mstrio: simple and secure access to MicroStrategy data

mstrio provides a high-level interface for Python and R and is designed to give data scientists and developers simple and secure access to MicroStrategy data. It wraps MicroStrategy REST APIs into simple workflows, allowing users to connect to their MicroStrategy environment, fetch data from cubes and reports, create new datasets, and add new data to existing datasets. And, because it enforces MicroStrategy’s user and object security model, you don’t need to worry about setting up separate security rules.

With mstrio, it’s easy to integrate cross-departmental, trustworthy business data in machine learning workflows and enable decision-makers to take action on predictive insights in MicroStrategy Reports, Dossiers, HyperIntelligence Cards, and customized, embedded analytical applications.

Table of contents


Installation is easy when using CRAN. Read more about installation on MicroStrategy’s product documentation.

Install the mstrio package



Functionalities may be added to mstrio either in combination with annual MicroStrategy platform releases or through updates to platform releases. To ensure compatibility with APIs supported by your MicroStrategy environment, it is recommended to install a version of mstrio that corresponds to the version number of your MicroStrategy environment.

The current version of mstrio is 11.2.0 and is supported on MicroStrategy 2019 Update 4 (11.1.4) and later. To leverage the MicroStrategy for RStudio application, mstrio 11.2.0 and MicroStrategy 2019 Update 4 (11.1.4) are required.

If you intend to use mstrio with MicroStrategy version older than 11.1.4, refer to the CRAN package archive to download mstrio 10.11.1, which is supported on: * MicroStrategy 2019 (11.1) * MicroStrategy 2019 Update 1 (11.1.1) * MicroStrategy 2019 Update 2 (11.1.2) * MicroStrategy 2019 Update 3 (11.1.3)

To install a specific, archived version of mstrio, first obtain the URL for the version you need from the package archive on CRAN, and install as follows:

packageurl <- ""
install.packages(packageurl, repos=NULL, type="source")

Main Features

Read the following tutorials to become more familiar with mstrio - Connect to your MicroStrategy environment - Import data from a Report into a R data frame - Import data from a Cube into a R data frame - Export data into MicroStrategy by creating datasets - Update, replace, or append new data to an existing dataset


Connect to MicroStrategy

The connection object manages your connection to MicroStrategy. Connect to your MicroStrategy environment by providing the URL to the MicroStrategy REST API server, your username, password, and the project (case-sensistive) to connect to. By default, the connect() function expects your MicroStrategy username and password.


base_url <- ""
username <- "username"
password <- "password"
project_name <- "MicroStrategy Tutorial"

conn <- connect_mstr(base_url=base_url, username=username, password=password, project_name=project_name)

The URL for the REST API server typically follows this format: Validate that the REST API server is running by accessing in your web browser.

Currently, supported authentication modes are Standard (the default) and LDAP. To use LDAP, add login_mode when creating your Connection object:

conn <- connect_mstr(base_url=base_url, username=username, password=password, project_name=project_name,

By default, SSL certificates are validated with each API request. To turn this off, use:

conn <- connect_mstr(base_url=base_url, username=username, password=password, project_name=project_name, 

Import data from Cubes and Reports

In mstrio, Reports and Cubes have the same API, so you can use these examples for importing Report data to a DataFrame, too. To import the contents of a published cube into a DataFrame for analysis in R, use the Cube class:

my_cube <- Cube$new(connection=conn, cube_id="...")
df <- my_cube$to_dataframe()

To import Reports into a DataFrame for analysis in R use the optimized Report class:

my_report <- Report$new(connection=conn, report_id="...")
df <- my_report$to_dataframe()

By default, all rows are imported when my_cube$to_dataframe() or my_report$to_dataframe() are called. Filter the contents of a cube/report by passing the object IDs for the metrics, attributes, and attribute elements you need. First, get the object IDs of the metrics, attributes that are available within the Cube/Report object instance:


If you need to filter by attribute elements, call my_cube$get_attr_elements() or my_report$get_attr_elements() which will fetch all unique attribute elements per attribute. The attribute elements are available within the Cube/Report object instance:


Then, choose those elements by passing their IDs to the my_cube$apply_filters() method. To see the chosen elements, call my_cube$filters and to clear any active filters, call my_cube$clear_filters().

   attributes=list("A598372E11E9910D1CBF0080EFD54D63", "A59855D811E9910D1CC50080EFD54D63"),
   attr_elements=list("A598372E11E9910D1CBF0080EFD54D63:Los Angeles", "A598372E11E9910D1CBF0080EFD54D63:Seattle"))
df <- my_cube$to_dataframe()

Export data into MicroStrategy with Datasets

Create a new dataset

With mstrio you can create and publish single or multi-table datasets. This is done by passing Pandas DataFrames to a dataset constructor which translates the data into the format needed by MicroStrategy.

stores_df <- data.frame("store_id" = c(1, 2, 3),
                        "location" = c("New York", "Seattle", "Los Angeles"),
                        stringsAsFactors = FALSE)
sales_df <- data.frame("store_id" = c(1, 2, 3),
                       "category" = c("TV", "Books", "Accessories"),
                       "sales" = c(400, 200, 100),
                       "sales_fmt" = c("$400", "$200", "$100"),
                       stringsAsFactors = FALSE)

ds = Dataset$new(connection=conn, name="Store Analysis")
ds$add_table(name="Stores", data_frame=stores_df, update_policy="add")
ds$add_table(name="Sales", data_frame=sales_df, update_policy="add")

By default Dataset$create() will upload the data to the Intelligence Server and publish the dataset. If you just want to create the dataset but not upload the row-level data, use Dataset$create(auto_upload=FALSE) followed by Dataset$update() and Dataset$publish().

When using Dataset$add_table(), R data types are mapped to MicroStrategy data types. By default, numeric data (integers and floats) are modeled as MicroStrategy Metrics and non-numeric data are modeled as MicroStrategy Attributes. This can be problematic if your data contains columns with integers that should behave as Attributes (e.g. a row ID), or if your data contains string-based, numeric looking data which should be Metrics (e.g. formatted sales data, [“$450”, “$325”]). To control this behavior, provide a list of columns that you want to convert from one type to another.

ds$add_table(name="Stores", data_frame=stores_df, update_policy="add",

ds$add_table(name="Sales", data_frame=sales_df, update_policy="add",

It is also possible to specify where the dataset should be created by providing a folder ID in Dataset$create(folder_id="...").

After creating the dataset, you can obtain its ID using Datasets$dataset_id. This ID is needed for updating the data later.

Update a dataset

When the source data changes and users need the latest data for analysis and reporting in MicroStrategy, mstrio allows you to update the previously created dataset.

ds <- Dataset$new(connection=conn, dataset_id="...")
ds$add_table(name="Stores", data_frame=stores_df, update_policy='update')
ds$add_table(name="Sales", data_frame=stores_df, update_policy='upsert')

The update_policy parameter controls how the data in the dataset gets updated. Currently supported update operations are add (inserts entirely new data), update (updates existing data), upsert (simultaneously updates existing data and inserts new data), and replace (truncates and replaces the data).

By default, the raw data is transmitted to the server in increments of 100,000 rows. On very large datasets (>1 GB), it is beneficial to increase the number of rows transmitted to the Intelligence Server with each request. Do this with the chunksize parameter:


Finally, note that updating datasets that were not created using the REST API is not supported.

More resources


RStudio and Shiny are trademarks of RStudio, Inc.