def testManyDuplicationsInOrder1234(self):
"""
A list with duplicates of all of 1 2 3 4 (introduced in that order)
should become the list 1 2 3 4 with only one of each.
"""
self.assertEquals([1, 2, 3, 4],
uniqueList([1, 2, 1, 2, 3, 1, 2, 4, 1, 3, 2]))
def testManyDuplicationsInOrder4321(self):
"""
A list with duplicates of all of 4 3 2 1 (introduced in that order)
should become the list 4 3 2 1 with only one of each.
"""
self.assertEquals([4, 3, 2, 1],
uniqueList([4, 4, 3, 4, 3, 2, 3, 4, 4, 3, 1, 2, 4]))
def testManyDuplicationsInOrder4321(self):
"""
A list with duplicates of all of 4 3 2 1 (introduced in that order)
should become the list 4 3 2 1 with only one of each.
"""
self.assertEquals([4, 3, 2, 1],
uniqueList([4, 4, 3, 4, 3, 2, 3, 4, 4, 3, 1, 2, 4]))
def testManyDuplicationsInOrder1234(self):
"""
A list with duplicates of all of 1 2 3 4 (introduced in that order)
should become the list 1 2 3 4 with only one of each.
"""
self.assertEquals([1, 2, 3, 4],
uniqueList([1, 2, 1, 2, 3, 1, 2, 4, 1, 3, 2]))
def extractAtnames(comment):
"""Find all @names in a comment.
@param comment: The C{unicode} comment text.
@return: A C{list} of C{@}names, with no duplicates, in the order they
appear in the comment.
"""
return uniqueList(findall(ATNAME_REGEX, comment))
def extractFiles(comment):
"""Find all files in a comment.
@param comment: The C{unicode} comment text.
@return: A C{list} of about values from the comment, with no duplicates,
in the order they appear in the comment.
"""
return uniqueList(findall(FILE_REGEX, comment))
def extractPlustags(comment):
"""Find all +plustags in a comment.
@param comment: The C{unicode} comment text.
@return: A C{list} of +plustags, with no duplicates, in the order they
appear in the comment.
"""
return uniqueList(findall(PLUSTAG_REGEX, comment))
def extractAtnames(comment):
"""Find all @names in a comment.
@param comment: The C{unicode} comment text.
@return: A C{list} of C{@}names, with no duplicates, in the order they
appear in the comment.
"""
return uniqueList(findall(ATNAME_REGEX, comment))
def extractFiles(comment):
"""Find all files in a comment.
@param comment: The C{unicode} comment text.
@return: A C{list} of about values from the comment, with no duplicates,
in the order they appear in the comment.
"""
return uniqueList(findall(FILE_REGEX, comment))
def extractPlustags(comment):
"""Find all +plustags in a comment.
@param comment: The C{unicode} comment text.
@return: A C{list} of +plustags, with no duplicates, in the order they
appear in the comment.
"""
return uniqueList(findall(PLUSTAG_REGEX, comment))
def testNoDuplicates(self):
"""A list with no duplicates should be untouched."""
self.assertEquals(['hey', 'you', 'guys'],
uniqueList(['hey', 'you', 'guys']))
def testSimpleDuplicate(self):
"""A list of two identical things gets reduced to length one."""
self.assertEquals(['hey'], uniqueList(['hey', 'hey']))
def testEmpty(self):
"""The empty list is the empty list when uniqued."""
self.assertEquals([], uniqueList([]))
def testSimpleDuplicate(self):
"""A list of two identical things gets reduced to length one."""
self.assertEquals(['hey'], uniqueList(['hey', 'hey']))
def create(self,
text,
username,
about=None,
importer=None,
when=None,
url=None):
"""Create a new comment.
@param text: The C{unicode} comment text.
@param username: the C{unicode} username of the commenter.
@param about: Optionally, a C{list} of C{unicode} values the comment is
about.
@param importer: A C{unicode} string giving the name of the importer.
@param when: A C{datetime.datetime} instance or C{None} if the
current time should be used.
@param url: A C{str} URL or C{None} if there is no URL associated with
this comment.
@raise L{FeatureError}: if (1) the comment text is C{None} or is all
whitespace, or (2) if the importer name contains the separator
(space) that we use in the about value for comment objects.
@return: A C{dict} as follows:
{
fluidinfo.com/info/about: A C{list} of all the about values
(i.e., URLs and hashtags) in the comment text, including
the thing the comment was about (if anything). The hashtags
are in lowercase.
fluidinfo.com/info/timestamp: The C{int} UTC timestamp (seconds
since the epoch) the comment was created at.
fluidinfo.com/info/url: The C{url}, as received.
fluidinfo.com/info/username: The C{username}, as received.
}
"""
if not text or text.strip() == '':
raise FeatureError('Comment text non-existent or just whitespace.')
if importer:
if ' ' in importer:
raise FeatureError('Comment importer name contains a space.')
else:
importer = u'fluidinfo.com'
when = when or datetime.utcnow()
floatTime = timegm(when.utctimetuple()) + float(when.strftime('0.%f'))
isoTime = when.isoformat()
if not url:
url = 'https://fluidinfo.com/comment/%s/%s/%s' % (
importer, username, isoTime)
# Put all the explicit about values into a list called abouts. Items
# are stripped and those that are not URLs are lowercased.
abouts = []
if about:
for item in map(unicode.strip, about):
abouts.append(item if URL_REGEX.match(item) else item.lower())
abouts.extend(self._extractAbouts(text))
abouts = uniqueList(abouts)
commentObjectAbout = u'%s %s %s' % (importer, username, isoTime)
commentID = self._objects.create(commentObjectAbout)
values = {
u'fluidinfo.com/info/about': abouts,
u'fluidinfo.com/info/username': username,
u'fluidinfo.com/info/text': text,
u'fluidinfo.com/info/url': url,
u'fluidinfo.com/info/timestamp': floatTime
}
self._tagValues.set({commentID: values})
if abouts:
# Get all the object IDs of the target objects. If an object does
# not exist, create it.
result = getAboutTagValues(values=abouts)
existingObjects = dict(
result.values(AboutTagValue.value, AboutTagValue.objectID))
missingAbouts = set(abouts) - set(existingObjects.iterkeys())
for aboutValue in missingAbouts:
existingObjects[aboutValue] = self._objects.create(aboutValue)
createComment(commentID, existingObjects.values(), username, when)
return values
def testDuplicatesWithInterveningElement(self):
"""A list with elements A B A should become just A B."""
self.assertEquals(['A', 'B'], uniqueList(['A', 'B', 'A']))
def testNoDuplicates(self):
"""A list with no duplicates should be untouched."""
self.assertEquals(['hey', 'you', 'guys'],
uniqueList(['hey', 'you', 'guys']))
def testDuplicatesWithInterveningElement(self):
"""A list with elements A B A should become just A B."""
self.assertEquals(['A', 'B'], uniqueList(['A', 'B', 'A']))
def testEmpty(self):
"""The empty list is the empty list when uniqued."""
self.assertEquals([], uniqueList([]))
def create(self, text, username, about=None, importer=None, when=None,
url=None):
"""Create a new comment.
@param text: The C{unicode} comment text.
@param username: the C{unicode} username of the commenter.
@param about: Optionally, a C{list} of C{unicode} values the comment is
about.
@param importer: A C{unicode} string giving the name of the importer.
@param when: A C{datetime.datetime} instance or C{None} if the
current time should be used.
@param url: A C{str} URL or C{None} if there is no URL associated with
this comment.
@raise L{FeatureError}: if (1) the comment text is C{None} or is all
whitespace, or (2) if the importer name contains the separator
(space) that we use in the about value for comment objects.
@return: A C{dict} as follows:
{
fluidinfo.com/info/about: A C{list} of all the about values
(i.e., URLs and hashtags) in the comment text, including
the thing the comment was about (if anything). The hashtags
are in lowercase.
fluidinfo.com/info/timestamp: The C{int} UTC timestamp (seconds
since the epoch) the comment was created at.
fluidinfo.com/info/url: The C{url}, as received.
fluidinfo.com/info/username: The C{username}, as received.
}
"""
if not text or text.strip() == '':
raise FeatureError('Comment text non-existent or just whitespace.')
if importer:
if ' ' in importer:
raise FeatureError('Comment importer name contains a space.')
else:
importer = u'fluidinfo.com'
when = when or datetime.utcnow()
floatTime = timegm(when.utctimetuple()) + float(when.strftime('0.%f'))
isoTime = when.isoformat()
if not url:
url = 'https://fluidinfo.com/comment/%s/%s/%s' % (
importer, username, isoTime)
# Put all the explicit about values into a list called abouts. Items
# are stripped and those that are not URLs are lowercased.
abouts = []
if about:
for item in map(unicode.strip, about):
abouts.append(item if URL_REGEX.match(item) else item.lower())
abouts.extend(self._extractAbouts(text))
abouts = uniqueList(abouts)
commentObjectAbout = u'%s %s %s' % (importer, username, isoTime)
commentID = self._objects.create(commentObjectAbout)
values = {u'fluidinfo.com/info/about': abouts,
u'fluidinfo.com/info/username': username,
u'fluidinfo.com/info/text': text,
u'fluidinfo.com/info/url': url,
u'fluidinfo.com/info/timestamp': floatTime}
self._tagValues.set({commentID: values})
if abouts:
# Get all the object IDs of the target objects. If an object does
# not exist, create it.
result = getAboutTagValues(values=abouts)
existingObjects = dict(result.values(AboutTagValue.value,
AboutTagValue.objectID))
missingAbouts = set(abouts) - set(existingObjects.iterkeys())
for aboutValue in missingAbouts:
existingObjects[aboutValue] = self._objects.create(aboutValue)
createComment(commentID, existingObjects.values(), username, when)
return values