summaryrefslogtreecommitdiff
path: root/pt-bloom
diff options
context:
space:
mode:
authorPatrick Simianer <p@simianer.de>2016-07-05 11:01:46 +0200
committerPatrick Simianer <p@simianer.de>2016-07-05 11:01:46 +0200
commit2b1d7f881c19c4d4b5afae194e02d3300c7675d0 (patch)
tree5a06ee7de98640a39244b57bb369697176b44ebf /pt-bloom
parent69949dda35c3ea21d8e926e5f0a596a0a0f61c6a (diff)
mv
Diffstat (limited to 'pt-bloom')
-rwxr-xr-xpt-bloom24
1 files changed, 24 insertions, 0 deletions
diff --git a/pt-bloom b/pt-bloom
new file mode 100755
index 0000000..5c2cf01
--- /dev/null
+++ b/pt-bloom
@@ -0,0 +1,24 @@
+#!/usr/bin/env ruby
+
+require 'bloom-filter'
+require 'trollop'
+
+STDIN.set_encoding 'utf-8'
+STDOUT.set_encoding 'utf-8'
+
+conf = Trollop::options do
+ opt :size, "number of entries in the filter", :type => :int, :required => true
+ opt :error_rate, "error rate", :type => :float, :default => 0.01
+end
+
+f = BloomFilter.new conf[:size], conf[:error_rate]
+while line = STDIN.gets
+ src, tgt = splitpipe(line)[0..1]
+ src.strip!
+ tgt.strip!
+ f.insert(src+" ||| "+tgt)
+end
+
+f.dump('pt.bloom')
+f.close
+