1.9.0

Highlights

New optional use_iterative_json parameter in execute_mdx_dataframe/execute_view_dataframe functions #646, #612

This new feature allows TM1py to use iterative JSON parsing.

When use_iterative_json is True TM1py requires a significantly smaller memory footprint (down to ~ 20% of the original value) at an almost negligible cost of performance (single percentage digit):

from TM1py import TM1Service

with TM1Service(
        base_url="https://localhost:12354",
        user="admin",
        password="apple") as tm1:
    
    df = tm1.cells.execute_view_dataframe(cube_name="Sales", view_name="Very Large View", private=False,
                                          use_iterative_json=True)

This is a handy feature when dealing with large or very large data volumes (1M to 10M cells) in an environment with limited RAM.

New skip_non_updateable argument in write/write_dataframe functions #657

This optional argument to the write/write_dataframe functions asks TM1py to filter out cells that can not be updated before attempting to write. If not used, TM1py will fail with an error when attempting to write to rule-derived or consolidated cells.

from TM1py import TM1Service

with TM1Service(
        base_url="https://localhost:12354",
        user="admin",
        password="apple") as tm1:

    cells = {
        ('Belgium', 'Revenue', '2022', 'Apr', 'Revenue'): 10_000,
        ('Belgium', 'Revenue', '2022', 'May', 'Revenue'): 15_000,
        ('Belgium', 'Revenue', '2022', 'Jun', 'Revenue'): 20_000,
        ('Belgium', 'Revenue', '2022', 'Q1', 'Revenue'): 45_000,
    }

    tm1.cells.write("Sales", cells, skip_non_updateable=True)

This is a useful feature, as it saves the TM1py user from verifying the validity of the cell updates yourself when not working with flawless data sources.
Only errors w.r.t. updatability are suppressed! Attempts, for instance, to write to not existing elements will still raise errors.

New search functions to search through cubes, processes, and chores #660, #663, #665

from TM1py import TM1Service

with TM1Service(
        base_url="https://localhost:12354",
        user="admin",
        password="apple") as tm1:

    processes = tm1.processes.search_string_in_code(search_string="Sales")

from TM1py import TM1Service

with TM1Service(
        base_url="https://localhost:12354",
        user="admin",
        password="apple") as tm1:

    processes = tm1.processes.search_string_in_name(name_contains="Sales")

from TM1py import TM1Service

with TM1Service(
        base_url="https://localhost:12354",
        user="admin",
        password="apple") as tm1:

    cubes = tm1.cubes.search_for_dimension(dimension_name="Product", skip_control_cubes=True)

New Features

• add write to_message_log_function #621
• allow skip_zeros in execute_mdx_values function #659
• use python built-in csv module to create csv strings in execute_mdx_csv function #678
• new rename function in ApplicationService #682
• support compact_json in some execute_mdx_ functions #650

Improvements and Bugfixes

• get_last_message_from_processerrorlog to return text ac0d72f
• raise exception when session creation fails a42753e
• don't raise error if cellset is already deleted 3246391
• improve execute_set_mdx function 61fe2ea
• add update_or_create function for document from file 7f94f71
• drop obsolete arguments in TM1Service constructor #652
• support attribute type change #664
• fix get_values function issue to work with alternate hierarchies seamlessly #680
• improve handling of alternate hierarchies in write function #679
• optional measure_dimension_elements dictionary argument to write and write_dataframe function to improve performance b21ac47

Acknowledgments

Big thanks to @rkvinoth, @jrobinsonAG, @gbryant-dev, @raeldor, @adscheevel, @jordanjeremy for contributing code to this release, and many others for reporting bugs and requesting new features.

How to upgrade TM1py

To upgrade TM1py, just use the following command:

pip install TM1py --upgrade