analyze_audit.py

def analyze_audit_history(self):
    """ Parse audit history response and separate out deletes vs modifications/creations """
    last_modified_by_type, modified_ids, deleted_ids = {}, set(), set()
    last_modified_by_type['glbatch'] = arrow.now()  # Always consider glbatches modified

    for audit_trail in self._get_audit_trail():
        for event in audit_trail.response.data.find_all('audithistory'):
            object_type = event.objecttype.get_text()

            # Consolidate journal and glbatch to be the same, since they hit the same endpoint
            object_type = 'glbatch' if object_type == 'journal' else object_type

            if object_type in self.methods_by_object_type:
                access_mode = event.accessmode.get_text()
                if access_mode in ['Deleted', 'Delete Event', 'D']:
                    deleted_ids.add(event.objectkey.get_text())
                elif access_mode in ['Modify Event', 'Modified', 'M', 'Add Event', 'Created', 'C']:
                    modified_ids.add(event.objectkey.get_text())

            event_time = arrow.get(event.accesstime.get_text(), self.INTACCT_DATE_FORMAT)
            existing_time = last_modified_by_type.get(object_type)
            max_time = max(event_time, existing_time) if existing_time else event_time

            last_modified_by_type[object_type] = max_time

    return last_modified_by_type, modified_ids, deleted_ids

def _get_audit_trail(self):
    """ A generator that returns pages of audit trail history via the Intacct API """
    last_synced = self.erp_connection.metadata.get('last_transaction_sync')
    last_synced = arrow.get(last_synced).format(self.INTACCT_DATE_FORMAT) if last_synced else None

    num_remaining, offset = 1, 0
    while num_remaining:
        query = f'''
            <object>AUDITHISTORY</object>
            <select>
                <field>ACCESSTIME</field>
                <field>ACCESSMODE</field>
                <field>OBJECTTYPE</field>
                <field>OBJECTKEY</field>
                <field>SOURCE</field>
            </select>
            <orderby>
                <order>
                    <field>ACCESSTIME</field>
                    <descending/>
                </order>
            </orderby>
            <pagesize>2000</pagesize>
            <offset>{offset}</offset>
        '''

        if last_synced:
            query += f'''
                <filter>
                    <greaterthan>
                        <field>ACCESSTIME</field>
                        <value>{last_synced}</value>
                    </greaterthan>
                </filter>
            '''
        query = '<query>' + query + '</query>'

        root = IntacctClient._make_xml_request(self.erp_connection, query)

        num_remaining = int(root.response.data['numremaining'])
        offset += int(root.response.data['count'])

        yield root