From 351da575db883411bf2c7a0d9c974efd716da497 Mon Sep 17 00:00:00 2001
From: urlbot <urlbot@eagle.local.yeeer.net>
Date: Wed, 1 Oct 2014 09:33:21 +0200
Subject: [PATCH] (not in use) add string_similarity() for testing

---
 strsim.py | 63 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 63 insertions(+)
 create mode 100755 strsim.py

diff --git a/strsim.py b/strsim.py
new file mode 100755
index 0000000..fbcf8e2
--- /dev/null
+++ b/strsim.py
@@ -0,0 +1,63 @@
+#!/usr/bin/python3
+
+import re
+
+def str_sim(a, b):
+	a = a.lower()
+	b = b.lower()
+
+	a_parts = re.split('[\W_]+', a)
+	b_parts = re.split('[\W_]+', b)
+
+	# this is a "simple" way to declare out[a][b]
+	out = list(map(list, [[0]*len(b_parts)]*len(a_parts)))
+
+	for i in range(0, len(a_parts)-1):
+		for j in range(0, len(b_parts)-1):
+			if a_parts[i] == b_parts[j]:
+				out[i][j] += 1
+	
+	i = 0
+	for j in range(0, len(b_parts)):
+		print('  |'*i + ' '*2 + '.- ' + b_parts[j])
+		i += 1
+	print('  |'*i)
+
+	for i in range(0, len(a_parts)):
+		print(' ' + str(out[i]) + ' ' + a_parts[i])
+
+	return out
+
+def sum_array(array):
+	_sum = 0
+	for a in array:
+		if list == type(a) or tuple == type(a) or hash == type(a):
+			_sum += sum_array(a)
+		elif int == type(a) or float == type(a):
+			_sum += a
+	return _sum
+
+def wrapper_print(a, b, comment=''):
+	ret = str_sim(a, b)
+	if '' != comment: comment = ' ^ ' + comment
+	print('[%2dx%2d::%2d]%s' %(len(ret), len(ret[0]), sum_array(ret), comment))
+
+pairs = (
+	(
+		'http://de.wikipedia.org/wiki/Monte_Kali_%28Heringen%29',
+		'Monte Kali (Heringen)'
+	),
+	(
+		'http://www.spiegel.de/politik/ausland/buddhisten-treffen-in-colombo-blitzender-moench-a-994447.html',
+		'Buddhisten-Treffen in Colombo: Blitzender Mönch - SPIEGEL ONLINE'
+	)
+)
+
+wrapper_print('foo bar baz', 'foo bar boom')
+
+for (url, title) in pairs:
+	wrapper_print(title, url, comment='raw')
+	url_no_proto = re.sub(r'https?://[^/]*/', '', url)
+	wrapper_print(title, url_no_proto, comment='no proto/domain')
+	url_no_proto_no_digits = re.sub(r'[0-9]*', '', url_no_proto)
+	wrapper_print(title, url_no_proto_no_digits, comment='no proto/domain/[0-9]')