Exemple #1
0
    def getSessionDetails(self, session):
        '''Gets the session and visits from the table.

    Parameters
    ----------
    session : Session
      The session requested from the table.

    Returns
    -------
    data : dict
      The result of getting the session from the table. This contains either
      the error that occurred or the session and its visits.
    '''
        try:
            result = self.client.query(
                TableName=self.tableName,
                IndexName='GSI2',
                KeyConditionExpression='#gsi2 = :gsi2',
                ExpressionAttributeNames={'#gsi2': 'GSI2PK'},
                ExpressionAttributeValues={':gsi2': session.gsi2pk()},
                ScanIndexForward=True)
            if len(result['Items']) == 0:
                return {'error': 'Session not in table'}
            data = {'visits': []}
            for item in result['Items']:
                if item['Type']['S'] == 'visit':
                    data['visits'].append(itemToVisit(item))
                elif item['Type']['S'] == 'session':
                    data['session'] = itemToSession(item)
            # DynamoDB is limited in 1MB of query results. Continue to query from the
            # 'LastEvaluatedKey' when this condition is met.
            if 'LastEvaluatedKey' in result.keys():
                still_querying = True
                while still_querying:
                    result = self.client.query.scan(
                        TableName=self.tableName,
                        IndexName='GSI2',
                        KeyConditionExpression='#gsi2 = :gsi2',
                        ExpressionAttributeNames={'#gsi2': 'GSI2PK'},
                        ExpressionAttributeValues={':gsi2': session.gsi2pk()},
                        ScanIndexForward=True,
                        ExclusiveStartKey=result['LastEvaluatedKey'])
                    for item in result['Items']:
                        if item['Type']['S'] == 'visit':
                            data['visits'].append(itemToVisit(item))
                        elif item['Type']['S'] == 'session':
                            data['session'] = itemToSession(item)
                    if 'LastEvaluatedKey' not in result.keys():
                        still_querying = False
            return data
        except ClientError as e:
            print(f'ERROR getSessionDetails: { e }')
            return {'error': 'Could not get session from table'}
def test_processPages(table_name):
    ip = randomIP()
    this_event = event(ip, table_name)
    visits = [
        itemToVisit(record['dynamodb']['NewImage'])
        for record in this_event['Records']
        if record['dynamodb']['NewImage']['Type']['S'] == 'visit'
    ]
    client = DynamoClient(table_name)
    client.addVisitor(Visitor(ip))
    client.addVisits(visits)
    client.addBrowsers([
        itemToBrowser(record['dynamodb']['NewImage'])
        for record in this_event['Records']
        if record['dynamodb']['NewImage']['Type']['S'] == 'browser'
    ])
    for session in [
            itemToSession(record['dynamodb']['NewImage'])
            for record in this_event['Records']
            if record['dynamodb']['NewImage']['Type']['S'] == 'session'
    ]:
        client.addSession(session)
    assert processPages( client, this_event ) == 'Successfully added ' + \
      f'{ len( { visit.slug for visit in visits } ) } pages and updated 0 ' + \
      f'from { len( visits ) } records.'
Exemple #3
0
def _parsePageDetails( data, result ):
  '''Parses the DynamoDB items to their respective objects.

  Parameters
  ----------
  data : dict
    The parsed data as a dictionary.
  result : dict
    The result of the DynamoDB query.

  Returns
  data : dict
    The original parsed data combined with the new parsed data.
  '''
  for item in result['Items']:
    if item['Type']['S'] == 'visit':
      data['visits'].append( itemToVisit( item ) )
    elif item['Type']['S'] == 'page':
      data['page'] = itemToPage( item )
    elif item['Type']['S'] == 'day':
      data['days'].append( itemToDay( item ) )
    elif item['Type']['S'] == 'week':
      data['weeks'].append( itemToWeek( item ) )
    elif item['Type']['S'] == 'month':
      data['months'].append( itemToMonth( item ) )
    elif item['Type']['S'] == 'year':
      data['years'].append( itemToYear( item ) )
  return data
Exemple #4
0
def _parseVisitorDetails(data, result):
    '''Parses the DynamoDB items to their respective objects.

  Parameters
  ----------
  data : dict
    The parsed data as a dictionary.
  result : dict
    The result of the DynamoDB query.

  Returns
  data : dict
    The original parsed data combined with the new parsed data.
  '''
    for item in result['Items']:
        if item['Type']['S'] == 'visitor':
            data['visitor'] = itemToVisitor(item)
        elif item['Type']['S'] == 'visit':
            data['visits'].append(itemToVisit(item))
        elif item['Type']['S'] == 'session':
            data['sessions'].append(itemToSession(item))
        elif item['Type']['S'] == 'location':
            data['location'] = itemToLocation(item)
        elif item['Type']['S'] == 'browser':
            data['browsers'].append(itemToBrowser(item))
    return data
Exemple #5
0
def processPages( dynamo_client, event ):
  '''Creates the page and day/week/month/year from a DynamoDB event.

  Parameters
  ----------
  dynamo_client : DynamoClient
    The DynamoDB client used to access the table.
  event : dict
    The DynamoDB PUT event.

  Returns
  -------
  result : str
    The result of the number of pages processed.
  '''
  # Parse the visits from the event.
  visits = [
    itemToVisit( record['dynamodb']['NewImage'] )
    for record in event['Records']
    if record['dynamodb']['NewImage']['Type']['S'] == 'visit'
  ]
  if len( visits ) == 0:
    return 'No pages to process'
  # Store the number of new and updated pages
  new_pages = 0
  update_pages = 0
  # The unique slugs are iterated over because there may be multiple visits to
  # the same slug.
  for visit in [
    [ visit for visit in visits if visit.slug == slug ][0]
      for slug in { visit.slug for visit in visits }
  ]:
    # Query the page and its details
    page_details = dynamo_client.getPageDetails( visit )
    # Raise an exception when an error occurs.
    if 'error' in page_details.keys():
      raise Exception( page_details['error']  )
    # Update the page with new results
    update_result = dynamo_client.updatePage( page_details['visits'] )
    if 'error' in update_result.keys():
      raise Exception( update_result['error'] )
    # Add the page to the table when there isn't one
    if 'page' not in page_details.keys():
      new_pages += 1
    # Update the page when there is one
    else:
      update_pages += 1
  # Return what was done during execution
  return f'Successfully added { new_pages } pages and updated ' + \
    f'{ update_pages } from { len( visits ) } records.'
Exemple #6
0
def test_itemToVisit():
    visit = Visit(visitor_id, visit_date, user_number, page_title, page_slug,
                  session_start, scroll_events)
    item = visit.toItem()
    newVisit = itemToVisit(item)
    assert newVisit.id == visit.id
    assert newVisit.date == visit.date
    assert newVisit.user == visit.user
    assert newVisit.title == visit.title
    assert newVisit.slug == visit.slug
    assert newVisit.sessionStart == visit.sessionStart
    assert newVisit.prevTitle == visit.prevTitle
    assert newVisit.prevSlug == visit.prevTitle
    assert newVisit.nextTitle == visit.nextTitle
    assert newVisit.nextSlug == visit.nextSlug
    assert newVisit.timeOnPage == visit.timeOnPage
Exemple #7
0
def test_itemToVisit():
    visit = Visit('2020-12-23T20:32:26.000Z', '0.0.0.0', '0', 'Tyler Norlund',
                  '/', '2020-12-23T20:32:26.000Z')
    item = visit.toItem()
    newVisit = itemToVisit(item)
    assert newVisit.date == visit.date
    assert newVisit.ip == visit.ip
    assert newVisit.user == visit.user
    assert newVisit.title == visit.title
    assert newVisit.slug == visit.slug
    assert newVisit.sessionStart == visit.sessionStart
    assert newVisit.prevTitle == visit.prevTitle
    assert newVisit.prevSlug == visit.prevTitle
    assert newVisit.nextTitle == visit.nextTitle
    assert newVisit.nextSlug == visit.nextSlug
    assert newVisit.timeOnPage == visit.timeOnPage
Exemple #8
0
def test_itemToVisit_exception():
    with pytest.raises(Exception) as e:
        assert itemToVisit({})
    assert str(e.value) == "Could not parse visit"