|
|
|
18
|
-
date: 2011-
|
|
19
|
-
default_executable:
|
|
18
|
+
date: 2011-10-28 00:00:00 Z
|
|
20
19
|
dependencies:
|
|
21
20
|
- !ruby/object:Gem::Dependency
|
|
22
21
|
name: unicode
|
|
@@ -68,7 +67,6 @@ files:
|
|
|
68
67
|
- lib/string.rb
|
|
69
68
|
- ext/string_hashing/extconf.rb
|
|
70
69
|
- ext/string_hashing/string_hashing.c
|
|
71
|
-
has_rdoc: true
|
|
72
70
|
homepage: http://github.com/bookmate/simhash
|
|
73
71
|
licenses: []
|
|
74
72
|
|
|
@@ -98,7 +96,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
|
|
|
98
96
|
requirements: []
|
|
99
97
|
|
|
100
98
|
rubyforge_project: simhash
|
|
101
|
-
rubygems_version: 1.
|
|
99
|
+
rubygems_version: 1.8.10
|
|
102
100
|
signing_key:
|
|
103
101
|
specification_version: 3
|
|
104
102
|
summary: "Gives you possbility to convert string into simhashes to futher use: finding near-duplicates, similar strings, etc."
|