AB scenario fails with copied file

[marc-vdm]

Updating AB

• Yes, I have updated AB and still experience this issue

What happened?

SDF file copied from AB, so errors are unlikely. Fails with the error below.

Files are confidential, will try to re-create later with non-confidential info.

Related:

• Scenario LCA calculation failed. #1122

Interestingly, not all files break:
Working version
princ-env-sci-db.xlsx
test_scenario.xlsx

Relevant errors

No data in scenario table 1, skipping
separator == ';'
Loading Scenario file. This may take a while for large files
<function ScenarioImportWidget.scenario_db_check at 0x000002D9E7620670> -- 0.0
<function SuperstructureManager.fill_empty_process_keys_in_exchanges at 0x000002D9E2AFBA60> -- 0.0010080337524414062
<function SuperstructureManager.verify_scenario_process_keys at 0x000002D9E2AFBB80> -- 0.01266932487487793
<function SuperstructureManager.check_duplicates at 0x000002D9E2AFBDC0> -- 0.0029997825622558594
<function SuperstructureManager.check_scenario_exchange_values at 0x000002D9E2AFBCA0> -- 0.0010004043579101562
<function SuperstructureManager.merge_flows_to_self at 0x000002D9E2AFB790> -- 0.003999948501586914
<function ScenarioImportWidget.sync_superstructure at 0x000002D9E7620550> -- 0.03170967102050781
<function ScenarioImportWidget.load_action at 0x000002D9E7620430> -- 5.103520154953003
Traceback (most recent call last):
  File "C:\Users\meidemtvander\PycharmProjects\AB\activity_browser\layouts\tabs\LCA_results_tab.py", line 62, in generate_setup
    new_tab = LCAResultsSubTab(data, self)
  File "C:\Users\meidemtvander\PycharmProjects\AB\activity_browser\layouts\tabs\LCA_results_tabs.py", line 136, in __init__
    self.setup_tabs()
  File "C:\Users\meidemtvander\PycharmProjects\AB\activity_browser\layouts\tabs\LCA_results_tabs.py", line 143, in setup_tabs
    self._update_tabs()
  File "C:\Users\meidemtvander\PycharmProjects\AB\activity_browser\layouts\tabs\LCA_results_tabs.py", line 154, in _update_tabs
    tab.update_tab()
  File "C:\Users\meidemtvander\PycharmProjects\AB\activity_browser\layouts\tabs\LCA_results_tabs.py", line 634, in update_tab
    self.lca_scores_widget.update_tab()
  File "C:\Users\meidemtvander\PycharmProjects\AB\activity_browser\layouts\tabs\LCA_results_tabs.py", line 687, in update_tab
    super().update_tab()
  File "C:\Users\meidemtvander\PycharmProjects\AB\activity_browser\layouts\tabs\LCA_results_tabs.py", line 309, in update_tab
    self.update_plot()
  File "C:\Users\meidemtvander\PycharmProjects\AB\activity_browser\layouts\tabs\LCA_results_tabs.py", line 693, in update_plot
    df = self.parent.mlca.get_results_for_method(method_index)
  File "C:\Users\meidemtvander\PycharmProjects\AB\activity_browser\bwutils\superstructure\mlca.py", line 214, in get_results_for_method
    return pd.DataFrame(
  File "C:\Users\meidemtvander\.conda\envs\ab_development_stable\lib\site-packages\pandas\core\frame.py", line 785, in __init__
    mgr = ndarray_to_mgr(
  File "C:\Users\meidemtvander\.conda\envs\ab_development_stable\lib\site-packages\pandas\core\internals\construction.py", line 336, in ndarray_to_mgr
    _check_values_indices_shape_match(values, index, columns)
  File "C:\Users\meidemtvander\.conda\envs\ab_development_stable\lib\site-packages\pandas\core\internals\construction.py", line 420, in _check_values_indices_shape_match
    raise ValueError(f"Shape of passed values is {passed}, indices imply {implied}")
ValueError: Shape of passed values is (6, 3), indices imply (5, 3)

Operating system

Windows 10

Conda environment

No response

[marc-vdm]

Update:

The problem seems to be in this section:

activity-browser/activity_browser/bwutils/multilca.py

Line 174 in c722b17

self.func_unit_translation_dict = {

The code here makes a dict (I don't know why, actually) of the functional units. I guess for easier reading, but the information is not shown to users??

In my error above, the code is reduced to 8 items instead of 9 as some FU have the same activity name (different product name though).
This also happens in non-scenario LCA, but apparently we don't use the variable self.func_key_list when that is the case as the calculation works as expected in that case.

I think we should review if we can get rid of it entirely (as the TODO states).

• Review why self.func_key_list is needed for scenarios currently
• Rewrite code so it is either not needed, or if not possible, re-write such that the keys are actually unique.