synapseClient (version 1.14-0)

QueryResult: Retrieve Query Results

Description

QueryResult is a reference class that enables paging through query results.

Details

Synapse has the ability to query for entities using a SQL-like query language. See the Query service API wiki for details on how to form a query: http://sagebionetworks.jira.com/wiki/display/PLFM/Repository+Service+API#RepositoryServiceAPI-QueryAPI

The QueryResult object is returned from the synapseQuery function when blockSize is specified. The client can then call either fetch() or collect() to get the next block of results as a data.frame. The difference is that with collect(), the results are accumulated. For example, after 3 calls to collect(), with a blockSize of 10 we would expect a data.frame with 30 rows.

The collectAll() method repeatedly requests blocks of blocksize query results until there are no more or the limit is reached. This can be slow, so use with caution. It's usually safer to set a limit.

See Also

synapseQuery

Examples

## Not run: 
#   ## get a QueryResult object back from a query
#   qr <- synapseQuery("select id, name, parentId from file limit 1000", blockSize=100)
# 
#   ## get a data.frame with 100 results
#   df <- qr$fetch()
# 
#   ## accumulate the next 300 results
#   df <- qr$collect()
#   df <- qr$collect()
#   df <- qr$collect()
# 
#   # get number of rows accumulated
#   # should be 400 with initial fetch plus 3 calls to collect
#   length(qr)
# 
#   # return the selected column names
#   names(qr)
# 
#   ## accumulate the remaining results.
#   ## df will be a data.frame with all 1000 accumulated rows
#   df <- qr$collectAll()
# ## End(Not run)