This works, When I remove the comparator_type
validators = {
'tid': 'IntegerType',
'approved': 'BooleanType',
'text': 'UTF8Type',
'favorite_count':'IntegerType',
'retweet_count': 'IntegerType',
'expanded_url': 'UTF8Type',
'tuid': 'LongType',
'screen_name': 'UTF8Type',
'profile_image': 'UTF8Type',
'embedly_data': 'BytesType',
'created_at': 'UTF8Type',
}
SYSTEM_MANAGER.create_column_family('Narrative','Twitter_search',
default_validation_class='UTF8Type', key_validation_class='UTF8Type',
column_validation_classes=validators)
On Wed, Dec 11, 2013 at 12:23 PM, Kumar Ranjan <[email protected]> wrote:
> I am using ccm cassandra version
>
> *1.2.11*
>
>
> On Wed, Dec 11, 2013 at 12:19 PM, Kumar Ranjan <[email protected]>wrote:
>
>> validators = {
>>
>> 'approved': 'BooleanType',
>>
>> 'text': 'UTF8Type',
>>
>> 'favorite_count':'IntegerType',
>>
>> 'retweet_count': 'IntegerType',
>>
>> 'expanded_url': 'UTF8Type',
>>
>> 'tuid': 'LongType',
>>
>> 'screen_name': 'UTF8Type',
>>
>> 'profile_image': 'UTF8Type',
>>
>> 'embedly_data': 'CompositeType',
>>
>> 'created_at': 'UTF8Type',
>>
>> }
>>
>> SYSTEM_MANAGER.create_column_family('Narrative','Twitter_search_test',
>> comparator_type='CompositeType', default_validation_class='UTF8Type',
>> key_validation_class='UTF8Type', column_validation_classes=validators)
>>
>>
>> throws:
>>
>> *InvalidRequestException*: InvalidRequestException(why='Invalid
>> definition for comparator org.apache.cassandra.db.marshal.CompositeType.
>>
>> Can you please explain why?
>>
>>
>> On Wed, Dec 11, 2013 at 12:08 PM, Tyler Hobbs <[email protected]> wrote:
>>
>>> What options are available depends on what version of Cassandra you're
>>> using.
>>>
>>> You can specify the row key type with 'key_validation_class'.
>>>
>>> For column types, use 'column_validation_classes', which is a dict
>>> mapping column names to types. For example:
>>>
>>> sys.create_column_family('mykeyspace', 'users',
>>> column_validation_classes={'username': UTF8Type, 'age': IntegerType})
>>>
>>>
>>> On Wed, Dec 11, 2013 at 10:32 AM, Kumar Ranjan <[email protected]>wrote:
>>>
>>>> What are the all possible values for cf_kwargs ??
>>>>
>>>> SYSTEM_MANAGER.create_column_family('Narrative','Twitter_search_test',
>>>> comparator_type=UTF8Type, ???? )
>>>>
>>>> ???? - Here I want to specify, Column data types and row key type. How
>>>> can I do that ?
>>>>
>>>>
>>>> On Thu, Aug 15, 2013 at 12:30 PM, Tyler Hobbs <[email protected]>wrote:
>>>>
>>>>> The column_validation_classes arg is just for defining individual
>>>>> column types. Glad you got it figured out, though.
>>>>>
>>>>>
>>>>> On Thu, Aug 15, 2013 at 11:23 AM, Pinak Pani <
>>>>> [email protected]> wrote:
>>>>>
>>>>>> Thanks for quick reply. Apparantly, I was trying this to get working
>>>>>>
>>>>>> cf_kwargs = {'default_validation_class':COUNTER_COLUMN_TYPE}
>>>>>> sys.create_column_family('my_ks', 'vote_count',
>>>>>> column_validation_classes=cf_kwargs) #1
>>>>>>
>>>>>> But this works:
>>>>>>
>>>>>> sys.create_column_family('my_ks', 'vote_count', **cf_kwargs) #2
>>>>>>
>>>>>> I thought #1 should work.
>>>>>>
>>>>>>
>>>>>>
>>>>>> On Thu, Aug 15, 2013 at 9:15 PM, Tyler Hobbs <[email protected]>wrote:
>>>>>>
>>>>>>> The only thing that makes a CF a counter CF is that the default
>>>>>>> validation class is CounterColumnType, which you can set through
>>>>>>> SystemManager.create_column_family().
>>>>>>>
>>>>>>>
>>>>>>> On Thu, Aug 15, 2013 at 10:38 AM, Pinak Pani <
>>>>>>> [email protected]> wrote:
>>>>>>>
>>>>>>>> I do not find a way to create a counter column family in Pycassa.
>>>>>>>> This[1] does not help.
>>>>>>>>
>>>>>>>> Appreciate if someone can help me.
>>>>>>>>
>>>>>>>> Thanks
>>>>>>>>
>>>>>>>> 1.
>>>>>>>> http://pycassa.github.io/pycassa/api/pycassa/system_manager.html#pycassa.system_manager.SystemManager.create_column_family
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> Tyler Hobbs
>>>>>>> DataStax <http://datastax.com/>
>>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Tyler Hobbs
>>>>> DataStax <http://datastax.com/>
>>>>>
>>>>
>>>>
>>>
>>>
>>> --
>>> Tyler Hobbs
>>> DataStax <http://datastax.com/>
>>>
>>
>>
>
