Skip to main content

Access olap data sources through xmla

Project description

olap.xmla

This package is meant for accessing xmla datasources - see http://en.wikipedia.org/wiki/XML_for_Analysis

Builing

In this directory, run:

python setup.py build

Testing

Tests are done against the Mondrian, SSAS, icCube XMLA providers. The testsDiscover module tests behavior with different XMLA providers with the Discover command while testsExecute does the same with the Execute command. Note that you likely need to modify the sources if you want to test yourself since they contain specifics (namely the location of the services and names of the data sources).

Sample

Here is an example how to use it:

import olap.xmla.xmla as xmla

p = xmla.XMLAProvider()
# mondrian
c = p.connect(location="http://localhost:8080/mondrian/xmla")
# to analysis services (if iis proxies requests at /olap/msmdpump.dll)
# you will need a valid kerberos principal of course
# c = p.connect(location="https://my-as-server/olap/msmdpump.dll",
#               sslverify="/path/to/my/as-servers-ca-cert.pem")
# to icCube
# c = p.connect(location="http://localhost:8282/icCube/xmla", username="demo",
#               password="demo")

# getting info about provided data
print c.getDatasources()
print c.getMDSchemaCubes()
# for ssas a catalog is needed, so the call would be like
# get a catalogname from a call to c.getDBSchemaCatalogs()
# c.getMDSchemaCubes(properties={"Catalog":"a catalogname"})

# execute a MDX (working against the foodmart sample catalog of mondrian)
cmd= """select {[Measures].ALLMEMBERS} * {[Time].[1997].[Q2].children} on columns,
[Gender].[Gender].ALLMEMBERS on rows
from [Sales]
"""

res = c.Execute(cmd, Catalog="FoodMart")
#return only the Value property from the cells
res.getSlice(properties="Value")
# or two props
res.getSlice(properties=["Value", "FmtValue"])

# to return some subcube from the result you can
# return all
res.getSlice()
# carve out the 4th column
res.getSlice(Axis0=3)
# same as above, SlicerAxis is ignored
res.getSlice(Axis0=3, SlicerAxis=0)
# return the data sliced at the 2nd and 3rd row
res.getSlice(Axis1=[1,2])
# return the data sliced at the 2nd and 3rd row and at the 4th column
res.getSlice(Axis0=3, Axis1=[1,2])

Using the procedural interface:

import olap.xmla.xmla as xmla

p = xmla.XMLAProvider()
c = p.connect(location="http://localhost:8080/mondrian/xmla")
s = p.getOLAPSource()

# import olap.interfaces as oi
# oi.IOLAPSource.providedBy(s) == True

s.getCatalogs()
s.getCatalog("FoodMart").getCubes()
s.getCatalog("FoodMart").getCube("HR").getDimensions()
s.getCatalog("FoodMart").getCube("HR").getDimension("[Department]").\
getMembers()
s.getCatalog("FoodMart").getCube("HR").getDimension("[Department]").\
getMember("[Department].[14]")

cmd= """select {[Measures].ALLMEMBERS} * {[Time].[1997].[Q2].children} on columns,
[Gender].[Gender].ALLMEMBERS on rows
from [Sales]
"""
res=s.getCatalog("FoodMart").query(cmd)
res.getSlice()

Note

The contained vs.wsdl originates from the following package: http://www.microsoft.com/en-us/download/confirmation.aspx?id=9388 and was subsequently modified (which parameters go in the soap header) to work with the suds package.olap.xmla

CHANGES

0.7.2

  • now relies on requests 1.2.3

  • fixed race condition in kerberos auth

0.7.1

  • kerberos-auth was sent twice

0.7

  • works now with requests 0.14- and 1.-series of requests

  • selection of cell properties failed

0.6

  • dependency on kerberos and s4u2p is now optional

  • added optional kerberos-sspi package for kerberos on windows through sspi via pywin32

0.5

  • as_user and spn are no longer ignored in the kerberos authentication

  • implemented the procedural interface from olap.interfaces

  • fixed problem when no sliceraxis info is returned

  • parameter property of getSlice now spells properties

0.4

  • keyword kerberos is gone. kerberos auth need is detected automatically

  • BeginSession and EndSession provide XMLA Sessionsupport

  • changes to work with icCube XMLA provider

0.3

  • changed keyword doKerberos in XMLProvider.connect to kerberos

  • added sslverify keyword to XMLProvider.connect defaulting to True. This will be handed to requests get method, so you can point it to your certificate bundle file.

0.2

  • removed dependencies on specific versions in setup.py

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

xmla-0.7.2.tar.gz (18.5 kB view hashes)

Uploaded Source

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page