rasdapy 1.1.0

Python interface to rasdaman

*RasdaPy is a client API for rasdaman that enables building and executing rasql queries with python.*

Requirements

• numpy, grpcio, protobuf.

• a running rasdaman instance, see http://rasdaman.org/wiki/Download

Installation

1. Make sure you have Python 2.7 or newer if using Python 2 and Python 3.4 or newer if using Python 3. If in doubt, run:

   $ python –version

2. If you do not have setuptools, numpy, grpcio, and protobuf installed, note that they will be downloaded as dependencies of rasdapy with: pip install rasdapy.

Usage

A full client is available that demonstrates how to use rasdapy to send queries to rasdaman and handle the results: http://rasdaman.org/browser/applications/rasdapy/rasql.py

Import RasdaPy core API

$ from rasdapy.db_connector import DBConnector
$ from rasdapy.query_executor import QueryExecutor

Connect to rasdaman

The DBConnector maintains the connection to rasdaman. In order to connect it is necessary to specify the host and port on which rasmgr is running, as well as valid rasdaman username and password.

$ db_connector = DBConnector("localhost", 7001, "rasadmin", "rasadmin")

Create the query executor

QueryExcutor is the interface through which rasql queries (create, insert, update, delete, etc.) are executed.

$ query_executor = QueryExecutor(db_connector)

Open the connection to rasdaman

$ db_connector.open()

Execute sample queries

The query below returns a list of all the collections available in rasdaman.

$ collection_list = query_executor.execute_read("select c from RAS_COLLECTIONNAMES as c")
$ print(collection_list)

Calculate the average of all values in collection mr2.

$ result = query_executor.execute_read("select avg_cells(c) from mr2 as c")
$ type(result)

Depending on the query the result will have a different type (e.g. scalar value, interval, array). Each data type is wrapped in a corresponding class: http://rasdaman.org/browser/applications/rasdapy/rasdapy/models

Select a particular subset of each array in collection mr2. This query will return raw array data that can be converted to a Numpy ndarray.

$ result = query_executor.execute_read("select m[0:10 ,0:10] from mr2 as m")
$ numpy_array = result.to_array()

Encode array subset to PNG format and write the result to a file.

$ result = query_executor.execute_read("select encode(m[0:10 ,0:10], "png") from mr2 as m")
$ with open("/tmp/output.png", "wb") as binary_file:
$   binary_file.write(result.data)

Create a rasdaman collection. Note that you should be connected with a user that has write permission; by default this is rasadmin/rasadmin in rasdaman, but this can be managed by the administrator.

$ query_executor.execute_write("create collection test_rasdapy GreySet")

Insert data from a PNG image into the collection. Similarly you need to have write permissions for this operation.

$ query_executor.execute_write("insert into test_rasdapy values decode($1)", "your_path/rasdaman/systemtest/testcases_services/test_all_wcst_import/test_data/wcps_mr/mr_1.png")

Alternatively, you can import data from a raw binary file; in this case it is necessary to specify the spatial domain and array type.

$ query_executor.execute_update_from_file("insert into test_rasdapy values $1", "your_path/rasdaman/systemtest/testcases_mandatory/test_select/testdata/101.bin", "[0:100]", "GreyString")

Further example queries and a general guide for rasql can be found in the rasql Query Language guide (http://rasdaman.org/browser/manuals_and_examples/manuals/doc-guides).

Close the connection to rasdaman

$ db_connector.close()

Best practices:

It is recommended to follow this template in order to avoid problems with leaked transactions:

from rasdapy.db_connector import DBConnector
from rasdapy.query_executor import QueryExecutor

db_connector = DBConnector("localhost", 7001, "rasadmin", "rasadmin")
query_executor = QueryExecutor(db_connector)

db_connector.open()

try:
    query_executor.execute_read("...")
    query_executor.execute_write("...")
    # ... more Python code
finally:
    db_connector.close()

Development Warning

The Python implementation of Protocol Buffers is not as mature as the C++ and Java implementations. It may be more buggy, and it is known to be pretty slow at this time. Since this library relies heavily on Protocol Buffers and GRPC, it might be prone to occasional hiccups and unexpected behaviour.

Contributors

• Bang Pham Huu

• Siddharth Shukla

• Dimitar Misev

Thanks also to

• Alex Mircea Dumitru

• Vlad Merticariu

• George Merticariu

• Alex Toader

• Peter Baumann