Create a function to parse amazon urls and reduce code redundancy.

A test for amazon url parsing was added.

picard/const.py

@@ -43,9 +43,6 @@ FPCALC_NAMES = ['fpcalc', 'pyfpcalc']
 # Various Artists MBID
 VARIOUS_ARTISTS_ID = '89ad4ac3-39f7-470e-963a-56509c546377'
 
-# Amazon asin url
-AMAZON_ASIN_URL_REGEX = re.compile(r'^http://(?:www.)?(.*?)(?:\:[0-9]+)?/.*/([0-9B][0-9A-Z]{9})(?:[^0-9A-Z]|$)')
-
 # Release formats
 RELEASE_FORMATS = {
     u'CD': N_('CD'),

picard/coverart.py

@@ -26,9 +26,8 @@ import traceback
 import picard.webservice
 
 from picard import config, log
-from picard.const import AMAZON_ASIN_URL_REGEX
 from picard.metadata import Metadata, is_front_image
-from picard.util import partial, mimetype
+from picard.util import partial, mimetype, parse_amazon_url
 from PyQt4.QtCore import QUrl, QObject
 
 # data transliterated from the perl stuff used to find cover art for the
@@ -282,17 +281,15 @@ def _process_url_relation(try_list, relation):
     return False
 
 def _process_asin_relation(try_list, relation):
-    match = AMAZON_ASIN_URL_REGEX.match(relation.target[0].text)
-    if match is not None:
-        asinHost = match.group(1)
-        asin = match.group(2)
-        if asinHost in AMAZON_SERVER:
-            serverInfo = AMAZON_SERVER[asinHost]
+    amz = parse_amazon_url(relation.target[0].text)
+    if amz is not None:
+        if amz['host'] in AMAZON_SERVER:
+            serverInfo = AMAZON_SERVER[amz['host']]
         else:
             serverInfo = AMAZON_SERVER['amazon.com']
         host = serverInfo['server']
-        path_l = AMAZON_IMAGE_PATH % (asin, serverInfo['id'], 'L')
-        path_m = AMAZON_IMAGE_PATH % (asin, serverInfo['id'], 'M')
+        path_l = AMAZON_IMAGE_PATH % (amz['asin'], serverInfo['id'], 'L')
+        path_m = AMAZON_IMAGE_PATH % (amz['asin'], serverInfo['id'], 'M')
         _try_list_append_image_url(try_list, QUrl("http://%s:%s" % (host, path_l)))
         _try_list_append_image_url(try_list, QUrl("http://%s:%s" % (host, path_m)))
 

picard/mbxml.py

@@ -19,8 +19,8 @@
 
 import re
 from picard import config
-from picard.util import format_time, translate_from_sortname
-from picard.const import RELEASE_FORMATS, AMAZON_ASIN_URL_REGEX
+from picard.util import format_time, translate_from_sortname, parse_amazon_url
+from picard.const import RELEASE_FORMATS
 
 
 _artist_rel_types = {
@@ -98,12 +98,11 @@ def _relations_to_metadata(relation_lists, m):
                     work_to_metadata(relation.work[0], m)
         elif relation_list.target_type == 'url':
             for relation in relation_list.relation:
-                if relation.type == 'amazon asin':
-                    url = relation.target[0].text
-                    match = AMAZON_ASIN_URL_REGEX.match(url)
-                    if match is not None and 'asin' not in m:
-                        m['asin'] = match.group(2)
-                if relation.type == 'license':
+                if relation.type == 'amazon asin' and 'asin' not in m:
+                    amz = parse_amazon_url(relation.target[0].text)
+                    if amz is not None:
+                        m['asin'] = amz['asin']
+                elif relation.type == 'license':
                     url = relation.target[0].text
                     m.add('license', url)
 

picard/util/__init__.py

@@ -325,3 +325,14 @@ def load_release_type_scores(setting):
 
 def save_release_type_scores(scores):
     return " ".join(["%s %.2f" % v for v in scores.iteritems()])
+
+
+def parse_amazon_url(url):
+    """Extract host and asin from an amazon url.
+    It returns a dict with host and asin keys on success, None else
+    """
+    r = re.compile(r'^http://(?:www.)?(?P<host>.*?)(?:\:[0-9]+)?/.*/(?P<asin>[0-9B][0-9A-Z]{9})(?:[^0-9A-Z]|$)')
+    match = r.match(url)
+    if match is not None:
+        return match.groupdict()
+    return None

test/test_amazon_urls.py

@@ -0,0 +1,39 @@
+# -*- coding: utf-8 -*-
+
+import unittest
+from picard.util import parse_amazon_url
+
+
+class ParseAmazonUrlTest(unittest.TestCase):
+
+    def test_1(self):
+        url = 'http://www.amazon.com/dp/020530902X'
+        expected = {'asin': '020530902X', 'host': 'amazon.com'}
+        r = parse_amazon_url(url)
+        self.failUnlessEqual(r, expected)
+
+    def test_2(self):
+        url = 'http://ec1.amazon.co.jp/gp/product/020530902X'
+        expected = {'asin': '020530902X', 'host': 'ec1.amazon.co.jp'}
+        r = parse_amazon_url(url)
+        self.failUnlessEqual(r, expected)
+
+    def test_3(self):
+        url = 'http://amazon.com/Dark-Side-Moon-Pink-Floyd/dp/B004ZN9RWK/ref=sr_1_1?s=music&ie=UTF8&qid=1372605047&sr=1-1&keywords=pink+floyd+dark+side+of+the+moon'
+        expected = {'asin': 'B004ZN9RWK', 'host': 'amazon.com'}
+        r = parse_amazon_url(url)
+        self.failUnlessEqual(r, expected)
+
+    def test_4(self):
+        #incorrect ASIN
+        url = 'http://www.amazon.com/dp/A20530902X'
+        expected = None
+        r = parse_amazon_url(url)
+        self.failUnlessEqual(r, expected)
+
+    def test_5(self):
+        #incorrect ASIN
+        url = 'http://www.amazon.com/dp/020530902x'
+        expected = None
+        r = parse_amazon_url(url)
+        self.failUnlessEqual(r, expected)