Script doesn't work properly with the Cassandra 2.1.12 and ka version about ctools HOT 8 CLOSED

Some column values are dumped in hex format. What is the type of column which is not displayed correctly. Also you can set debug = 1 in sstable.py to enable detailed logging. If still you are having issues, please share me sample SSTable if it is OK.

from ctools.

I am getting Datetime value to be like this 2016-05-20 12\\:21+0530 where as in cqlsh it shows like this 2016-05-20 12:22:42+0530. Also one more value of cell is coming out to be like this 73746f636b69643d3233337c6369747969643d317c70663d327c656e7649643d31 where as in cqlsh it is like this key1=1334|key2=1|key3=2|key4=1. Could it be due to my value contain | (pipe) and = (equal) operator

from ctools.

I will look in to the Datetime column value decoding. The other cell output is in hex format to comply with sstable2json format. You can convert it to binary by decoding hex data. You can use Python binascii module.
$ python -c "import binascii; print binascii.unhexlify("73746f636b69643d3233337c6369747969643d317c70663d327c656e7649643d31")"
stockid=233|cityid=1|pf=2|envId=1

from ctools.

I reviewed the Datetime column value. Cassandra uses character ':' to identify the clustering key. If your CQL table schema consists of compound primary key, then the clustering key is appended with character ':' . You can check the internal storage structure using cassandra-cli LIST command. In this case the character ':' in Datetime value is escaped with character ''. So in the json output you see these escape characters

from ctools.

Clustering key and integers are not handled in your code properly. While converting it in json it gives the garbage values. Kindly just review it once. I have checked cluster key buffer contains the size of nextkey then value after that seperator value which is : again size of next key and repeat the same process. But in your code it is not handled properly.

from ctools.

I will change the code to identify the column name type and print based on it

from ctools.

I modified sstable2json.py to take additional argument to specify composite type. For example if your column name is compounded using date and utf8 string please specify the argument as TimestampType,UTF8Type.
This is a stand alone tool and it does't have access to the CQL table schema. So you need to pass the compound column types to the script. Currently it can decode only TimestampType and UTF8Type. I will add the support for other types soon.

from ctools.

Modified to use CQL type names timestamp, varchar and text. Please specify the argument as a comma separated list of CQL types. Currently only timestamp, varchar and text are supported.

from ctools.