Context
Found while working jepegit/cellpy#668 (v1.x batch notebook): after a custom update_cell → make_step_table() → make_summary(), the legacy bridge crashes with:
AttributeError: 'list' object has no attribute 'strip'
in cellpycore.cell_core._cycle_mode_to_test_mode when OldCellpyCellCore.cycle_mode is still list-shaped (often double-nested from cellpy-file meta, e.g. [['anode']]).
Why this is a core issue (not just v1.x)
cellpy master already mitigates on the consumer load path via recursive test_meta._unwrap after meta_test_dependent.update(as_list=True, …) (and has unit tests).- The engine/bridge still assumes a scalar string:
OldCellpyCellCore.cycle_mode getter only does one-level m[0]- setter keeps list values (
[x.lower() for x in cycle_mode]) instead of storing a scalar
_cycle_mode_to_test_mode calls cycle_mode.strip() with no list/tuple handling
Any path that skips consumer unwrap (legacy loads, in-memory meta mutation, older cellpy pins) can still hit this. v1.x (#668) will add a consumer-side unwrap backport; core should still be defensive.
What to do
- Getter (
OldCellpyCellCore.cycle_mode): recursively unwrap 1-element list/tuple to a scalar (same semantics as cellpy test_meta._unwrap).
- Setter: unwrap first, then store a scalar string (or
None) — do not persist a list of lowered strings.
_cycle_mode_to_test_mode: if given a list/tuple, unwrap before .strip(); keep existing string / None / unknown-spelling behaviour.- Tests covering at least:
'anode' / 'full_cell'['anode'][['anode']]None- make_summary / make_core_summary does not raise on nested list meta
Acceptance
Links
Context
Found while working jepegit/cellpy#668 (v1.x batch notebook): after a custom
update_cell→make_step_table()→make_summary(), the legacy bridge crashes with:in
cellpycore.cell_core._cycle_mode_to_test_modewhenOldCellpyCellCore.cycle_modeis still list-shaped (often double-nested from cellpy-file meta, e.g.[['anode']]).Why this is a core issue (not just v1.x)
cellpymaster already mitigates on the consumer load path via recursivetest_meta._unwrapaftermeta_test_dependent.update(as_list=True, …)(and has unit tests).OldCellpyCellCore.cycle_modegetter only does one-levelm[0][x.lower() for x in cycle_mode]) instead of storing a scalar_cycle_mode_to_test_modecallscycle_mode.strip()with no list/tuple handlingAny path that skips consumer unwrap (legacy loads, in-memory meta mutation, older cellpy pins) can still hit this. v1.x (#668) will add a consumer-side unwrap backport; core should still be defensive.
What to do
OldCellpyCellCore.cycle_mode): recursively unwrap 1-elementlist/tupleto a scalar (same semantics as cellpytest_meta._unwrap).None) — do not persist a list of lowered strings._cycle_mode_to_test_mode: if given a list/tuple, unwrap before.strip(); keep existing string /None/ unknown-spelling behaviour.'anode'/'full_cell'['anode'][['anode']]NoneAcceptance
cycle_modeno longer crashes summarization through the legacy bridge.Links
cellpy.readers.cellpy_file.read/test_meta._unwrap