def testImportBad(self):
data1 = file(testdata_filename).read()
data2 = data1.replace('<review id="17542">', '<somethingelse>')
data2 = data2.replace('</review>', '</somethingelse>', 1)
data1 = sehepunkte_parser.parse(data1)
data2 = sehepunkte_parser.parse(data2)
# Two because I am lazy. The first result contains two books that results in two results
# the replace above can only break the first import.
self.assertEquals(len(list(data1)), len(list(data2)) + 2)
def __call__(self):
data = []
before = datetime.datetime.now()
review_count = 0
for url in self._getTargetURLs():
try:
data.append(sehepunkte_parser.parse(urllib.urlopen(url).read()))
except IOError:
pass # The library takes care of logging a failure
for review in chain(*data):
try:
self._addReview(self._convertVocabulary(\
convertToString(\
review)))
review_count += 1
except:
log.exception("Warning, sehepunkte import failed! Exception "
"has not been catched, but bubbled. Probably sehepunkte "
"is broken now! Please see #4656 for fixes")
raise
total = (datetime.datetime.now() - before).seconds / 1.0
log.info("Sehepunkte finished. Imported %i reviews "
"in %f seconds. %f reviews/s",
review_count, total, total and review_count / total or review_count)
return "Success"
def testImportGood(self):
data = [x for x in sehepunkte_parser.parse(
file(testdata_filename).read())]
self.maxDiff = None
self.assertEquals(testdata_parsed, list(data))