Example #1
0
def calc_hot_window_score(submission_id, hot_point_window=timedelta(hours=6)):
    """
    Retrieve vote/score information for a given submission over hot_point_window time.
    Count votes received in the last hot_point_window time. If this is higher than normal,
    the story will be considered hot.

    @param submission_id: the submission to calculate
    @param hot_point_window: timedelta object representing acceptable vote timeframe from now
    """
    try:
        story_votes = dbsession.query(
            Vote.submission_id,
            func.sum(Vote.points).label('points')).filter(
                Vote.added_on < general.now_in_utc(), Vote.added_on >
                (general.now_in_utc() - hot_point_window)).filter(
                    Vote.submission_id == submission_id).group_by(
                        Vote.submission_id).one()
    except sqlalchemy.orm.exc.NoResultFound:
        # no votes on this story yet
        return 0
    from raggregate.queries import submission
    story = submission.get_story_by_id(submission_id)
    story.hot_window_score = story_votes.points
    story.hot_window_score_timestamp = general.now_in_utc()
    dbsession.add(story)
    return story_votes.points
Example #2
0
def calc_all_hot_window_scores(hot_eligible_age=timedelta(hours=48)):
    # by default, only stories from the last 48 hours are eligible for "hot" status
    stories = dbsession.query(Submission).filter(
        Submission.deleted == False).filter(
            Submission.added_on > (general.now_in_utc() -
                                   hot_eligible_age)).all()

    for s in stories:
        calc_hot_window_score(s.id)

    general.set_key_in_stat('mass_hot_window_timestamp',
                            general.now_in_utc(),
                            type='datetime')

    dbsession.flush()

    return 0
Example #3
0
def count_story_votes(submission_id):
    from raggregate.queries import submission
    dv_num = dbsession.query(Vote).filter(Vote.points < 0).filter(Vote.submission_id == submission_id).count()
    story = submission.get_story_by_id(submission_id)
    story.downvote_tally = dv_num
    story.downvote_tally_timestamp = general.now_in_utc()

    uv_num = dbsession.query(Vote).filter(Vote.points > 0).filter(Vote.submission_id == submission_id).count()
    story.upvote_tally = uv_num
    story.upvote_tally_timestamp = general.now_in_utc()

    story.total_vote_tally = uv_num + dv_num
    story.total_vote_timestamp = general.now_in_utc()

    dbsession.add(story)
    dbsession.flush()

    return [dv_num, uv_num]
Example #4
0
def calc_hot_average(hot_point_window=timedelta(hours=6)):
    """
    Calculate the average point assignment of stories over the last hot_point_window time.
    """
    story_votes = dbsession.query(
        Vote.submission_id,
        func.sum(Vote.points).label('points')).filter(
            Vote.added_on < general.now_in_utc(), Vote.added_on >
            (general.now_in_utc() - hot_point_window)).group_by(
                Vote.submission_id).all()

    aggregate = 0
    count = 0

    for sv in story_votes:
        aggregate += sv.points
        count += 1

    print "aggregate: " + str(aggregate)
    print "count: " + str(count)

    if aggregate <= 0:
        hot_avg = 0
    else:
        # @TODO: this floating point division is basically extraneous
        # but I want it to look like something is happening for now
        # someone who cares about performance might want to remove this
        # especially since we are just creating an integer further down
        # if this is >= 1, which it will be in most sites.
        # especially since decimal comparison is meaningless here; < 1 = 0
        # users only vote in integers
        hot_avg = float(aggregate) / float(count)

    if hot_avg >= 1.0:
        hot_avg = math.floor(hot_avg)

    general.set_key_in_stat('hot_avg', hot_avg)
    general.set_key_in_stat('hot_avg_timestamp',
                            general.now_in_utc(),
                            type='datetime')

    print("hot_avg: {0}".format(hot_avg))

    return hot_avg
Example #5
0
def count_story_votes(submission_id):
    from raggregate.queries import submission
    dv_num = dbsession.query(Vote).filter(Vote.points < 0).filter(
        Vote.submission_id == submission_id).count()
    story = submission.get_story_by_id(submission_id)
    story.downvote_tally = dv_num
    story.downvote_tally_timestamp = general.now_in_utc()

    uv_num = dbsession.query(Vote).filter(Vote.points > 0).filter(
        Vote.submission_id == submission_id).count()
    story.upvote_tally = uv_num
    story.upvote_tally_timestamp = general.now_in_utc()

    story.total_vote_tally = uv_num + dv_num
    story.total_vote_timestamp = general.now_in_utc()

    dbsession.add(story)
    dbsession.flush()

    return [dv_num, uv_num]
Example #6
0
def calc_all_hot_window_scores(hot_eligible_age = timedelta(hours = 48)):
    # by default, only stories from the last 48 hours are eligible for "hot" status
    stories = dbsession.query(Submission).filter(Submission.deleted == False).filter(Submission.added_on > (general.now_in_utc() - hot_eligible_age)).all()

    for s in stories:
        calc_hot_window_score(s.id)

    general.set_key_in_stat('mass_hot_window_timestamp', general.now_in_utc(), type = 'datetime')

    dbsession.flush()

    return 0
Example #7
0
def recentize_contro(recalc_timediff=timedelta(seconds=1),
                     age_timediff=timedelta(hours=48)):
    """
    Ensure that upvote and downvote tallies of stories added in the last age_timediff time
    are no older than recalc_timediff time. Recalculate if so, do nothing if not.
    """
    stories = dbsession.query(Submission).filter(
        Submission.deleted == False).filter(Submission.added_on > (
            general.now_in_utc() - age_timediff)).filter(
                sqlalchemy.or_(
                    sqlalchemy.or_(
                        Submission.upvote_tally_timestamp <
                        (general.now_in_utc() - recalc_timediff),
                        Submission.downvote_tally_timestamp <
                        (general.now_in_utc() - recalc_timediff)),
                    sqlalchemy.or_(
                        Submission.upvote_tally_timestamp == None,
                        Submission.downvote_tally_timestamp == None))).all()
    for s in stories:
        count_story_votes(s.id)

    return 0
Example #8
0
def get_hot_stories(hot_eligible_age=timedelta(hours=48)):
    """
    Get a story list that is suitable for display as "hot".
    @param timediff: timedelta object representing hot story eligibility age
    """
    hot_avg = general.get_key_from_stat('hot_avg')['value']
    stories = dbsession.query(Submission).options(
        joinedload('submitter')).filter(Submission.deleted == False).filter(
            Submission.added_on > (
                general.now_in_utc() - hot_eligible_age)).filter(
                    Submission.hot_window_score > hot_avg).order_by(
                        Submission.hot_window_score.desc())
    return stories
Example #9
0
def get_controversial_stories(
        timediff=timedelta(hours=48), contro_threshold=5, contro_min=10):
    """
    Get a story list that is suitable for display as "controversial".
    @param timediff: timedelta representing controversial eligibility age
    @param contro_threshold: the maximum difference between up and down votes to be considered "controversial"
    @param contro_min: the minimum number of points required for something to appear as controversial
    @return: SA query ready to get list of stuff
    """
    stories = dbsession.query(Submission).options(
        joinedload('submitter')).filter(Submission.deleted == False).filter(
            Submission.added_on > (general.now_in_utc() - timediff)).filter(
                func.abs(Submission.upvote_tally - Submission.downvote_tally)
                <= contro_threshold).filter(
                    Submission.total_vote_tally > contro_min).order_by(
                        Submission.added_on.desc())
    return stories
Example #10
0
def calc_hot_window_score(submission_id, hot_point_window = timedelta(hours = 6)):
    """
    Retrieve vote/score information for a given submission over hot_point_window time.
    Count votes received in the last hot_point_window time. If this is higher than normal,
    the story will be considered hot.

    @param submission_id: the submission to calculate
    @param hot_point_window: timedelta object representing acceptable vote timeframe from now
    """
    try:
        story_votes = dbsession.query(Vote.submission_id, func.sum(Vote.points).label('points')).filter(Vote.added_on < general.now_in_utc(), Vote.added_on > (general.now_in_utc() - hot_point_window)).filter(Vote.submission_id == submission_id).group_by(Vote.submission_id).one()
    except sqlalchemy.orm.exc.NoResultFound:
        # no votes on this story yet
        return 0
    from raggregate.queries import submission
    story = submission.get_story_by_id(submission_id)
    story.hot_window_score = story_votes.points
    story.hot_window_score_timestamp = general.now_in_utc()
    dbsession.add(story)
    return story_votes.points
Example #11
0
def calc_hot_average(hot_point_window = timedelta(hours = 6)):
    """
    Calculate the average point assignment of stories over the last hot_point_window time.
    """
    story_votes = dbsession.query(Vote.submission_id, func.sum(Vote.points).label('points')).filter(Vote.added_on < general.now_in_utc(), Vote.added_on > (general.now_in_utc() - hot_point_window)).group_by(Vote.submission_id).all()

    aggregate = 0
    count = 0

    for sv in story_votes:
        aggregate += sv.points
        count += 1

    print "aggregate: " + str(aggregate)
    print "count: " + str(count)

    if aggregate <= 0:
        hot_avg = 0
    else:
        # @TODO: this floating point division is basically extraneous
        # but I want it to look like something is happening for now
        # someone who cares about performance might want to remove this
        # especially since we are just creating an integer further down
        # if this is >= 1, which it will be in most sites.
        # especially since decimal comparison is meaningless here; < 1 = 0
        # users only vote in integers
        hot_avg = float(aggregate) / float(count)

    if hot_avg >= 1.0:
        hot_avg = math.floor(hot_avg)

    general.set_key_in_stat('hot_avg', hot_avg)
    general.set_key_in_stat('hot_avg_timestamp', general.now_in_utc(), type = 'datetime')

    print("hot_avg: {0}".format(hot_avg))

    return hot_avg
Example #12
0
def recentize_hots(hot_recalc_threshold=timedelta(hours=1),
                   hot_point_window=None,
                   hot_eligible_age=None):
    """
    Recalculate the hot page every hot_recalc_threshold time.
    This depends on the timestamp keys placed in Stat in other calls.
    """

    count = 0

    def call_hot_average():
        if hot_point_window:
            calc_hot_average(hot_point_window)
        else:
            calc_hot_average()

    def get_hot_avg_time():
        hot_avg_timestamp = None

        try:
            hot_avg_timestamp = general.get_key_from_stat('hot_avg_timestamp',
                                                          type='datetime')
        except sqlalchemy.orm.exc.NoResultFound:
            call_hot_average()

        if type(hot_avg_timestamp) == dict:
            return hot_avg_timestamp['value']
        else:
            return None

    hot_avg_timestamp = None

    while hot_avg_timestamp == None and count < 5:
        hot_avg_timestamp = get_hot_avg_time()
        count += 1

    if count >= 5:
        print(
            "\t-----WARNING WARNING WARNING-----\n\tSomething has gone tragically wrong with the hotness: AVG.\n\t-----WARNING WARNING WARNING-----"
        )

    hot_avg_timestamp = hot_avg_timestamp.replace(tzinfo=pytz.utc)

    if hot_avg_timestamp < (general.now_in_utc() - hot_recalc_threshold):
        call_hot_average()

    count = 0

    def call_all_hot_window_scores():
        if hot_eligible_age:
            calc_all_hot_window_scores(hot_eligible_age)
        else:
            calc_all_hot_window_scores()

    def get_mass_time():
        mass_timestamp = None

        try:
            mass_timestamp = general.get_key_from_stat(
                'mass_hot_window_timestamp', type='datetime')
        except sqlalchemy.orm.exc.NoResultFound:
            call_all_hot_window_scores()

        if type(mass_timestamp) == dict:
            return mass_timestamp['value']
        else:
            return None

    mass_timestamp = None

    while mass_timestamp == None and count < 5:
        mass_timestamp = get_mass_time()
        count += 1

    if count >= 5:
        print(
            "\t-----WARNING WARNING WARNING-----\n\tSomething has gone tragically wrong with the hotness: MASS.\n\t-----WARNING WARNING WARNING-----"
        )

    mass_timestamp = mass_timestamp.replace(tzinfo=pytz.utc)

    if mass_timestamp < (general.now_in_utc() - hot_recalc_threshold):
        call_all_hot_window_scores()

    return 0
Example #13
0
def recentize_contro(recalc_timediff = timedelta(seconds = 1), age_timediff = timedelta(hours = 48)):
    """
    Ensure that upvote and downvote tallies of stories added in the last age_timediff time
    are no older than recalc_timediff time. Recalculate if so, do nothing if not.
    """
    stories = dbsession.query(Submission).filter(Submission.deleted == False).filter(Submission.added_on > (general.now_in_utc() - age_timediff)).filter(sqlalchemy.or_(sqlalchemy.or_(Submission.upvote_tally_timestamp < (general.now_in_utc() - recalc_timediff), Submission.downvote_tally_timestamp < (general.now_in_utc() - recalc_timediff)), sqlalchemy.or_(Submission.upvote_tally_timestamp == None, Submission.downvote_tally_timestamp == None))).all()
    for s in stories:
        count_story_votes(s.id)

    return 0
Example #14
0
def get_controversial_stories(timediff = timedelta(hours = 48), contro_threshold = 5, contro_min = 10):
    """
    Get a story list that is suitable for display as "controversial".
    @param timediff: timedelta representing controversial eligibility age
    @param contro_threshold: the maximum difference between up and down votes to be considered "controversial"
    @param contro_min: the minimum number of points required for something to appear as controversial
    @return: SA query ready to get list of stuff
    """
    stories = dbsession.query(Submission).options(joinedload('submitter')).filter(Submission.deleted == False).filter(Submission.added_on > (general.now_in_utc() - timediff)).filter(func.abs(Submission.upvote_tally - Submission.downvote_tally) <= contro_threshold).filter(Submission.total_vote_tally > contro_min).order_by(Submission.added_on.desc())
    return stories
Example #15
0
def get_hot_stories(hot_eligible_age = timedelta(hours = 48)):
    """
    Get a story list that is suitable for display as "hot".
    @param timediff: timedelta object representing hot story eligibility age
    """
    hot_avg = general.get_key_from_stat('hot_avg')['value']
    stories = dbsession.query(Submission).options(joinedload('submitter')).filter(Submission.deleted == False).filter(Submission.added_on > (general.now_in_utc() - hot_eligible_age)).filter(Submission.hot_window_score > hot_avg).order_by(Submission.hot_window_score.desc())
    return stories
Example #16
0
def recentize_hots(hot_recalc_threshold = timedelta(hours = 1), hot_point_window = None, hot_eligible_age = None):
    """
    Recalculate the hot page every hot_recalc_threshold time.
    This depends on the timestamp keys placed in Stat in other calls.
    """

    count = 0
    def call_hot_average():
        if hot_point_window:
            calc_hot_average(hot_point_window)
        else:
            calc_hot_average()

    def get_hot_avg_time():
        hot_avg_timestamp = None

        try:
            hot_avg_timestamp = general.get_key_from_stat('hot_avg_timestamp', type = 'datetime')
        except sqlalchemy.orm.exc.NoResultFound:
            call_hot_average()

        if type(hot_avg_timestamp) == dict:
            return hot_avg_timestamp['value']
        else:
            return None

    hot_avg_timestamp = None

    while hot_avg_timestamp == None and count < 5:
        hot_avg_timestamp = get_hot_avg_time()
        count += 1

    if count >= 5:
        print("\t-----WARNING WARNING WARNING-----\n\tSomething has gone tragically wrong with the hotness: AVG.\n\t-----WARNING WARNING WARNING-----")

    hot_avg_timestamp = hot_avg_timestamp.replace(tzinfo=pytz.utc)

    if hot_avg_timestamp < (general.now_in_utc() - hot_recalc_threshold):
        call_hot_average()

    count = 0

    def call_all_hot_window_scores():
        if hot_eligible_age:
            calc_all_hot_window_scores(hot_eligible_age)
        else:
            calc_all_hot_window_scores()

    def get_mass_time():
        mass_timestamp = None

        try:
            mass_timestamp = general.get_key_from_stat('mass_hot_window_timestamp', type = 'datetime')
        except sqlalchemy.orm.exc.NoResultFound:
            call_all_hot_window_scores()

        if type(mass_timestamp) == dict:
            return mass_timestamp['value']
        else:
            return None

    mass_timestamp = None

    while mass_timestamp == None and count < 5:
        mass_timestamp = get_mass_time()
        count += 1

    if count >= 5:
        print("\t-----WARNING WARNING WARNING-----\n\tSomething has gone tragically wrong with the hotness: MASS.\n\t-----WARNING WARNING WARNING-----")

    mass_timestamp = mass_timestamp.replace(tzinfo=pytz.utc)

    if mass_timestamp < (general.now_in_utc() - hot_recalc_threshold):
        call_all_hot_window_scores()

    return 0